THE BIOLOGICAL BULLETIN PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY Associate Editors SEP 9 1998 ;, MA JAMES A. BLAKE, ENSR Marine & Coastal Center, Woods Hole Louis E. BURNETT, Grice Marine Biological Laboratory, College of Charleston WILLIAM D. COHEN, Hunter College, City University of New York CHARLES D. DERBY, Georgia State University SHINYA INOUE, Marine Biological Laboratory RUDOLF A. RAFF, Indiana University Editorial Board PETER B. ARMSTRONG, University of California, Davis ANDREW R. CAMERON, California Institute of Technology THOMAS H. DIETZ, Louisiana State University RICHARD B. EMLET, Oregon Institute of Marine Biology, University of Oregon DAVID EPEL, Hopkins Marine Station, Stanford Univer- sity DAPHNE GAIL FAUTIN, University of Kansas WILLIAM F. GILLY, Hopkins Marine Station, Stanford University ROGER T. HANLON, Marine Biological Laboratory GREGORY HINKLE, University of Massachusetts, Dart- mouth MAKOTO KOBAYASHI, Hiroshima University of Eco- nomics MICHAEL LABARBERA, University of Chicago DONAL T. MANAHAN, University of Southern California MARGARET MCFALL-NGAI, Kewalo Marine Laboratory, University of Hawaii MARK W. MILLER, Institute of Neurobiology, University of Puerto Rico TATSUO MOTOKAWA, Tokyo Institute of Technology YOSHITAKA NAGAHAMA, National Institute for Basic Biology, Japan SHERRY D. PAINTER, Marine Biomedical Institute, Uni- versity of Texas Medical Branch K. RANGA RAO, University of West Florida BARUCH RINKEVICH, Israel Oceanographic & Limno- logical Research Ltd. RICHARD STRATHMANN, Friday Harbor Laboratories, University of Washington STEVEN VOGEL, Duke University J. HERBERT WAITE, University of Delaware SARAH ANN WOODIN, University of South Carolina RICHARD K. ZIMMER-FAUST, University of California, Los Angeles Editor: MICHAEL J. GREENBERG, The Whitney Laboratory, University of Florida Managing Editor: PAMELA L. CLAPP, Marine Biological Laboratory AUGUST, 1998 Printed and Issued by LANCASTER PRESS, Inc. 3575 HEMPLAND ROAD LANCASTER, PA Cover Expression of endostyle-specific genes in a 1- month-old adult of an ascidian ( dona intesti- nalis ), revealed by in situ hybridization in a whole mount. Ascidians, or tunicates, are ubiquitous ma- rine animals; they are also primitive chordates and therefore share the characteristic features of the phylum Chordata with the cephalochordates and vertebrates. But the endostyle, a pharyngeal secre- tory structure associated with food collection, is also found in ascidians, amphioxus, and larval lampreys; moreover, the endostyle is transformed into the thyroid gland in the course of vertebrate evolution. Therefore, the endostyle-specific genes are key markers that should help to elucidate, not only the development of the organ itself, but also the evolution of the chordates. Michio Ogasawara and Noriyuki Satoh of Kyoto University are pursu- ing these matters; details of their work are reported in this issue. CONTENTS IMAGING AND MICROSCOPY PHYSIOLOGY Piston, David W. Concepts in Imaging and Microscopy. Choosing objective lenses: the importance of numerical aperture and magnification in digital optical microscopy RESEARCH NOTES Harosi, Ferenc I., lone Hunt von Herbing, and Jeffrey R. Van Keuren Sickling of anoxic red blood cells in fish Stuart, J. A., E. L. Ooi, J. McLeod, A. E. Bourns, and J. S. Ballantyne D- and t-/3-hydroxybutyrate dehydrogenases and the evolution of ketone body metabolism in gastropod molluscs Rodhouse, Paul G. Physiological progenesis in cephalopod molluscs . . 12 17 Leys, Sally P., and Henry M. Reiswig Transport pathways in the neotropical sponge Aplysma 30 Nair, P. Satish, and William E. Robinson Calcium speciation and exchange between blood and extrapallial fluid of the quahog Mercenaria mtrcenaria (L.) 43 Nakatani, Is. mm. Yoshinori Okada, and Takuji Kitahara Induction of extra claws on the chelipeds of a cray- fish, Procambarus clarkii 52 ECOLOGY AND EVOLUTION Ogasawara, Michio, and Noriyuki Satoh Isolation and characterization of endostyle-specific genes in the ascidian dona intestmalis 60 Maruyama, Tadashi, Masaharu Ishikura, Satoru Yama- zaki, and Satoru Kanai Molecular phylogeny of zooxanthellate bivalves ... 70 DEVELOPMENT AND REPRODUCTION Holm, Eric R., Brian T. Nedved, Eugenio Carpizo- Ituarte, and Michael G. Hadfield Metamorphic-signal transduction in Hydroides elegans (Polychaeta: Serpulidae) is not mediated by a G protein NEUROBIOLOGY AND BEHAVIOR Cobb, Christopher S., and Roddy Williamson Electrophysiology and innervation of the photosensi- tive epistellar body in the lesser octopus Eledone cirrhosa . 78 21 Annual Report of the Marine Biological Laboratory . . Rl THE BIOLOGICAL BULLETIN THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts 02543 Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts 02543. Subscription per year (six issues, two volumes): $195 for libraries; $95 for individuals. Subscription per volume (three issues): $97.50 for libraries; $50 for individuals. Back and single issues (subject to availability): $40 for libraries; $20 for individuals. Communications relative to manuscripts should be sent to Michael J. Greenberg, Editor-in-Chief, or Pamela L Clapp, Managing Editor, at the Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts 02543. Telephone: (508) 289-7428. FAX: 508-457-1924. E-mail: pclapp@mbl.edu. http://www.mbl.edu/BiologicalBulletin/ The home page for the electronic companion to THE BIOLOGICAL BULLETIN the Marine Models Electronic Record and other BIOLOGICAL BULLETIN publications is available on the World Wide Web at the address shown above. THE BIOLOGICAL BULLETIN is indexed in bibliographic services including Index Medicus and MEDLINE, Chemical Abstracts, Current Contents. Elsevier BIOBASE/Current Awareness in Biological Sciences, and Geo Abstracts. Printed on acid free paper, effective with Volume 180, Issue I, 1991. POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543. Copyright 1998. by the Marine Biological Laboratory Periodicals postage paid at Woods Hole, MA, and additional mailing offices. ISSN 0006-3185 INSTRUCTIONS TO AUTHORS The Biological Bulletin accepts outstanding original re- search reports of general interest to biologists throughout the world. Papers are usually of intermediate length ( 10-40 manu- script pages). A limited number of solicited review papers may be accepted after formal review. A paper will usually appear within four months after its acceptance. Very short, especially topical papers (less than 9 manu- script pages including tables, figures, and bibliography) will be published in a separate section entitled "Research Notes." A Research Note in The Biological Bulletin follows the format of similar notes in Nature. It should open with a summary para- graph of 150 to 200 words comprising the introduction and the conclusions. The rest of the text should continue on without subheadings, and there should be no more than 30 references. References should be referred to in the text by number, and listed in the Literature Cited section in the order that they appear in the text. Unlike references in Nature, references in the Re- search Notes section should conform in punctuation and ar- rangement to the style of recent issues of The Biological Bulle- tin. Materials and Methods should be incorporated into appro- priate figure legends. See the article by Lohmann etal. (October 1990, Vol. 179: 214-218) for sample style. A Research Note will usually appear within two months after its acceptance. The Editorial Board requests that regular manuscripts con- form to the requirements set below; those manuscripts that do not conform will be returned to authors for correction before review. 1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies of photographs are not ac- ceptable for review purposes.) The submission letter accompa- nying the manuscript should include a telephone number, a FAX number, and (if possible) an E-mail address for the corresponding author. The original manuscript must be typed in no smaller than 12 pitch or 10 point, using double spacing (including figure leg- ends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper, 8 by 1 1 inches. Please, no right justification. Manu- scripts should be proofread carefully and errors corrected legibly in black ink. Pages should be numbered consecutively. Margins on all sides should be at least 1 inch (2.5 cm). Manuscripts should conform to the Council of Biology Editors Style Manual, 5th Edi- tion (Council of Biology Editors, 1983) and to American spelling. Unusual abbreviations should be kept to a minimum and should be spelled out on first reference as well as defined in a footnote on the title page. Manuscripts should be divided into the following components: Title page. Abstract (of no more than 200 words). Introduction, Materials and Methods, Results, Discussion, Ac- knowledgments, Literature Cited, Tables, and Figure Legends. In addition, authors should supply a list of words and phrases under which the article should be indexed. 2. Title page. The title page consists of a condensed title or running head of no more than 35 letters and spaces, the manuscript title, authors' names and appropriate addresses, and footnotes listing present addresses, acknowledgments or contri- bution numbers, and explanation of unusual abbreviations. 3. Figures. The dimensions of the printed page. 7 by 9 inches, should be kept in mind in preparing figures for publica- tion. We recommend that figures be about 1 times the linear dimensions of the final printing desired, and that the ratio of the largest to the smallest letter or number and of the thickest to the thinnest line not exceed 1:1.5. Explanatory matter generally should be included in legends, although axes should always be identified on the illustration itself. Figures should be prepared for reproduction as either line cuts or halftones. Figures to be reproduced as line cuts should be unmounted glossy photo- graphic reproductions or drawn in black ink on white paper, good-quality tracing cloth or plastic, or blue-lined coordinate paper. Those to be reproduced as halftones should be mounted on board, with both designating numbers or letters and scale bars affixed directly to the figures. All figures should be num- bered in consecutive order, with no distinction between text and plate figures and cited, in order, in the text. The author's name and an arrow indicating orientation should appear on the reverse side of all figures. Color: The Biological Bulletin will publish color figures and plates, but must bill authors for the actual additional cost of printing in color. The process is expensive, so authors with more than one color image should consistent with editorial concerns, especially citation of figures in order combine them into a single plate to reduce the expense. On request, when supplied with a copy of a color illustration, the editorial staff will provide a pre-publication estimate of the printing cost. 4. Tables, footnotes, figure legends, etc. Authors should follow the style in a recent issue of The Biological Bulletin in preparing table headings, figure legends, and the like. Because of the high cost of setting tabular material in type, authors are asked to limit such material as much as possible. Tables, with their headings and footnotes, should be typed on separate sheets, numbered with consecutive Roman numerals, and placed after the Literature Cited. Figure legends should contain enough in- formation to make the figure intelligible separate from the text. Legends should be typed double spaced, with consecutive Ara- bic numbers, on a separate sheet at the end of the paper. Foot- notes should be limited to authors' current addresses, acknowl- edgments or contribution numbers, and explanation of unusual abbreviations. All such footnotes should appear on the title page. Footnotes are not normally permitted in the body of the text. 5. Literature cited. In the text, literature should be cited by the Harvard system, with papers by more than two authors cited as Jones et al., 1980. Personal communications and mate- rial in preparation or in press should be cited in the text only, with author's initials and institutions, unless the material has been formally accepted and a volume number can be supplied. The list of references following the text should be headed Litera- ture Cited, and must be typed double spaced on separate pages, conforming in punctuation and arrangement to the style of recent issues of The Biological Bulletin. Citations should include complete titles and inclusive pagination. Journal abbre- viations should normally follow those of the U. S. A. Standards Institute (USASI), as adopted by BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor differences set out below. The most generally useful list of biological journal titles is that published each year by BIOLOGICAL ABSTRACTS (BIOSIS List of Serials; the most recent issue). Foreign authors, and others who are accustomed to using THH WORLD LIST OF SCIENTIFIC PERIOD- ICALS, may find a booklet published by the Biological Council of the U.K. (obtainable from the Institute of Biology, 41 Queen's Gate, London, S.W.7, England, U.K.) useful, since it sets out the WORLD LIST abbreviations for most biological jour- nals with notes of the USASI abbreviations where these differ. CHEMICAL ABSTRACTS publishes quarterly supplements of addi- tional abbreviations. The following points of reference style for THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD LIST) usage: A. Journal abbreviations, and book titles, all underlined (for italics) B. All components of abbreviations with initial capitals (not as European usage in WORLD LIST e.g., J. Cell. Cornp. Physiol. NOT J. cell. comp. Physiol. ) C. AH abbreviated components must be followed by a pe- riod, whole word components must not (i.e.. J. Cancer Res.) D. Space between all components (e.g., J. Cell. Comp. Physiol.. not J.Cell.Comp.Physiol.) E. Unusual words in journal titles should be spelled out in full, rather than employing new abbreviations invented by the author. For example, use Rit Vfsindafjelags Islendinga without abbreviation. F. All single word journal titles in full (e.g., Veliger, Ecol- ogy, Brain). G. The order of abbreviated components should be the same as the word order of the complete title (i.e., Proc. and Trans, placed where they appear, not transposed as in some BIOLOGICAL ABSTRACTS listings). H. A few well-known international journals in their pre- ferred forms rather than WORLD LIST or USASI usage (e.g.. Nature, Science. Evolution NOT Nature, Land., Science, N.Y.; Evolution, Lancaster, Pa.) 6. Reprints, page proofs, and charges. Authors of arti- cles in black and white (no color figures) receive their first 50 reprints (without covers) free of charge. Color reprints and additional black-and-white reprints may be purchased; authors will receive order forms. Reprints normally will be delivered about 2 to 3 months after the issue date. Authors (or delegates for foreign authors) will receive page proofs of articles shortly before publication. They will be charged the current cost of printers' time for corrections to these (other than corrections of printers' or editors' errors). Other than these charges for au- thors' alterations. The Biological Bulletin does not have page charges. Reference: Biol. Bull. 195: 1-4. (August, 1998) Concepts in Imaging and Microscopy Choosing Objective Lenses: The Importance of Numerical Aperture and Magnification in Digital Optical Microscopy DAVID W. PISTON Department of Molecular Physiology and Biophysics, Vanderbilt University, Nit.slmlle, Tennessee 37232-0615 Abstract. Microscopic images are characterized by a number of microscope-specific parameters numerical ap- erture (NA), magnification (M), and resolution (R) and by parameters that also depend on the specimen for ex- ample, contrast, signal-to-noise ratio, dynamic range, and integration time. In this article, issues associated with the microscope-specific parameters NA, M, and R are discussed with respect to both widefield and laser scanning confocal microscopies. Although most of the discussion points apply to optical microscopy in general, the main application con- sidered is fluorescence microscopy. Introduction The objective lens is arguably the most important com- ponent of any light microscope (Keller, 1995). Advances in digital imaging have completely changed the way that optical microscopy is performed, and have also changed the relevant specifications for objective lenses. Although lens design, construction, and quality have improved to keep up with the requirements of modern light micros- copy, the markings on the lenses remain as they have been for decades. On the objective lens shown in Figure Received 13 February 1998; accepted 28 May 1998. E-mail: dave.piston@mcmail.vanderbilt.edu This is the second in a series of articles entitled "Concepts in Imaging and Microscopy." This series is supported by the Opto-Precision Instru- ments Association (OPIA) and was introduced with an editorial in the April 1998 issue of this journal (Biol. Bull. 194: 99). The first article in the series was written by Dr. Kenneth R. Castleman and appeared in the same issue (Biol. Bull. 194: 100-107). 1. the word "FLUAR" describes the type of lens design; although all manufacturers use similar types of designs, the nomenclature varies from company to company. The next most notable feature on the objective lens is the magnification (M), which in the illustration is 100X. It is written in the largest font of all the specifications, yet as is discussed here, it is not the most important parame- ter. This distinction belongs to the numerical aperture (NA), which is written next to the magnification, but in a smaller font, and in this case is 1.30. The immersion medium for this objective is also given. Below the magni- fication and numerical aperture, the tube length (^) and the coverslip thickness (0.17 mm) are given. Currently all manufacturers are offering infinity-corrected optics (de- noted by the ^ symbol), and most lenses are optimally corrected for a number 1.5 coverslip, nominally 170-pm thick. Both of these parameters are important, but the objective lens will still function adequately for many ap- plications with other tube lengths and coverslip thick- nesses. However, because the manufacturers perform chromatic corrections in different ways, multi-color ex- periments for instance, co-localization of two different colored immunofluorescent probes should be per- formed using only sets of optics that were designed to work together. This applies not only to mixing lenses of different manufacturers, but also to mixing older and newer lines of optics. The working distance of the objec- tive (the depth into the sample to which the lens can focus before it runs into the sample) is also a very important parameter, especially for confocal microscopy in thick biological samples. Despite the critical nature of this spec- D. W. PISTON Figure 1. An objective lens with typical iiKiikmgs. Although this is a Zeiss lens, most manufacturers use similar markings. itication, the working distance is not marked on most lenses. Nikon has now started writing the working dis- tance on their CFI60 optics, and it is hoped that this trend will be followed by the other manufacturers. This short article describes the relative importance of magnification and numerical aperture for digital optical mi- croscopy. Traditionally, observations made with optical mi- croscopes were detected by eye. and in this case, the size of the detector pixels given by physiological factors in the human eye is not optimal, so the magnification was increased so the sample could be "seen" better. In digital imaging, however, the magnification can be determined by the combination of resolution and detector pixel size. To understand the relative importance of NA and magnifica- tion, we must consider the basics of image formation and the effect of lens parameters on the resolution and informa- tion content in optical microscopy. Because the resolving power of an optical microscope is dependent only on the numerical aperture, magnification should be thought of as a secondary parameter whose optimal value can be deter- mined by the NA, detector pixel size, and other instrument- independent imaging parameters. Thus, NA is a more im- portant parameter than magnification in digital imaging. The practical implications of this conclusion are described tor two commonly used modes of fluorescence imasiii": widefield epi-fluorescence microscopy with a CCD camera as the detector, and laser scanning confocal microscopy with photomultiplier tube detectors. Basics of Image Formation As might be guessed by looking at the markings on any objective lens, the magnification and numerical aper- ture are important for the image-formation properties of an optical microscope. Magnification for an optical instru- ment is defined as the relative enlargement of the image over the object. Although at first glance it would seem best to use the highest magnification possible, the maxi- mal useful magnification is limited by the resolution of the imaging instrument (as described in the next para- graph). The definition of numerical aperture is more com- plicated. NA is defined by the half-angle of the objective's collection cone (a) and the index of refraction of the immersion medium (n), and is expressed by NA = n-sin(o) (Inoue and Spring. 1997, p. 32). The larger the cone of collected light, the higher the NA, and the more light that will be collected. Thus in practice. NA can be thought of as the amount of light that is collected by the objective lens; a high-NA lens collects more light than a lovv-NA lens. An analogy is with telescopes: a larger telescope collects more light just as a lens with a larger NA collects more light. Most optical microscopes also otter the option of secondary magnification between the objective lens and the detector. Use of such extra magni- fication may sometimes be required (see Table I). "but should be avoided if possible since extra light loss is introduced. As suggested above, resolution (as determined by the basic diffraction principles of light) limits the useful mag- nification in an optical microscope. Resolution (R) is de- fined as the smallest distance that two objects can be apart and still be discerned as two separate objects. There are many mathematical definitions for resolution, but a simple and reasonable approximation is R = X/(2-NA). where \ is the wavelength of the light (Inoue and Spring, 1997, p. 3 1 ). This relationship indicates that when using a high- NA lens and 500-nm (blue-green) light, the smallest re- solvable distance is -200 nm, or 0.2 pm. which agrees well with experimental values. One frequent point of con- tusion for microscope users is the difference between spatial resolution (the ability to distinguish multiple ob- jects) and spatial precision (the ability to localize a single object). Many image-processing enhancements can be used to increase the precision of localization. For exam- ple, the path of a single microtubule can be determined to --lOnm precision by pixel-fitting (e.g.. Ghosh and Webb, 1994) or deconvolution methods (e.g., Agard et ul.. 1989; Carrington et a/.. 1990; Holmes et ui. 1995). However, this is not 10-nm resolution: 10-nm resolution means that two microtubules that are 10-nm apart can be recogni/ed as two separate tubules. If multiple objects are small and close together (that is, close enough that they cannot he resolved), then no amount of image pro- cessing can differentiate between several individual ob- jects and a single object. Since there is a minimum resolvable distance for every microscope, continuing to increase the magnification past a certain point will no longer increase the information CHOOSING OBJECTIVE LENSES content of the image. Further magnification beyond this point is sometimes referred to as "empty" or meaningless magnification. This is analogous to any digital image on a computer, where pixelation is observed when an image is magnified on the screen (this can be seen, for example, by repeatedly using the "zoom in" command in Adobe Photoshop). So the question obviously arises, how should the correct magnification be chosen? A good rule is to use the Nyquist criterion, which basically says that one should collect two points per resolution size (I none and Spring, 1997, p. 513). Collecting images in this manner maximizes the information content. The use of \A2-NA) for the resolution criterion, and of R/2 for the sampling rate are both arbitrary. Many microscopists select other resolution criteria, but all of these choices are only mathematical approximations of the same physical properties. Use of any other resolution criteria would not affect the arguments presented here, although the numbers (e.g.. those shown in Table I ) would change slightly. Some attention should also be given to special consid- erations for fluorescence microscopy (Rost, 1992). Since fluorescence is subject to fluorophore saturation and pho- tobleaching effects that do not affect other optical meth- ods, the highest possible light collection efficiency is de- sirable. This consideration dictates that the highest possi- ble NA should be used. However, the highest NA lenses (NA = 1.40) are usually of a "plan-Apochromat" design; this type of lens consists of up to 14 elements and has a lower transmittance than a "Fluor" design. Also, if aque- ous samples are used, the actual NA is limited to ~ 1 .2 because of total internal reflection for higher collection angles at the interface between water and coverslip (Inoue and Spring, 1997, pp. 53-55). For these reasons, fluores- cence from an aqueous sample appears brighter through a 100X/1.30 NA FLUAR (as shown in Fig. 1) than through a 100X/1.40 plan-Apochromat objective lens. Finally, it should be noted that improvements in almost every aspect of lens design and construction (e.g.. com- puter design of complex lens combination, automated grinding of arbitrary lens shapes, new optical materials for both lenses and coatings, and computer-controlled thin film deposition for precise optical coatings) make modern objective lenses superior to and more reliable than older lenses. Although many older lenses are superb, variables during their construction made finding a good one some- what challenging, and many researchers just took what came. Today's lenses are consistently of high quality, and also offer higher transmission efficiency and lower autofluorescence than did older lenses. Numerical aperture, magnification, and resolution in widefield microscopy In a digital widefield (conventional fluorescence) mi- croscope, the image is projected onto an imaging detector (usually a CCD camera) that takes the place of the eye. Thus, to optimize the information content of the resulting digital image, the pixels on the detector must be matched to the desired image resolution. As described above, the radius of a diffraction-limited spot, R dllT ~ X/(2 NA). is a good quantity to use for the definition of resolution. In the image plane, this spot will still give the smallest resolvable object, but the width of the spot will now be M R. Based on the Nyquist criterion, the desired sam- pling rate should be twice the resolution, so we want a pixel size in the object of R vm , n ~ R d ,n/2. In practice, the pixel size in widefield microscopy is fixed by the imaging camera used, so the magnification is the only variable that can be adjusted. For the purposes of these calculations, we can assume \ = 0.5 pm (a good approximation for fluorescein (FITC) imaging). We can determine the opti- mal M to be used for a given pixel size by matching the sampling resolution in the image plane to the pixel size (P) by P = M-R s ., mp . Table I shows the results of this calculation for two typical pixel sizes: 24 //m (an older SITe 512D CCD chip) and 6.8 /jm (the more modern Kodak KAF1400 CCD chip). As can be seen from the table, the older chips (with their larger pixel sizes) require higher magnification. For these larger pixels, an extra intermediate magnification of 2.5 would be required to maximize the information content of an image collected with a lOOx/1.3 NA objective lens, and in fact cameras that use the older SITe chip usually have some extra magnification built into them. As micro-fabrication tech- nology continues to advance, however, the need for high magnification lenses will decrease. Obviously, it is not possible to purchase a 72X/1.30 NA lens (although given the popularity of the KAF1400 CCD chip, perhaps it should be), so these optimal magnifications can only serve as a guide for selection of the best objective lens. Further calculations, such as those presented in the table, reveal that a camera with a pixel size of 5.4 //m would be ideal for use with many existing lenses, such as 60X/1.4 NA. 40X/0.90 NA. and 25X/0.60 NA. It should be noted, however, that as pixel sizes get smaller, the dynamic range of the detector may be reduced. For instance, the 5.4-/jm pixels would likely be filled by fewer than 20.000 counts, which would limit the detector to 14-bit dynamic range. This is in contrast to larger pixel sizes (i.e., the 24/L/m in the SITe 512D CCD chip), which can easily deliver > 16-bit dynamic range. For applications that require high precision, such as deconvolution methods. smaller pixels may be unworkable. Numerical aperture, magnification, anil resolution in laser scanning microscopy Much has been made of the improvement in resolution provided by confocal microscopy. But this improvement is at best minimal, and is only attained for extivmcK 4 D. W. PISTON Table I Optimal magnification for detectors with different pixel sizes calculated for five numerical apertures Numerical aperture (NAl .40 * Represents the SITe512D CCD chip. t Represents the Kodak KAF1400 CCD chip. 0.90 0.60 Calculated resolutions {//m) R-diff 0.18 0.19 0.28 0.42 0.83 samp 0.090 0.095 0.140 0.210 0.415 Magnification (x) 24-^m pixels* 267 253 171 114 58 6.8-/jm pixelst 76 72 49 32 16 small pinholes. In practical fluorescence microscopy, the pinhole must be opened somewhat to increase the effi- ciency of fluorescence collection. In fact, the pinhole is almost always opened enough to negate any resolution enhancement (Sandison et ui, 1995). In this practical case, there is an improvement in rejection of out-of-focus background, but the resolution is still given by R d , IT ~ \J (2 NA), so the ideal sampling resolution remains as shown in Table I. A key point in laser scanning microscopy is that there is no longer a fixed pixel size. Because the field over which the laser is scanned can be varied (this variation is usually called "zoom," or more appropriately "elec- tronic zoom," and is analogous to using an optical zoom lens), the sampling resolution can be easily changed. For this reason, users often have a lot of trouble choosing lenses when they switch to laser scanning confocal mi- croscopy. For instance, a 40X lens with a zoom factor of 2.5 is basically equivalent to a lOOx lens with no extra zoom. Thus, a 40X/1.3 NA lens should be chosen over an equivalent lOOx/1 .3 NA lens, because it offers a poten- tially larger field of view with no fall-off in light collec- tion or resolution. In laser scanning confocal microscopy, two other pa- rameters must be considered. First is the size of the detec- tor pinhole, which depends on the magnification. Most confocal microscopes have an adjustable pinhole that is easily set to match the magnification (e.g., for equiva- lence, a 60X lens needs a pinhole 1.5-fold larger than a 40x lens). Secondly, the lens design for confocal micros- copy may, in some cases, be more important than either M or NA. This is especially true for co-localization exper- iments, in which the chromatic corrections of a plan- Apochromat make it preferable despite its lower light- collection efficiency (the same trade-off must be consid- ered for any three-dimensional microscopies based on widefield and deconvolution methods, as well). Regard- less of the lens design, however, a lower magnification lens (of equivalent NA) is almost always preferable, be- cause it offers a larger field-of-view, and delivers equiva- lent resolution. Acknowledgments This work was supported by an Arnold and Mabel Beckman Foundation Young Investigator Award, NIH grant DK53434. and the Vanderbilt Cell Imaging Re- source (underwritten by CA68485 and DK20593). Literature Cited Agard, D. A., Y. Hiraoka. P. Shaw, and J. W. Sedat. 1989. Fluores- cence microscopy in three dimensions. Methods Cell Bio/. 30: 353- 377. Carrington, W. A., K. E. Fogarty. and F. S. Fay. 1990. 3D fluores- cence imaging of single cells using image restoration. Pp. 53-72 in Non-invasive Techniques in Cell Biology, i. K. Foskett and S. Grinstein, eds. Wiley-Liss, New York. Ghosh, R. N., and W. W. Webb. 1994. Automated detection and tracking of individual and clustered cell surface low density lipopro- tein receptor molecules. Biophys. J. 66: 1301-1318. Holmes, T. J., S. Bhatlacharyya, J. A. Cooper, D. Hanzel, V. Krish- 11:1111111 1 hi. W. Lin, B. Roysam, D. H. Szarowski, and J. N. Turner. 1995. Light microscopic images reconstructed by maxi- mum likelihood deconvolution. Pp. 389-402 in The Handbook of Biological Confocal Microscopy. 2nd Edition. J. Pawley, ed. Plenum, New York. Inoue, S., and K. R. Spring. 1997. Video Microscopy: the Fundamen- tals, 2nd Edition. Plenum, New York. Keller, H. E. 1995. Objective lenses for confocal microscopy. Pp. 1 1 1 - 1 26 in The Handbook of Biological Confocal Microscopy. 2nd Edition. J. Pawley. ed. Plenum. New York. Rost, F. W. D. 1992. Fluorescence Microscopy. Cambridge Univer- sity Press, Cambridge, UK. Sandison, D. R., D. W. Piston, R. M. Williams, and W. W. Webb. 1995. Resolution, background rejection, and signal-to-noise in widefield and confocal microscopy. App. Optics 34: 3576-3588. Reference: Biol. Bull. 195: 5-11. (August, 1998) Sickling of Anoxic Red Blood Cells in Fish FERENC I. HAROSI 1 *, IONE HUNT VON HERBING 2 , AND JEFFREY R. VAN KEUREN 3 ' Laboratory of Sensory Physiology, Marine Biological Laboratory, Woods Hole, Massachusetts 02543, and Department of Physiology, Boston University School of Medicine, Boston, Massachusetts 02118; 'School of Marine Sciences, University of Maine, Orono, Maine 04469; and 3 Biology Department, Woods Hole Oceanographic Institution. Woods Hole, Massachusetts 02543 The occurrence of the mutant hemoglobin Hb S in hu- man red blood cells results in sickle cell anemia. This disease, including its genetic and molecular bases, has been extensively investigated and is well understood (1.2). The presence of deoxy-induced sickling of animal erythro- cytes is largely unknown, however. We examined red blood cells (RBCs)from several fish species in vitro under aerated and anoxic conditions. Our polarized light micro- scopic techniques were aimed at establishing correlations bet\\'een erythrocyte morphology, state of oxygenation. spectral absorbance, linear dichroism, and linear bire- fringence. We found no fish with intracellular HbO 2 poly- merization; but there were intraerythrocytic aggregations of deoxy Hb with a high degree of either molecular order or disorder. The ordered aggregates in the RBCs of Atlan- tic cod, haddock, and toadfish were remarkably similar in dichroic ratio magnitudes and birefringence to those in human RBCs that contain HbS. Therefore, fish hemo- globins appear to be good models of sickling disorders and polymerization-related phenomena. The conse- quences of sickling on animal health and fish aquaculture remain to be studied. Aggregates of hemoglobin in fish erythrocytes have been reported since 1865 (cited in ref. 3). Nevertheless, former studies have failed to establish causal relationships between Hb aggregation, the state of oxygenation, and the attendant optical properties of erythrocytes (4-6). As a chance micro- spectrophotometric observation, we discovered linearly di- chroic absorption of light by Hb in larval fish erythrocytes. This linear dichroism i.e., the dependence of the absorp- Received 25 February 1998; accepted 5 June 1998. * To whom correspondence should be addressed. E-mail: fharosi@ mbl.edu tion coefficient on polarization direction was associated with abnormal RBC shape, or "sickling," the common term for this condition. The cells with the distorted morphol- ogy contained intracellular "bodies" with high concentra- tions of deoxyhemoglobin; for the sake of simplicity we call these bodies "erythrosomes." Erythrosomes are therefore intraerythrocytic. aniso- tropic aggregates of polymerized deoxy Hb that become visible in the light microscope because the refractive in- dex of Hb is higher than that of the surrounding medium. Erythrosomes exhibit linear dichroism as well as linear birefringence and occur in varied forms. In the RBCs of larval haddock and cod, for example, they may appear as V. U. or "banana" -shaped structures; but at other stages, or in different species, they may be shaped like needles or rods. The sickled fish erythrocytes encountered to date have always been associated with deoxy Hb, and not the oxygenated form. Indeed, the sickling that appeared was readily reversible by aeration; that is, the erythrosomes disappear and the affected cells resume their normal shape upon oxygenation. Moreover, repeated cycles of anoxia and normoxia cause concomitant cycles of intracellular hemoglobin aggregation and dispersion. In these charac- teristics, sickling in fish mimics sickling in man. Our initial observations of sickling were made on 8- and 60-day-old larval haddock (Melanogrammus aeglefinus, family Gadidae) as well as on 118- and 125-day-old Atlantic cod (Gadus morhua, family Ga- didae), all of which derived from stocks originating in the Bay of Fundy, Canada. Because we did not know whether sickling in fish is restricted to the developmental stages of particular species or is a more widespread phe- nomenon, we extended our investigation to wild-caught and cultured specimens of several species from four other geographical locations. F. I. HAROSI ET AL Typical spectral behavior of larval Atlantic cod RBC is illustrated in Figure 1. Erythrocytes with undistorted shape yielded the expected absorption spectrum for HbO : as shown in panel a: the Soret (or y ) band peaks at 414 nm and is not dichroic (dichroic ratio of 1 ). This is consistent with the presence of a random ensemble of HbO : mole- cules (a- and /3-band peaks are at 574 and 540 nm, respec- tively). In marked contrast, the Soret-band in panel b peaks at 430 nm and shows significant linear dichroism with a peak ratio of about 2. The latter is indicative of an orderly arrangement of the chromophores in situ (no a-band and shifted /? peak to 554 nm); thus the cod Hb behaves like the human Hb S. RBCs of juvenile Atlantic cod (age, 227 and 272 days) obtained from the Narra- gansett Laboratory of the National Marine Fisheries Ser- vice (Rhode Island), also sickled extensively (92%-95% of all cells) under anoxic conditions. Therefore, RBC sickling in cod and haddock is limited neither to an iso- lated stock nor to a specific age group, but is present in both larval and juvenile stages. Of the 12 fish species tested, however, only 4 species exhibited sickling under our experimental conditions. These were the two gadids (Atlantic cod and haddock), the toadfish (Opsunus l O ro .a 01 -Q 00 Normal, oxyHb 300 350 400 450 500 550 600 650 Wavelength [nm] Sickled, deoxyHb 300 350 400 450 500 550 600 650 Wavelength [nm] Figure 1. Larval Atlantic cod erythrocyte spectra under oxygenated (a) and deoxygenated (b) conditions: absorbance. A; linear dichroism. LD: and dichroic ratio, R. Absorbance was defined as A = log I/,//,), where /, and /, are transmitted fluxes through sample and reference volumes of the preparation. Linear dichroism was proportional to trans- mission modulation by sample anisotropy. which was calibrated with an external polarizer to yield 1.0 full-scale value. Polarized absorbance components were determined (26, 27) parallel with erythrosome length for Ay. and perpendicular to it for Aj_. The dichroic ratio was computed as R = A i /A,. HEMOGLOBIN POLYMERIZATION IN FISH Figure 2. Photomicrographs of adult toadtish erythrocytes (nucle- ated, as all non-mammalian erythrocytes [9]) obtained with a Zeiss Axiophot 2 microscope in differential interference contrast mode (objec- tive: Plan Apochromat, 63/1.4, oil immersion), (a) Normal cells con- taining HbO 2 with spectra like those of Fig. la. (b-d). Sickled cells under anoxic condition, each with spectra similar to those depicted in Fig. Ib. Scale burs, 10/jni. the erythrocytes, usually forming elliptical and closed structures. The transverse-to-tangential dichroic ratio around the cells were nearly uniform, with R ~ 1 .5. These results imply a flexible fibrous aggregation of Hb in which a degree of parallelism prevails. Unlike the toadfish case, tautog Hb polymerization seems to be too feeble to distort cells into "sickles"; rather it occurs along existing cellu- lar features, such as hooplike microtubules (9). Hemoglobin gelation in the summer flounder (Para- lichthys dentatus, family Bothidae) is markedly different from that in either the toadfish or the tautog. Anoxic erythrocytes in the flounder may contain intraerythrocytic clumps without sickling. Aggregates of deoxy Hb in flounder remained isotropic and did not gel into regular structures. They exhibited no linear dichroism (R 1). To distinguish isotropic from anisotropic aggregation, we regarded a cell to be sickled only when the Soret-band's linear dichroism was at least 10% above isotropic (i.e.. R a 1.1; see Table I). Because a linearly dichroic sample should also have anisotropic polarizability, sickle cells would be expected to exhibit linear birefringence as well. Indeed, this has been observed (10) and measured (3) for human sickle cells. We found fish erythrosomes to be also birefringent. Figure 3a illustrates how toadfish erythrosomes may "light up" between crossed polarizers in various color- ations depending on their orientation with respect to the polarizer's passing direction; they are the darkest at and 90. and brightest near 45 and 1 35. Retardance and absorbance spectra obtained from a typical toadfish erythrosome are depicted in Figure 3b. The retardance spectrum shows two peaks, one at 445- 450 nm and the other at 585-590 nm; these peaks corre- spond to the longwave half-maxima of the y and /3-bands (with peaks at 430 and 556 nm, respectively). Although the latter behavior is consistent with anomalous disper- sion (\ 1,12), we currently lack a quantitative interpreta- tion. The theory of anomalous dispersion describes varia- tions in the relationship between refractive index and wavelength through regions of strong absorption: the in- dex first declines with increasing wavelength, and then sharply rises before declining again to a higher plateau on the longwave side of an absorption band. The presence of the erythrosome's linear dichroism complicates mat- ters. Here, the stronger absorption coincides with the slow direction of retardance. and the weaker absorption with the fast direction. Therefore, the two principal refractive indices undergo unequal dispersion and thus yield an "anomalous retardance." Whereas we expected erythro- somes to have t\vo intrinsic birefringent components, one due to the heme groups and the other to the globin chains, their further delineation does not appear possible at pres- ent. Based on multiple determinations of retardance shown in Figure 3b (and assuming that the width of the erythrosome equals its thickness), the average specific retardance (n = 16) was in the range of 8.4 1.4 nm/pm (450 nm) to 2.7 0.6 nm//im (540 nm). Thus, the average difference in the principal refractive indices, n^.ii,, \ . was variable in the visible spectrum between 2.7-8.4 X 10 \ This range of values is to be compared with that of crystal- line quartz, where the difference between the principal refractive indices is 9 X 10"'. The explanation for eryth- rosome hues also follows from Figure 3b. Since light transmission of an erythrosome between crossed polariz- ers is proportional to retardance at each wavelength, and because peak retardance occurs around 450 nm, the trans- mitted flux should cause blue visual sensation at white illumination. If, on the other hand, retardance is also ele- vated for longer wavelengths (expected for thicker eryth- rosomes), increased transmission would result throughout the spectrum, yielding desaturated shades of blue. The principal axes of birefringence were directly ob- servable in polarized light by sample rotation. For toadfish erythrosomes. these correspond to their short and long dimensions, as can be seen in Figure 3a. However, a compensator is also needed if the slow direction is to be distinguished from the fast. To accomplish this, we used the Pol-Scope ( 13, 14), which is equipped with an auto- matic compensator. Figure 3c depicts brightness-encoded retardance images of several anoxic toadfish erythrocytes. F. I. HAROSI ET AL Table I Fish erythrocyte properties: absorbance maxima and sickling OxyHb \ mav [nm] DeoxyHb \ mjx [nm] Species Stage Sickling Sickling Atlantic cod (Gadus morhua) Larval 414 540 574 430 554 + (-100) Juven. 414 540 574 430 554 + (-95) Haddock (Melanogrammus aeglefinus) Larval T t t t 430 556 + (-100) Juven. 414 540 575 430 556 + (95) Adult 414 542 575 430 556 + (91) Toadfish (Opsanus tun) Adult 414 540 576 430 556 + (99) Tautog (Taittoga onitis) Larval 414 540 575 430 556 + (-90) Juven. 414 542 576 430 556 + (-1) Adult 414 542 576 430 556 Summer flounder ( Paralichthys di'inanix ) Juven. 414 542 576 430 560 Killitish (Fundulus heleroclinis) Juven. 414 544 575 430 556 Sheepshead minnow ( Cvprinodon variegatus ) Juven. 415 540 578 428 560 Blueback herring (Alosa aestevalis) Juven. 414 539 575 430 555 Atlantic silverside (Menidia menidia) Juven. 415 540 576 430 560 Smooth dogfish (Musteltis canis) Adult 414 544 575 430 556 Little skate (Raja erinacea) Adult 413 538 576 429 555 Common white sucker ( Catostomus commersoni) Adult 414 540 576 430 555 * In this column. + indicates the occurrence of optically anisotropic intraerythrocytic aggregates; percentage was obtained by visual inspection of cell morphology; other fractions were based on measurable Soret-band linear dichroism (R =: 1.1). The numeric values we regard only as approximate indicators of sickling, because polymerization is reversibly dependent on subtle variations in environmental conditions. Factors promoting sickling were anoxic conditions, low pH (6.9) and elevated temperature (29C). analogous to those reported previously for Hb S. t Unavailable data. revealing bundles of linear molecular aggregation. An- other cell from the same preparation is shown in Figure 3d, where an erythrosome retardance image with superim- posed black lines indicates local slow axis direction: this direction is approximately transverse to the long dimen- sion of the erythrosome. The line scans of erythrosome retardance are consistent with a model of fibrous molecu- lar order which, taken together with the slow direction of retardation and coincident strong absorption, is in har- mony with the presence of bundles of polymerized hemo- globin molecules with their planar porphyrin groups ori- ented nearly perpendicular to the length of the aggrega- tion. These features have also been found in normal human reduced-hemoglobin crystals and in sickle cells from anemia patients (3). Our polarized light microscopic observations are com- plementary to the structural information provided by elec- tron microscopy. Thomas (15) used the latter technique on red cells of cod (Gadus cullarias) and reported a para- crystalline organization for hemoglobin in erythrocytes. For atomic-scale comparison of fish hemoglobins with Hb S, it will be necessary to combine data from several techniques, as has been done for the various forms of human hemoglobin, including refined methods in electron microscopy (16), X-ray crystallography (17), and analy- ses of globin mutations based on amino acid sequence information ( 18). The current investigation notwithstanding, we can only speculate how widespread sickling is among the nearly 20,000 extant fish species. But given the richly varied environments that fishes inhabit, a great variability in fish hemoglobins is probable. Also probable is that the evolution of fish hemoglobins has responded to various environmental stresses in many species. Thus, we antici- pate fish hemoglobins to emerge as a rich resource for genetic, evolutionary, molecular, and pathophysiological studies. The variety of sickling hemoglobins may lead to new model systems, not only for a human disease, but also for studies of fundamental importance such as poly- merization, force generation (19), and the identification of molecular contacts that make paracrystal formation possible. The phenomenon of sickling may also have practical relevance to marine fish aquaculture. Fish in culture are often under heightened stress (e.g., insufficient oxygen- ation, high bacterial load, and low pH), and they are consequently prone to infections and high mortality rates. Because a propensity for sickling could increase mortali- HEMOGLOBIN POLYMERIZATION IN FISH s 400 500 600 Wavelength [nm] 700 CD O c CO Distance [ [im] 20 CD O CO CO o5 K T Distance [|j,m] 19 Figure 3. Toadfish erythrocytes. (a| Unstained, anoxic preparation between crossed polarizers at white light illumination (photographed in a Zeiss Axiophot 2 microscope on Kodak Ektachrome 160T film), (b) Retardance (solid) and absorbance (dashed) spectra of an isolated erythrosome (3.6 x 9 fjm). (c) Retardance images from sickled and ghost cells measured at 546 nm with the Pol-Scope ( 1 3, 1 4). The retardance for each picture element (pixel) was encoded in gray between (black) and 16 nm (white), (d) Another Pol-Scope image of retardance with two line scans (inserts): longitudinal (L) and transverse (T) to the erythrosome. Scale bars, 1 ^m. ties, the molecular characterization of sickling hemoglo- bins in fishes and their attendant physiological manifesta- tions are of paramount importance. The phenomenon is also relevant to restocking programs aimed at reestablish- ing dwindling groundfish populations. Summary Sickling in fish red blood cells is described, and the phenomenon is shown to occur in several species. By correlating the morphology of RBC with spectral absorp- tion of Hb in situ, sickling is associated with the presence of deoxyhemoglobin. The polymerization of deoxy Hb is inferred from optical anisotropies measured in single cells. The magnitude of linear dichroism and the nature of linear birefringence detected in fish "erythrosomes" show close kinship to the corresponding properties deter- mined previously from human sickle cells and hemoglo- bin crystals. The range of oxygen tension for sickling has not been determined and. therefore, we do not know whether it is within physiological limits. Also unknown is whether sickling in fish is pathological or, perhaps, reflective of a physiological condition with as yet obscure significance. In either case, however, further inw 10 F. I. HAROSI ET AL. lions promise to provide new insights into the requisite conditions that produce intracellular hemoglobin poly- merization. Methods Sample preparation. Experiments were conducted at room temperature (20-23C). Our primary buffer (ma- rine teleost Ringer's solution) contained (in mM): NaCl 140, KC1 2.5, CaCl 2 1.5, MgSO 4 1, NaH,PO 4 0.5, NaHCO, 0.5. and HEPES 10, adjusted to pH 7.3 with NaOH. This was used (air-saturated) in about 10-fold excess to dilute whole blood for oxygenated preparations. Samples for microscopy were prepared in between two No. 1 'A cover glasses sealed around the edge with a mol- ten mixture of paraffin and Vaseline. The usual procedure was to place 3 /J\ of blood on a thoroughly cleaned (20. 2 1 ) microscope cover glass, either with or without a poly- L-lysine (P-1274. Sigma) coating (22. 23). add and mix 15-30 /jl of desired buffer, cover, blot, and seal. Anoxic preparations were obtained in one of four ways: ( 1 ) By letting metabolically active blood cells deplete O : . The effectiveness of this method is variable: for extensive sickling to occur, it may take from a few hours to 2- 3 days. (2) By mounting erythrocytes and metabolically active retinal cells together in a sealed preparation, where complete deoxygenation may take only 30-120 min. This was the condition under which we first observed sickling. (3) By using N ; -purged dithionite solution (from 100 mM Na : S:O 4 stock) (24) and N : -purged buffer ( 1:1 ) in prepar- ing blood samples. (4) By the use of a calcium-free buffer (marine teleost Ringer without added calcium, containing 1 mM EGTA. at pH 6.9. and purged with N : for 1 h) and blood stored under light mineral oil (Paraffin, Fisher Scientific). The purpose of the oil is to keep the blood airtight, as previously reported (10). Additionally, we found that the oil prevents clotting for long periods (tested up to 8 weeks for toadfish blood); the latter method was found most effective in producing large, well-aligned erythrosomes. Determination of uhsorhancc and linetir ilichro- /.v;/i. We used the dichroic microspectrophotometer (25- 27) with a measuring beam cross section of about 0.6 X 2 fj,m on cells in optical isolation. This instrument is a single-beam, phase-modulated, polarized light micro- scope equipped with automatic, wavelength-scanning and recording photometry. Glycerin immersion-type micro- scope objectives were used for both condenser (32X/0.4 Ultrafluar. Zeiss) and objective ( lOOx/1.3 UV-F100, Ni- kon). Visualization was implemented with infrared light and a video camera system with tape recording. Lineur birefringence determination. The laboratory in- strument was built around a modified Axiovert 10 (Zeiss) microscope utilizing quartz components and fiberoptic links between a 150-W xenon light source (Oriel Corp.) and a spectrograph-detector combination. Its adjustable cross section microbeam (about 1.5 X 4 /vin) was passed through a rotatable polarizer (Glan-Thompson type); the condensor (32X/0.4 Ultrafluar, Zeiss); the preparation af- fixed to a rotatable, sliding stage; the objective (lOOx/ 1.2 Ultrafluar, Zeiss); and a slidable analyzer (HN38S, Meadowlark Optics). The transmitted light was collected and focused on the entrance slit of a spectrograph (MonoSpec 18, Jarrell Ash), which in turn dispersed it over an intensified diode array detector (Model 1455B, EG&G PARC). Electronic scanning of the latter provided records of spectral responses in the range of 350-650 nm. This instrument was also capable of determining ab- sorbance and linear dichroism from static transmission measurements taken at appropriate sample orientations and polarizer settings. For the determination of linear birefringence, the preparation was positioned between crossed polarizers with the sample's principal ("slow" and "fast") axes oriented at 45 to the direction of polarization. The basis for signal processing was provided by the formula /(\) = /,,(\){siir[F(\)/2] + 1/EF). where /(\) is the transmitted flux at sample retardance of F(\) radians. /,,(\) the flux through the system for polarizer and analyzer axes set parallel [with F(\) = 0]. and EF is the extinction factor defined as EF = / r (\)//,(\) with /, (\) being the transmitted flux for crossed polarizers (28). It followed from the foregoing that sin : [F(\)/2] = |/(\) - /, (\)]//,,(\). where from the retardance (in nm) at wave- length \ (nm) could be obtained as F(\) = ([/(\) - /, (\)]/ / ; ,(\)} l/: X/TT (assuming equality between the sine of small angles and their values in radians). The latter relationship was implemented on data sets from spectral scans of 7(\). /,(\) and /,,(\) by software operations using ORIGIN (Mi- crocal Software). Birefringence is linked to retardance by the optical path difference in the sample between the fast and slow waves propagating through thickness \i>lic cuiiipiiriments of gastropod hepatopancreas * Enzyme activity (/jmol/min/g wet tissue weight )$ Genus Fractiont CCO CS DL-BHBDH D-BHBDH L-BHBDH LDH Stylommatophora Archachatina hepatopancreas tissue n.ni n.ni- n.m. 0.13 0.06 0.77 0.09 n.m. heart tissue n.m n m. n.m. 35.74 8.60 n.d. n.m. kidney tissue n.ni n m. n.m. 0.54 0.10 0.58 0.08 n.m Brad\baena mito 1.25 0.10 0.83 0.1 1 0.14 0.08 0.21 0.12 0.14 0.08 0.33 0.1 cytosol 0.16 0.06 0.22 0.06 0.85 0.06 0.23 0.08 0.98 0.14 4.35 0.8 Arion mito 1.10 0.11 1.90 0.10 n.d. n.d. n.d. n.m cytosol 0.03 0.01 0.21 0.04 2.44 0.09 0.07 0.04 2.60 0.11 n.m Basommatophora Physa mito 0.87 0.08 4.71 0.50 1 .02 0.28 3.07 0.73 n.d. n.m cytosol 0.17 0.08 1.37 0.34 0.53 0.18 0.99 0.19 0.22 0.04 n.m Helisoma mito 4.85 0.65 3.58 0.29 0.52 0.15 1.94 0.44 n.d. n.ni cytosol 0.04 0.04 0.39 0.08 0.12 0.05 0.13 0.04 0.09 0.01 n.m Stagnicola hepatopancreas mito 2.81 0.21 1.40 0.20 2.97 0.14 7.00 0.67 n.d. 0.38 0.2 cytosol 0.76 0.08 0. 1 8 0.06 0.67 0.10 1.01 0.19 n.d. 1.25 0.2 heart mito 9.48 0.85 5.72 3.66 n.ni. 13.20 1.96 n.d. 6.00 3.3 cytosol 5.01 2.96 1.95 0.55 n.m. 1 1.61 1.85 n.d. 34.57 1.7 Neritopsina Helicina mito 6.54 1.24 1.49 i* O.I 1 0.45 0.09 0.53 0.32 n.d. 0.17 0.1 cytosol 0.31 0.11 0.37 0.25 0.25 0.18 0.21 0.10 n.d. 1.72 0.8 Architaenioglossa Campeloma mito 2.12 0.12 1.82 0. 1 1 0.46 0.27 0.65 0.04 n.d. n.d. cytosol 0.01 0.01 0.06 0.02 0.07 0.02 0.15 0.06 n.d. n.d. Pomacea hepatopancreas mito 1.82 0.62 0.89 0.45 1 .68 0.63 1.53 0.70 n.d. 0.44 0.2 cytosol 0.03 0.01 0.25 0.06 0.19 0.1 1 0.33 0.13 n.d. 0.48 0.01 heart mito 3.25 1.35 17.05 1.87 n.m. 0.55 0.18 n.d. 0.32 0.1 cytosol 0.81 0.37 14.18 2.73 n.m. 0.09 0.05 n.d. 7.73 2.3 kidney mito 2.97 1.83 4.09 0.56 n.m. 0.47 0.09 n.d. 0.44 0.3 cytosol 0.15 0.10 0.31 1 0.05 n.m. 0.14 0.10 n.d. 0.92 0.4 * Most gastropods were collected from fields and ponds near the University of Guelph. in Guelph, Ontario, Canada, or purchased (Pomacea, Campeloma) from local aquarium stores. The terrestrial prosobranchs. Helicina orbiculata. were collected near Jacksonville. Florida. The giant African snails, Archachatina ventricosa, were from our laboratory population established with animals provided by the Toronto Metro Zoo. All snails and slugs were kept in terraria or aquaria at room temperature (22 2C) and fed lettuce. We observed no effect of duration of time spent under these conditions on BHBDH compartmentation or enantiomeric specificity. Taxonomic classification of gastropods was as in Ponder and Lindberg (101 (Table II). A. ventricosa tissues were not fractionated. They were prepared as in Stuart and Ballantyne. (4) For enzyme measurements in all other gastropods, excised tissues were placed in approximately 20 volumes of sucrose buffer ( 100 mM sucrose. 20 mM N-[2-hydroxyethyl]piperazine-N'-[2-ethanesul- fonic acid] (HEPESl. pH 7.51. Cells were disrupted by five passes of a Potter-Elvejhem homogenizer operated at low speed (<100 rpm). This homogenate was centrifuged at KXOOOx g for 10 min to separate it into mitochondnal (pellet) and cytosolic (supernatant) fractions. The mitochondrial pellet was resuspended in a volume of buffer equal to the original buffer addition, thus maintaining equal the dilution of both fractions. Both fractions were sonicated with three 5-s bursts of a Vibra-Cell sonicator (Sonics & Materials Inc., Danbury, CT) set to 80% output, 50 watts. These preparations were used directly in the measurement of enzymes. We used CCO to mark the mitochondrial membrane, CS to mark the leakage of matrix enzymes from mitochondria damaged in the fractionation process, and LDH to evaluate the contamination of the mitochondrial fraction with cytosolic enzymes. CCO, CS, and LDH were measured as in Stuart and Ballantyne. (4) The assays for D-, L- and DL-BHBDH contained 2 mM NAD and either 200 mM D- or L-BHB or 400 mM DL-BHB. respectively (BHB omitted for control) in 50 mM imidazole. pH 8.0. All chemicals were purchased from Sigma Chemical Co. (St. Louis, MO). t "tissue" = whole tissue homogenate: "mito" = mitochondrial fraction; "cytosol" = cytosolic fraction. t CCO = cytochrome C oxidase; CS = citrate synthase; LDH = lactate dehydrogenase. DL-. D- and L-BHBDH = activity measured with racemic mixture of DL-/3-hydroxybutyrate, high-purity D-/3-hydroxybutyrate, and high-purity L-/3-hydroxybutyrate. respectively, n.d. = not detected; n.m. = not measured. Values are means standard error; n = 5. 14 J. A. STUART ET AL low L-BHBDH activity observed in the mitochondrial fraction thus can be attributed to cytosolic contamination. Low D-BHBDH activities in Bradvbaena, Arion, and Archachatina hepatopancreas were likely due to contami- nating i -BHB in the commercial D-BHB preparation (L-BHB contamination was up to 2%). As we used a [o-BHB] of 20 mM in the assays, as much as 0.4mA/ L-BHB could have been present. Heart and kidney were fractionated in the same way as the hepatopancreas. Fractionation of heart tissue re- sulted in a lower recovery of mitochondria and greater leakage into the cytosol of mitochondrial matrix enzymes (Table I), perhaps because of the small size of these tis- sues (Stagnicola hearts averaged 0.8 mg; Pomacea hearts averaged 10 mg). Nonetheless, the distribution of BHBDH activity approximated those of CCO and CS in these tissues, indicating that BHBDH activity is localized to the mitochondria in basommatophoran heart and archi- taenioglossan heart and kidney. In all of these tissues, BHBDH was specific for D-BHB, with no oxidation of L-BHB observed. The pattern of D- and L-BHBDH distribution in Archti- chatimi (Table I) parallels that seen in Cepaea. In C. nemoralis tissues, no mitochondrial form of BHBDH was found (5). Instead, a cytosolic L-BHBDH was present in hepatopancreas and kidney, and a cytosolic D-BHBDH was found in heart and kidney. Similarly, D-BHB was oxidized by Archaclmtina heart and kidney homogenates and L-BHB was oxidi/ed by kidney and hepatopancreas. This configuration of BHBDH organization appears to be characteristic of the Stylommatophora. We investigated the evolution of cytosolic BHBDH isoforms by mapping the occurrence of the mitochondrial D-BHBDH and cytosolic L-BHBDH of gastropod hepato- pancreas onto a gastropod phylogeny (8) using the Mac- Clade (9) software program (Fig. 1 ). Mollusc phylogenies based upon morphological (10) and molecular (8, 11-13) data support the monophyly of Gastropoda, within which pulmonates are monophyletic. Within Pulmonata, the Sty- lommatophora and Basommatophora are each also mono- phyletic. For this analysis, we have considered the presence of mitochondrial D-BHBDH to be the ancestral condition. Though BHBDH activity is undetectable in marine gas- tropods, it is found in all freshwater and terrestrial gastro- pods that we have studied. Thus, the enzyme appears to have first arisen in a gastropod ancestral to these groups. Our phylogenetic analysis suggests that the cytosolic L- BHBDH isoform evolved a single time in an ancestral stylommatophoran. Gastropods of this order are almost exclusively terrestrial and compose the vast majority of terrestrial snails and slugs (14). To test whether the pres- ence of L-BHBDH correlates with terrestriality, we in- cluded the distantly related terrestrial gilled gastropod Helicina orbiculata Pomacea bridges* Campeloma decisum Bradybaena similans Cepaea nemoralis Anon subfuscus Archachatina ventncosa Physa gynna Biomphalaria glabrata O Stagnicola elodes Heltsoma tnvolvis Figure 1. Phylogenetic relationships among freshwater and terres- trial gastropod taxa used in this study. Unfilled lines represent the pres- ence of mitochondrial D-/3-hydroxybutyrate dehydrogenase (and absence of L-/3-hydroxybutyrate dehydrogenase). Filled lines denote the presence of cytosolic L-/3-hydroxybutyrate dehydrogenase (and absence of D-/3- hydroxybutyrate dehydrogenase). Hatched line = equivocal occurrence of L-/3-hydroxybutyrate dehydrogenase, denoting that the enzyme arose at an undetermined point along this lineage. Cepucti iirmoralis data are from Stuart and Ballantyne (5). Biomphalaria xlahmhi data are from Meyer ct al. (15). Helicinci (Neritomorpha) in our analysis. The absence of i -BHBDH in Hclichui. however, suggests that the occur- rence of this isoform correlates exclusively with phyloge- netic position. A general upregulation of BHBDH activities appears to have occurred in tissues of pulmonate gastropods. In pulmonate hearts, exceptionally high activities of D- BHBDH (approximately two orders of magnitude greater than in Pomacea heart) (Table I) suggest that D-BHB is particularly important as a metabolic substrate in these tissues. High activities of all enzymes of ketone body metabolism in Sttignicolu cloiles. C. lu'iuomlis, and A. vcntricosa indicate a substantial flux through this pathway in all pulmonates. D-BHB levels in hemolymph are as high as those of glucose in the basommatophoran pulmo- nate Biomplnihirici i>lcihrcitii (15). These snails, and tissues isolated from them, actively oxidize ketone bodies. BHB K.ETONE BODY METABOLISM IN GASTROPODS 15 Table II Higher classification f 10) of gastropod species used for measurements of BHBDH activity' Gastropoda Orthogastropoda Neritimorpha Helicinoidea Helicina orhiculata I terrestrial) Apogastropoda Architaenioglossa Viviparoidea Campeloma decisum (freshwater) Pomacea bridges: (freshwater) Heterobranchia Euthyneura Pulmonata Basommatophora Physa gyriiui (freshwater) Helisoma trivolvis (freshwater) Stagnicola elodes (freshwater) Stylommatophora Arcliacluirinu ventricosa (terrestrial I Brad\haena similaris (terrestrial) Arion subfuscits (terrestrial) appears, therefore, to be an important metabolic substrate in pulmonate snails. Unlike mammals, these snails show a decrease in D-BHB levels in the hemolymph during starvation, suggesting that pulmonates may routinely use ketone bodies as energy substrates, whereas mammals confine their use to times of starvation. This difference may be related to a decreased emphasis on amino acid metabolism and a low capacity for extrahepatic oxidation of fatty acids in pulmonates. The metabolic organization of tissues in these organisms suggests that they use ketone bodies which, unlike fatty acids, are freely soluble, as a means of transporting lipid carbon from central stores to peripheral tissues for oxidation. In stylommatophoran pulmonates, upregulation of BHBDH activity has been followed by the elaboration of new isoforms of the enzyme. This may have occurred through an initial loss of the transmembrane amino acid sequence from the membrane-bound mitochondria! BHBDH. to allow the enzyme to function in the cytosol. In stylommatophoran hepatopancreas and kidney, this en- zyme may have been further modified to act upon L- BHB rather than D-BHB. However, this scenario for the occurrence of cytosolic BHBDH isoforms assumes diver- gence from the ancestral mitochondria! D-BHBDH. Alter- natively, the cytosolic enzymes could derive from other proteins, unrelated to the mitochondria] o-BHBDH, and have achieved functional similarity through evolutionary convergence. This convergence appears to have occurred in the evolution of D- and L-LDHs in bacteria (16). On the other hand, divergence of cytosolic L-BHBDH from cytosolic D-BHBDH is suggested by the difficulty of sepa- rating these isoforms when they are electrophoresed to- gether on a gel that separates proteins on the bases of size and charge (5). Analyses of primary structures are necessary to determine the relatedness of the three BHBDH isoforms. The use of both enantiomers of a single metabolic sub- strate in routine energy metabolism is unusual in the ani- mal kingdom, though other examples of this phenomenon exist among molluscs. Both D- and L-alanine are found in tissues of some marine bivalves. The occurrence of D- alanine appears to be related to a role in osmoregulation (17). Both D- and L-specific isoforms of LDH also occur within individual cephalopods (18). although the physio- logical significance of these has not been established. The advantages of the stylommatophoran cytosolic BHBDH isoforms are also not immediately obvious. The existence of both D- and L-BHBDH may allow the metabolic parti- tioning of BHB between specific tissues. Enzyme activi- ties indicate that ketone bodies could be synthesized in the kidney from fatty acids under normal conditions. Both D- and L-BHBDH are present in stylommatophoran kid- neys, which are thus able to produce both forms of BHB. Each of these enantiomers of BHB may be specifically targeted to a tissue, with D-BHB being oxidized by heart and L-BHB by hepatopancreas. This adaptation could be related to the apparently greater role for ketone bodies in the intermediary metabolism of pulmonate gastropods i.e.. a refining of a much-used pathway. The phylogenetic pattern of hepatopancreas BHBDH stereospecificity and subcellular distribution in gastropods suggests that L-BHBDH. and perhaps also D-BHBDH. could be valuable characters for assessing phylogenetic relationships within the Gastropoda. Both enzymes can be rapidly and inexpensively assayed. The presence of hepatopancreas L-BHBDH may be a useful defining char- acteristic of the Stylommatophora. As such, it will be especially interesting to identify which isoforms of BHBDH are present in tissues of the Archiopulmonata, a group of much-debated phylogenetic position. Cer- tainly, the presence of L-BHBDH in stylommatophoran gastropods should be noted by population geneticists, as staining of electrophoretic gels of gastropod tissues with racemic DL-BHB mixtures will give results that are a function, in part, of the phylogenetic position and tissue of the snail examined. Acknowledgments We thank John Colbourne for help with phylogenetic analyses. Also, thanks to Drs. Fred G. Thompson and Harry G. Lee. and Tom Mason, Invertebrate Curator at the Toronto Metro Zoo, for their assistance with tindi and identifying gastropod species. 16 J. A. STUART ET AL. Literature Cited I . Nevvsholme, E. A., and A. R. Leech. 1983. Bioehemixtry for the Medical Sciences. John Wiley. New York. 2 Robinson, A. M., and D. H. Williamson. 1980. Physiological roles of ketone bodies as substrates and signals in mammalian tis- sues. Phyxinl. Rev. 60: 143-187. 3. Beis, A., V. A. Zamniit, and E. A. Newsholme. 1980. Activities of 3-hydroxybutyrate dehydrogenase, 3-oxoacid CoA-transferase and acetoacetyl-CoA thiolase in relation to ketone-body utilisation in muscles from vertebrates and invertebrates. Eur. J. Biochcni. 104: 209-215. 4. Stuart, J. A., and J. S. Ballantyne. 1996. Correlation of environ- ment and phytogeny with (he expression of /3-hydroxybutyrate de- hydrogenase in the mollusca. Camp. Biochem. Phyxiol. 114B: 153- 160. 5. Stuart, J. A., and J. S. Ballantyne. 1997. Tissue specific forms of /3-hydroxybutyrate dehydrogenase oxidize the D- or L-anantiom- ers of /3-hydroxybutyrate in the terrestrial gastropod Cepaea neino- rulix. J. Exp. Zool. 278: 115-1 IS. 6. Stuart, J. A., and J. S. Ballantyne. 1997. Importance of ketone bodies in the intermediary metabolism of the terrestrial snail Archu- clnirinii vi'ntrieoxu: evidence from enzyme activities. Camp. Bio- cliem. Phyxiol. 117B: 197-201. 7. Lehninger, A. L.. H. C. Sudduth, and J. B. Wise. 1960. D-beta- hydroxybutyric dehydrogenase of mitochondria. J. Bid. Chein. 235: 2450-2455. 8. Harasewych, M. G., S. L. Adamkevvicz, J. A. Blake, D. Saudek. T. Spriggs, and C. J. Bull. 1997. Phytogeny and relationships of pleurotomariid gastropods (Mollusca: Gastropoda): an assessment based on partial 18S rDNA and cytochrome c oxidase I sequences. Mol. Mar. Bin/. Biotech. 6: 1 -20. 9. Maddison, W. P., and D. R. Maddison. 1992. MucCltu/e: Analy- sis nt Pli\lg. Ser. 3: 317-328. 24 Forsythe. J. W., and W. F. Van Heukelem. 1987. Growth. Pp. 135-156 in Cephalopod Life Cycles. Vol. 2. Comparative Reviews. P. R. Boyle, ed. Academic Press, London. 25. Rosenberg, A. A., K. F. Wiborg, and I. M. Beck. 1980. Growth of Todarodes sagittatus (Lamarck) (Cephalopoda: Ommastrephi- dae) from the Northeast Atlantic, based on counts of statolith growth rings. Sarsiu 66: 53-57. 26. Forsythe, J. W., and R. T. Hanlon. 1989. Growth of the Eastern Atlantic squid, Loligo forbesi Steenstrup (Mollusca: Cephalopoda). Aquacult. Fish. Manage. 20: 1-14. 27. Forsythe, J. W. 1993. A working hypothesis on how seasonal temperature change may impact the growth of young cephalopods. Pp. 1 33- 143 in Recent Advances in Cephalopod Fisheries Biologv. T. Okutani. R. K. O'Dor. and T. Kubodera. eds. Tokai University Press. Tokyo. 28. Grist, E. P. M., and S. des Clers. 1998. How seasonal tempera- ture variations may influence the structure of annual squid popula- tions. IMA J. Math. Appl. Med. Biol. 15: I -23. 29. Arkhipkin, A., and A. Mikheev. 1992. Age and growth of the squid Sthenoteuthis pteropus (Oegopsida: Ommastrephidae) from the Central East Atlantic. J. E\p. Mar. Biol. Ecol. 163: 261-276. 30. Arkhipkin, A., and V. Laptikhovsky. 1994. Seasonal and in- terannual variability in growth and maturation of winter-spawning llle.f argentinus (Cephalopoda, Ommastrephidae) in the Southwest Atlantic. At/uat. Living Resour. 7: 221-232. 31. Gould, S. J. 1977. Ontogeny and Phvlogeny. Harvard University Press, Cambridge. 501 pp. 32. Jackson, G. D., and J. H. Choat. 1992. Growth in tropical cepha- lopods: an analysis based on statolith microstructure. Can. J. Fish. Ai/uat. Sci. 49: 218-228. 33. Alford, R. A., and G. D. Jackson. 1993. Do cephalopods and larvae of other taxa grow asymptotically? Am. Nat. 141: 7 17-728. 34. Pauly, D. In press. Why squids though not fish may be better understood by pretending they are. S. Afr. J. Mar. Sci. Reference: Bioi Bull. 195: 21-29. (August. 1998) Metamorphic-Signal Transduction in Hydroides elegans (Polychaeta: Serpulidae) Is Not Mediated by a G Protein ERIC R. HOLM, BRIAN T. NEDVED, EUGENIC CARPIZO-ITUARTE, AND MICHAEL G. HADFIELD* Kewalo Marine Laboratory, University of Hawaii, 41 Ahui St., Honolulu, Hawaii 96813 Abstract. Evidence from larvae of hydrozoans, gastro- pods, and barnacles suggests that G protein-coupled re- ceptors mediate induction of settlement and metamorpho- sis in response to environmental cues. We examined re- sponses of larvae of the serpulid polychaete Hydroides elegans to neuropharmacological agents to determine if G protein-coupled receptors or their associated signal- transduction pathways regulated induction of metamor- phosis by bacterial cues. Larvae of Hydroides elegans metamorphose rapidly and in high proportions when ex- posed to bacterial biofilms. Neither the G-protein activa- tor Gpp[NH]p nor the inhibitor GDP-/3-S affected meta- morphosis. Although the nonspecific phosphodiesterase inhibitors IBMX, theophylline, and papaverine induced larvae to metamorphose, RO-20-1724 (an inhibitor selec- tive for cAMP-specific phosphodiesterase IV) and the cy- clic nucleotide analogs db-cAMP and db-cGMP had no effect on metamorphosis. The adenylate cyclase activator forskolin inhibited responses of larvae to inductive bacte- rial biofilms. These apparently conflicting results may be due to side effects of IBMX, theophylline, papaverine, and forskolin on ion transport. The phorbol ester TPA, an activator of protein kinase C, also had no effect on larval metamorphosis. These experiments indicate that G protein-coupled receptors and signal transduction by the Received 23 December 1997; accepted 27 May 1998. * To whom correspondence should be addressed. E-mail: hadfield hawaii.edu Abbreviations: AC/cAMP. adenylate cyclase/cyclic AMP; db-cAMP. dibutyryl-cyclic AMP; db-cGMP, dibutyryl-cyclic GMP; DMSO, di- methyl sulfoxide; FSW, filtered seawater; GDP-/3-S, guanosine S'-O- (2-thiodiphosphate); Gpp[NH]p. 5'-guanylylimidodiphosphate; IBMX. 3-isobutyl-l-methylxanthine; PI/DAG/PKC, phosphatidylinositol/di- acylglycerol/protein kinase C; TPA, phorbol-12-myristate-13-acetate. adenylate cyclase/cyclic AMP or phosphatidyl-inositol/ diacylglycerol/protein kinase C pathways are not compo- nents of the morphogenetic pathway that is directly re- sponsible for processing metamorphic cues in H. elegans. Introduction Several decades of research have demonstrated the im- portance of biochemical and physical cues in determining, on small spatial scales, where larvae of marine inverte- brates settle and metamorphose (see Scheltema. 1974: Crisp, 1984; Svane and Young, 1989, for reviews). Identi- fication of the exact nature and ecological role of exoge- nous cues remains an active field of study. Over the last 20 years, however, increased attention has focused on the physiological mechanisms underlying detection of these cues and the initiation of metamorphosis. The approach generally taken in this work consists of exposing larvae to solutions of potentially neuroactive compounds, in the presence or absence of a cue, and noting the responses. Although such experiments may be difficult to interpret (Pawlik, 1990; Leitz, 1997), a common theme is devel- oping from the results. In the marine invertebrate larvae that have been most extensively studied, it appears that G protein-coupled re- ceptors and their associated signal-transduction pathways play an important role in regulating metamorphosis. In the hydrozoans Hydractinia echinata (Miiller, 1985; Leitz and Miiller, 1987; Leitz and Klingmann, 1990; Schneider and Leitz, 1994) and Mitrocomella polydiademata (Free- man and Ridgway, 1990), the phosphatidylinositol/diacyl- glycerol/protein kinase C (PI/DAG/PKC) pathway trans- duces the metamorphic signal provided by bacterial cues. In addition, Leitz and Wirth ( 1991 ) found that H. echinata 21 22 E. R. HOLM ET AL metamorphosed in response to the G-protein activators ortho- and metavanadate, suggesting that G protein-cou- pled receptors are involved in the process. Schneider and Leitz ( 1994) proposed that the bacterial inducer of meta- morphosis acted by first binding to a G protein-coupled receptor. In the barnacle Balanns ainphi trite, two signal - transduction systems, the adenylate cyclase/cyclic AMP (AC/cAMP) pathway (Rittschof et ui. 1986; Clare et ui, 1995) and the PI/DAG/PKC pathway (Yamamoto et al, 1995; Holm et al., unpubl. data), appear to regulate meta- morphosis. Clare (1996) proposed a model linking the two pathways; in this model the exogenous metamorphic cue was bound to a G protein-coupled receptor. Finally, in the abalone Ha/iotis rnfescens, the AC/cAMP pathway transduces the metamorphic signal provided by com- pounds found on the surface of crustose coralline algae, while the PI/DAG/PKC pathway facilitates the response to these compounds (Morse et al., 1980; Trapido-Rosen- thal and Morse, 1985, 1986; Baxter and Morse, 1987; Morse, 1990, 1991 ). The receptor of the metamorphic cue did not appear to be associated with a G protein, but evidence suggested that the separate, facilitating pathway was activated by binding of amino acids to a G protein- coupled receptor (Baxter and Morse, 1987. 1992; Wo- dicka and Morse, 1991). G protein-coupled receptors and their associated signal- transduction pathways present a compelling general model for detection of a variety of stimuli in an exception- ally broad range of organisms (Carr, 1992). It is not clear, however, whether the regulation of metamorphosis by G proteins is a general feature of marine invertebrate larvae (Hadfield. 1998). For larvae of Huliotis rnfescens, at least, a G protein-coupled receptor does not bind the metamor- phic signal (Baxter and Morse, 1987). There are several other classes of receptors (Feger et ai, 1994) that could conceivably function in marine larvae to detect exogenous cues for settlement and metamorphosis. These include lectins, which have been implicated in the metamorphosis of larvae of the spirorbid polychaete ./anna hmsiliensis (Kirchman et al., 1982; Maki and Mitchell, 1985), and ligand-gated ion channels. To test the hypothesis that G protein-coupled receptors or the AC/cAMP and PI/DAG/PKC signal-transduction pathways regulate metamorphosis in the serpulid poly- chaete Hydroides elegans, we examined the response of larvae of this species to various neuropharmacological agents. Hydroides elegans (Haswell, 1883) is a common member of the shallow subtidul fouling community throughout tropical and warm temperate seas (Hadfield et al.. 1994). In the laboratory, competent larvae settle and metamorphose after as little as 15 min of exposure to bacterial biofilms ( Hadfield et ul.. 1994; Carpizo-Ituarte and Hadfield, 1998). Experiments with larvae of other polychaetes, including Capilelln <idcs elegans. Except for Gpp[NH]p and RO-20-1724, the concentra- tion ranges we tested for these compounds spanned effec- tive concentrations previously determined to induce meta- morphosis in larvae of other marine invertebrates. We could find no comparable data for Gpp[NH]p and RO- 20-1724. RO-20-1724 was obtained from Calbiochem, San Diego, California; all other compounds were pur- chased from Sigma, St. Louis, Missouri. Statistical analysis The results of the experiments were analyzed using either two-sample /-tests or Wilcoxon rank-sum tests. The percentage of larvae metamorphosing in each replicate petri dish was subjected to the angular transformation, and the means and variances for each treatment were calculated. Variances for the two treatments to be com- pared were then tested for homogeneity using the F-max test (Sokal and Rohlf, 1981). If variances were homoge- neous, the transformed data were compared by two-sam- ple /-test. Otherwise the untransformed data were ana- lyzed by the nonparametric Wilcoxon rank-sum test. All /-tests were calculated by hand; the nonparametric tests were conducted using the NPAR 1 WAY procedure in SAS (SAS Institute Inc.. 1989). Significant results were identi- fied using the sequentially rejective Bonferroni procedure (Holm, 1979; Rice, 1989) to correct for multiple tests. Results Compounds affecting the activity of G proteins Neither the G-protein activator Gpp[NH]p nor the in- hibitor GDP-/?-S affected metamorphosis of the larvae of Hydroides elegans (Fig. 1A. B). In the absence of a stimulatory bacterial biofilm, Gpp[NH]p did not induce metamorphosis (Fig. 1A), and GDP-/2-S failed to inhibit metamorphosis in response to a biofilmed surface (Fig. IB). Compounds affecting the AC/cAMP transduction pathway In two separate trials, the phosphodiesterase inhibitors IBMX, theophylline, and papaverine all induced meta- morphosis in the absence of a biofilm (Fig. 2A, B). Con- centrations at which we observed maximal responses were the same in both trials; 10 4 M for IBMX (Mest. P 0.001. both trials), 10 - M for theophylline (/-test. P < 24 E. R. HOLM ET AL Table I Compounds tested for effects on metamorphosis of larvae o/Hydroides elegans Activity Predicted effect on metamorphosis Compounds affecting G proteins Gpp [NH] p GDP-/3-S Compounds affecting the AC/cAMP signal-transduction pathway IBMX Theophylline Papaverine RO-20-1724 db-cAMP db-cGMP Forskolin Compounds affecting the PI/DAG/PKC signal-transduction pathway TPA Activates G proteins Induction Inhibits activation of G proteins Inhibition Inhibits phosphodiesterases* Induction Inhibits phosphodiesterases* Induction Inhibits phosphodiesterases* Induction Inhibits cAMP-specific phosphodiesterase IV* Induction Increases intracellular cAMP Induction Increases intracellular cGMP Induction Activates adenylate cyclase* Induction Activates protein kinase C Induction A brief description of the activity of each compound is provided (Activity). Predicted Effect describes the result one would expect if G proteins. or the affected signal-transduction pathway, mediated responses of larvae to metamorphic cues. See text for details. * Activity increases intracellular cAMP or cGMP. 0.001, both trials), and 10 5 M for papaverine (/-test, P < 0.001, trial 1; Mest, P < 0.005, trial 2). IBMX at \0~ 3 M caused a large number of larvae to metamorphose without constructing a primary or secondary tube. Figure 2 shows only the percentage of larvae that had undergone normal metamorphosis (characterized by permanent at- tachment to the substratum and construction of either a primary or secondary tube, see Materials and Methods) when exposed to IBMX at IQ~* M. Papaverine was a much less effective inducer of metamorphosis than either IBMX or theophylline, with a maximum mean percentage of metamorphosis < 20%, as compared to > 80% for both IBMX and theophylline (Fig. 2A, B). Exposure of larvae to theophylline at 10~ 2 M was fatal. In the second trial (Fig. 2B) both IBMX (rank-sum test, P < 0.05) and papaverine (rank-sum test, P < 0.05) inhibited normal metamorphosis at their highest concentrations, although the effects were not strong. These compounds may have been toxic at the highest concentrations we tested. The selective phosphodiesterase inhibitor RO-20-1724 exhib- ited no inductive effects (Fig. 3). Neither the adenylate cyclase activator forskolin, the cAMP analog db-cAMP. nor the cGMP analog db-cGMP induced metamorphosis in the absence of biofilm (Fig. 4). In the presence of a stimulatory biofilm, however, 10~ 4 M forskolin strongly inhibited metamorphosis (Fig. 5A, /-test, forskolin vs. DMSO control, P < 0.001). Inhi- bition did not appear to be the result of toxicity, as larvae exposed to 10~ 4 A/ forskolin in this experiment and in the absence of biofilm swam and behaved normally. We observed no mortality in these trials. Neither db-cAMP nor db-cGMP inhibited responses of larvae to biofilms (Fig. 5B). Compounds affecting the PI/DAG/PKC transduction pathway Exposure of larvae to the phorbol ester TPA caused no significant induction of metamorphosis as compared to the appropriate DMSO controls (Fig. 6). Table II lists the compounds tested, their predicted ef- fect on metamorphosis of larvae of Hydroides elegans if G protein-coupled receptors or the AC/cAMP or PI/DAG/ PKC signal-transduction pathways mediated responses of larvae to metamorphic cues, and the effect we observed in our experiments. Discussion A growing body of research on the effects of neuro- pharmacological agents on marine invertebrate larvae suggests that G protein-coupled receptors and their associ- ated signal-transduction pathways are important regula- tors of larval settlement and metamorphosis, either di- rectly through sensation and processing of exogenous metamorphic cues, or indirectly by modifying responses to these cues. It is not clear, however, whether this role for G proteins in metamorphosis is a characteristic of all marine invertebrate larvae (Hadfield, 1998). Our experi- ments indicate that G protein-coupled receptors or signal transduction by the AC/cAMP or PI/DAG/PKC pathways METAMORPHOSIS OF HYDROIDES ELEGANS 25 A. 100 en "en o "I- o .2 "CD CD CD CL B. en 'en O 9- o -2 "55 CD O CD CL 90 10 Gpp[NH]p Biofilm Filtered Seawater Controls 10' 7 M 10' 6 M 10' 5 M Concentration 100 90 80 70 60 50 40 30 20 10 GDP-p-S + Biofilm Biofilm Filtered Seawater Controls 10' 7 M 10' 6 M 10' 5 M 10' 4 M Concentration Figure 1. Metamorphosis of larvae of Hydroides elegans exposed to activators and inhibitors of G proteins. (A) Response to the G-protein activator Gpp [NH]p. (B) Effect of the G-protein inhibitor GDP-/3-S on metamorphosis in response to bacterial biofilms. Points represent means of 4 replicates; error bars are standard deviations. AC/cAMP or PI/DAG/PKC pathways. Baxter and Morse (1987) found that Gpp[NH]p and GDP-/3-S did not di- rectly affect metamorphosis of larvae of Halintis rufes- cens, but larvae were induced to metamorphose by appli- cation of forskolin, IBMX, and theophylline, suggesting that the metamorphic signal was likely to be transduced by the AC/cAMP pathway, although not by a G protein- coupled receptor (Baxter and Morse, 1987; Morse, 1990). If signal transduction by the AC/cAMP pathway were a necessary event in the metamorphic sequence in Hydro- ides elegans, then competent larvae should have meta- morphosed when exposed to the phosphodiesterase inhib- itors, db-cAMP, and the adenylate cyclase activator for- skolin. The nonspecific phosphodiesterase inhibitors IBMX. theophylline, and papaverine induced significant levels of metamorphosis (Fig. 2); however, we observed no effect of the selective inhibitor RO-20-1724 (Fig. 3). Additionally, neither forskolin nor db-cAMP caused lar- vae to metamorphose (Fig. 4). Forskolin had the opposite A. en 'tn O _c Q. O CD o CD CL 100 -i 90 80 70 60 50 - 40 30 - 20 10 -I IBMX Theophylline Papaverine Biofilm Filtered Seawater Controls 10' 8 M 10' 7 M 10' 6 M 10' 5 M 10 J M 10' 3 M 10' 2 M Concentration do not mediate responses of larvae of the serpulid poly- chaete Hydroides elegans to metamorphic cues. If cues for settlement and metamorphosis of larvae of H. elegans were bound by a G protein-coupled receptor, we would have expected competent larvae to ( 1 ) meta- morphose when exposed to the GTP analog Gpp[NH]p. and (2) fail to respond to inductive bacterial biofilms when those biofilms were presented in combination with GDP-/J-S, a competitive inhibitor of G protein activation by GTP. We observed no increase in metamorphosis over FSW control treatments in the presence of Gpp|NH]p (Fig. 1 A), and no inhibition of response of larvae to bio- filmed surfaces when exposed to GDP-/3-S (Fig. IB). These results imply that the receptors responsible for de- tecting the metamorphic cue in H. elegans are not associ- ated with G proteins. The absence of evidence for the involvement of G pro- tein-coupled receptors does not exclude the possibility that the metamorphic signal is transduced by either the B. en 'en O "- O CD CD CL 100 - 90 - 80 - 70 - 60 50 40 30 20 10 Controls 10' 6 M 10' 5 M 10'" M Concentration 10" J M Figure 2. Metamorphosis of larvae of Hydroides elegans exposed to the phosphodiesterase inhibitors IBMX, theophylline, and papaverine. (A) and (B) present results from two separate trials. Note that the con- centration ranges (x axes) differ between trials. Points represent means of 5 replicates in (A), 4 replicates in (B); error bars are standard devia- tions. * - significantly different from the FSW control treatment at P == 0.05. 26 E. R. HOLM ET AL. 100 -, C/5 90 -- RO-20-1724 'c/5 e DMSO Control O t 80 - ^ Biofilm Q. 70 - v Filtered Seawater O E 60 - -2 50 - lr CD 40 - 30 - CD O 20 - CD CL 10 - - Controls 10' 7 M 10' 6 M 10' 5 M 10' 4 M 10' 3 M Concentration Figure 3. Metamorphosis of larvae of Hydroides elegans exposed to RO-20-1724, a selective inhibitor of the mammalian cAMP-specific phosphodiesterase IV. Exposure of larvae to DMSO served as the control for the RO-20-1724 treatment. Points represent means of 4 replicates; error bars are standard deviations. effect, inhibiting metamorphosis of//, elegans in response to bacterial biofilms (Fig. 5). Alternatively, the metamorphic signal may be trans- duced by cGMP. IBMX. theophylline, and papaverine are not selective for any of the cyclic nucleotide phosphodies- terase isozymes (Thompson, 1993); they inhibit phospho- diesterases that hydrolyze cGMP as well as those that hydrolyze cAMP. Exposure of larvae to db-cGMP. how- ever, neither induced (Fig. 4) nor inhibited (Fig. 5B) meta- morphosis. Although the results presented here appear to be contra- 100 CD C/5 O f 90 - 80 - B db-cAMP db-cGMP *- Q- 70 e DMSO Control O 60 - Forskolin 50 - r Biofilm "CD 40 - v Filtered Seawater - 30 o 20 CD Q_ 10 - - V Controls 1Q- 8 M 10' 7 M 10" 6 M 10' 5 M 10"" M Concentration Figure 4. Metamorphosis of larvae ot Hydroitles elegans exposed to the cyclic nucleotide analogs db-cAMP and db-cGMP, and the adenylate cyclase activator forskolin. Exposure of larvae to DMSO served as the control for the forskolin treatment. Points represent means of 4 repli- cates; error bars are standard deviations. A. 100 n 90 - '8 80 e- o CD 'CD CD Q_ 70 - 60 - 50 - 40 - 30 - 20 - 10 Forskolin + Biofilm DMSO + Biofilm Biofilm Filtered Seawater Controls 10' 8 M 10' 7 M 10' 6 M 10' 5 M 10' 4 M Concentration B. "e- o E ro "CD CD CL 100 -, 90 80 70 - 60 50 40 30 - 20 10 db-cAMP * Biofilm db-cGMP + Biofilm Biofilm Filtered Seawater Controls 10' 8 M 10 7 M 10' 6 M 10' 5 M 10 4 M Concentration Figure 5. Metamorphosis of larvae of Hydroides elegans exposed to bacterial biofilms in combination with forskolin. db-cAMP, and db- cGMP. (A) Response to biofilmed surfaces in the presence of forskolin. The DMSO + Biofilm treatment served as the control for larvae exposed to forskolin. (B) Response to biofilmed surfaces in the presence of db- cAMP and dh-cGMP. Points represent means of 4 replicates; error bars are standard deviations. * - significantly different from the control treat- ment at P s 0.05. dictory, they may be explained by the effects on ion trans- port of some of the compounds used. IBMX (Kopf ct ai, 1983; Simasko and Van, 1993; Usachev and Verkhratsky, 1995), theophylline (Kopf et cil., 1983), and papaverine (Fujioka, 1984; Iguchi et ai, 1992) all have effects on calcium transport that can be either independent of or stronger than their effects on intracellular cAMP levels. Forskolin can block both potassium (Watanabe and Gola, 1987; Coombs and Thompson. 1987; Hoshi ct til., 1988; Garber et at., 1990) and calcium (Park and Kim. 1996) channels, and desensitize acetylcholine receptors (Wag- oner and Pallottu. 1988; White. 1988). independently of its effects on cAMP. Calcium ions are important second messengers in the metamorphic pathway of some hydro- /oan larvae (Freeman and Ridgway, 1987, 1990), and may serve the same function in larvae of polychaetes (Ilan el a!., 1993) and barnacles (Rittschof ct tit., 1986; Clare, METAMORPHOSIS OF HYDROIDES ELEGANS 27 1996). Metamorphosis of larvae of several phyla of ma- rine invertebrates can be induced or inhibited by elevated concentrations of potassium (reviewed in Woollacott and Hadtield, 1996) or by application of potassium-channel blockers (reviewed in Pawlik, 1990). The exact mecha- nism by which potassium affects metamorphosis has yet to be determined. Ongoing research in our laboratory indicates that cal- cium and potassium play important roles in the metamor- phic morphogenesis of H. elegans. Addition of calcium ions to FSW enhances metamorphosis in the presence of biofilms, and pulse application of potassium induces metamorphosis in the absence of other cues (unpubl. data; Carpizo-Ituarte and Hadfield. 1996, 1998). Calcium and potassium channel blockers inhibit metamorphosis in re- sponse to cesium pulses and bacterial films, respectively (unpubl. data: Carpizo-Ituarte and Hadfield, 1998). It seems likely that the results we observed after application of the nonspecific phosphodiesterase inhibitors and for- skolin are due to their effects on ion transport, rather than on cAMP or cGMP levels. This conclusion is supported by the lack of inhibitory or stimulatory effects from db- cAMP, db-cGMP, or the selective phosphodiesterase in- hibitor RO-20-1724. On the basis of this evidence, we conclude that the AC/cAMP pathway does not transduce the metamorphic signal in larvae of H. elegans. Our data also indicate that metamorphosis does not require signal transduction by the PI/DAG/PKC pathway. We attempted to induce metamorphosis with the protein kinase C activator TPA, the phorbol ester found to be the most effective stimulator of metamorphosis in two species of hydrozoans (Miiller, 1985; Freeman and Ridgway, 1990). No effect was observed, implying that the PI/DAG/ PKC pathway does not directly participate in the meta- 100 -i TPA C/5 'c/5 O f 90 - 80 > DMSO Control r Biofilm I Q- 70 v Filtered Seawater O 60 - v CT3 50 CD 40 "c 30 CD O 20 CD Q_ 10 - - I _-_ _&= Controls 10' 9 M 10' 8 M 10 7 M Concentration 10' b M 10' 5 M Induction Induction Induction Induction Induction Induction Induction None Induction None Induction None Induction Inhibition Figure 6. Metamorphosis of larvae of Hvdroidex elegans exposed to the protein kinase C activator TPA. Exposure of larvae to DMSO served as the control for the TPA treatment. Points represent means of 4 replicates; error bars are standard deviations. Table II 1'iti/icled and ohscrvctl effects of compounds tested on lnn'ae of Hydroides elegans Effect on metamorphosis Predicted Observed Compounds affecting G proteins Gpp [NH] p Induction None GDP-^-S Inhibition None Compounds affecting the AC/cAMP signal-transduction pathway IBMX Theophylline Papaverine RO-20-1724 db-cAMP db-cGMP Forskolin Compounds affecting the PI/DAG/PKC signal-transduction pathway TPA Induction None See Table I and the text for descriptions of the activity of each compound. Predicted Effect describes the result one would expect if G proteins, or the affected signal-transduction pathway, mediated re- sponses of larvae to metamorphic cues. morphic process. The possibility that G proteins and the PI/DAG/PKC pathway transduce signals that modify re- sponses of larvae to the metamorphic cue, as in Haliotis rufescens (Trapido-Rosenthal and Morse, 1985; Baxter and Morse, 1987), cannot be eliminated, however. Our results indicate that neither G protein-coupled re- ceptors nor signal transduction by the AC/cAMP or PI/ D AG/PKC pathways are components of the morphogene- tic pathway (in the sense of Baxter and Morse, 1987) directly responsible for sensation and processing of bacte- rial settlement cues in larvae of Hydroides elegans. We are investigating the possibility that the receptors that detect these cues are either ligand-gated ion channels sim- ilar to amino acid taste receptors in channel catfish (Teeter ct a/.. 1990; Brand el a/., 1991), or lectins such as those suggested to mediate responses to bacterial biofilms in larvae of the spirorbid polychaete Janna brasiliensis (Kirchman et /., 1982; Maki and Mitchell, 1985). Acknowledgments This research was funded by grants from the Office of Naval Research to C. M. Smith and MGH (No. N00014- 95-1-0196), and MGH (N00014-95-1-1015). EC-I was supported in part by a scholarship for graduate studies from CONACYT/UABC, Mexico. We thank D. Rittschof and two anonymous reviewers for valuable comments on preliminary drafts of the manuscript. 28 E. R. HOLM ET AL. Literature Cited Baxter. G., and D. E. Morse. 1987. G protein and diacylglycerol regulate metamorphosis of planktonic molluscan larvae. Proc. Natl. Acad. Sci. USA 84: 1867-1870. Baxter, G. T., and D. E. Morse. 1992. Cilia from abalone larvae contain a receptor-dependent G protein transduction system similar to that in mammals. Biol. Bull, 183: 147-154. Biggers. VV. J., and H. Laufer. 1992. Chemical induction of settle- ment and metamorphosis of Capitella capitata Sp. I (Polychaeta) larvae by juvenile hormone-active compounds. Invenebr. Reprod. Dev. 22: 39-46. Brand, J. G., J. H. Teeter, T. Kumazawa, T. Huque, and D. L. Bayley. 1991. Transduction mechanisms for the taste of amino acids. Physio!. Behav. 49: 899-904. Bryan, P. J., P.-Y. Qian, J. L. Kreider, and F.-S. Chia. 1997. Induc- tion of larval settlement and metamorphosis by pharmacological and conspecific associated compounds in the serpulid polychaete Hydroides elegans. Mar. Ecol. Prog. Ser. 146: 81-90. Carpizo-Ituarte, E., and M. G. Hadfield. 1996. Biofilm and ionic stimulation of metamorphosis in the polychaete Hydroides elegans. Am. Zoo/. 36: 10A. Carpizo-Ituarte. E.. and M. G. Hadh'eld. 1998. Stimulation of meta- morphosis in the polychaete Hydroides elegans Haswell (Serpuli- dae). Biol. Bull. 194: 14-24. Carr, W. E. S. 1992. Recurring themes and variations: an overview and introduction. Bint Bull. 183: 143-146. Cheung, W. Y., and Y. M. Lin. 1974. Purification and characteriza- tion of cyclic 3', 5'-nucleotide phosphodiesterase from bovine brain. Methods En-ymol. 38: 223-239. Clare, A. S. 1996. Signal transduction in barnacle settlement: calcium re-visited. Biofmtling 10: 141-159. Clare, A. S., R. F. Thomas, and D. Rittschof. 1995. Evidence for the involvement of cyclic AMP in the pheromonal modulation of barnacle settlement. J. E.\p. Biol. 198: 655-664. Coombs, J., and S. Thompson. 1987. Forskolin's effect on transient K current in nudibranch neurons is not reproduced by cAMP. J. Neurosci. 7: 443-452. Crisp, D. J. 1984. Overview of research on marine invertebrate larvae. 1 940- 1 980. Pp. 1 03 - 1 26 in Marine Biodeterioration: an Interdisci- plinary Study, J. D. Costlow and R. C. Tipper, eds. Naval Institute Press. Annapolis. Maryland. Feger, J., S. Gil-Faloon, and C. Lamaze. 1994. Cell receptors: defi- nition, mechanisms and regulation of receptor-mediated endocytosis. Cell Mot Biol. 40: 1(139-1061. Freeman, G., and E. B. Ridgvtay. 1987. Endogenous photoproteins, calcium channels and calcium transients during metamorphosis in hydrozoans. Roux's Arch. Dev. Biol. 196: 30-50. Freeman, G., and E. B. Ridgway. 1990. Cellular and intracellular pathways mediating the metamorphic stimulus in hydrozoan planu- lae. Roux's Arch. Dev. Biol. 199: 63-79. Fujioka. M. 1984. Lack of a causal relationship between the vasodila- tor effect of papaverine and cyclic AMP production in the dog basilar artery. Br. J. Pharnuicot 83: 113-124. Garber, S. S., T. Hoshi, and R. W. Aldrich. 1990. Interaction of forskolin with voltage-gated K* channels in PC 12 cells. J. Neurosci. 10: 3361-3368. Gilman, A. G. 1987. G proteins: transducers of receptor-generated signals. Anini. Rev. Biochem. 56: 615-649. Hadfield, M. G. 1998. Research on settlement and metamorphosis of marine invertebrate larvae: past, present and future. Biofoitlint; 12: 9-29. Hadfield. M. G., C. C. Unabia, C. M. Smith, and T. M. Michael. 1994. Settlement preferences of (he ubiquitous fouler Hydroides elegans. Pp. 65-74 in Recent developments in Biofoii/ing Control. M. F. Thompson. R. Nagabhushanam, R. Sarojini. and M. Fin- german, eds. Oxford and IBH Pub. Co., New Delhi. Holm. S. 1979. A simple sequentially rejective multiple test proce- dure. Scu/ul. J. Statist. 6: 65-70. Hoshi, T., S. S. Garber. and R. W. Aldrich. 1988. Effect of forskolin on voltage-gated K* channels is independent of adenylate cyclase activation. Science 240: 1652-1655. Iguchi. M., T. Nakajima, T. Hisada, T. Sugimoto, and Y. Kurachi. 1992. On the mechanism of papaverine inhibition of the voltage- dependent Ca + * current in isolated smooth muscle cells from the guinea pig trachea. J. Pharmacot E\p. Ther. 263: 194-200. Ilan, M., R. A. Jensen, and D. E. Morse. 1993. Calcium control of metamorphosis in polychaete larvae. /. Exp. Zoot 267: 423-430. Jensen, R. A., and D. E. Morse. 1990. Chemically induced metamor- phosis of polychaete larvae in both the laboratory and ocean environ- ment. J. Chem. Ecol. 16: 911-930. Kirchman, D., S. Graham, D. Reish. and R. Mitchell. 1982. Lectins may mediate in the settlement and metamorphosis of Jainui fDexi- ospira) brasiliensis Grube (Polychaeta: Spirorbidae). Mar. Biol. Lett. 3: 131-142. Kopf, G. S., C. A. Lewis, and V. D. Vacquier. 1983. Methylxan- thines stimulate calcium transport and inhibit cyclic nucleotide phos- phodiesterases in abalone sperm. Dev. Biol. 99: 115-120. Leitz, T. 1997. Induction of settlement and metamorphosis of cni- darian larvae: signals and signal transduction. Invertehr. Reprod. Dev. 31: KW-122. Leitz, T., and G. Klingmann. 1990. Metamorphosis in Hvdractinia: studies with activators and inhibitors aiming at protein kinase C and potassium channels. Roux's Arch. Dev. Biol. 199: 107-113. Leitz, T., and W. A. Miiller. 1987. Evidence for the involvement of Pi-signaling and diacylglycerol second messengers in the initiation of metamorphosis in the hydroid Hydractinia echinata Fleming. Dev. Biol. 121: 82-89. Leitz, T., and A. Wirth. 1991. Vanadate, known to interfere with signal transduction. induces metamorphosis in Hydractinia (Coelen- terata; Hydrozoa) and causes profound alterations of the larval and postmetamorphic body pattern. Differentiation 47: 1 19-127. Maki, J. S., and R. Mitchell. 1985. Involvement of lectins in the settlement and metamorphosis of marine invertebrate larvae. Bull. Mar. Sci. 37: 675-683. Morse, A. N. C. 1991. How do planktonic larvae know where to set- tle? Am. Sci. 79: 154-167. Morse, D. E. 1990. Recent progress in larval settlement and metamor- phosis: closing the gaps between molecular biology and ecology. Hull Mar. Sci. 46: 465-483. Morse, D. E.. H. Duncan, N. Hooker, A. Baloun, and G. Young. 1980. GABA induces behavioral and developmental metamorpho- sis in planktonic molluscan larvae. Fed. Proc. 39: 3237-3241. Miiller, VV. A. 1985. Tumor-promoting phorbol esters induce meta- morphosis and multiple head formation in the hydroid Hvdractinia. Differentiation 29: 216-222. Park, T.-J., and K.-T. Kim. 1996. Cyclic AMP-independent inhibi- tion of voltage-sensitive calcium channels by forskolin in PC 1 2 cells. J. Neurochem. 66: 83-88. Pawlik, J. R. 1990. Natural and artificial induction of metamorphosis of Phragmatopoma lapidosa californica (Polychaeta: Sabellariidae), with a critical look at the effects of bioactive compounds on marine invertebrate larvae. Bull. Mar. Sci. 46: 512-536. Posternak, T., and G. VVeimann. 1974. The preparation of acylated derivatives of cyclic nucleotides. Methods Enzwno/. 38: 399-409. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43: 223-225. Rittschof, I)., J. Maki, R. Mitchell, and J. D. Costlow. 1986. Ion and METAMORPHOSIS OF HYDROIDES ELEGANS 29 neuropharmacological studies of barnacle settlement. Netherlands J. Sea Res. 20: 269-275. SAS Institute Inc. 1989. SAS/STAT User's Guide. Version f>. 4th <'. North Carolina. 846 pp. Scheltema, R. S. 1974. Biological interactions determining larval set- tlement of marine invertebrates. Thalassia Jugosl. 10: 263-296. Schneider, T., and T. Leitz. 1994. Protein kinase C in hydrozoans: involvement in metamorphosis of Hydractinia and in pattern forma- tion of Hydra. Roux's Arch. Dev. Biol. 203: 422-428. Seanion, K. B., W. Padgett, and J. W. Daly. 1981. Forskolin: unique diterpene activator of adenylate cyclase in membranes and intact cells. Proc. Ntitl. Acad. Sci. USA 78: 3363-3367. Simasku, S. M., and S. Van. 1993. 3-Isobutyl-l-methylxanthine in- hibits sustained calcium current independently of cyclic AMP in neuronal and endocrine cells. Mol. Pharmacol. 44: 622-627. Sokal, R. R., and F. J. Rohlf. 1981. Biometry. W. H. Freeman. San Francisco. 859 pp. Svane, I., and C. M. Young. 1989. The ecology and behaviour of ascidian larvae. Oceanogr. Mar. Biol. Annu. Re\'. 27: 45-90. Teeter, J. H., J. G. Brand, and T. Kumazawa. 1990. A stimulus- activated conductance in isolated taste epithelial membranes. Bio- /)/iv.v. J. 58: 253-259. Thompson, W. J. 1993. Cyclic nucleotide phosphodiesterases: phar- macology, biochemistry and function. Pp. 287-313 in Intracellnlar Messengers. C. W. Taylor, ed. Pergamon Press, Oxford. Trapido-Rosenthal, H. G., and D. E. Morse. 1985. i.-a, u'-diamino acids facilitate GABA induction of larval metamorphosis in a gastro- pod mollusc (Haliotis rufescens). J. Comp. Physio/. B. 155: 403- 414. Trapido-Rosenthal, H. G., and D. K. Morse. 1986. Availability of chemosensory receptors is down-regulated by habituation of larvae to a morphogenetic signal. Pmc. Null. Acad. Sci. USA 83: 7658- 7662. Usachev, Y., and A. Verkhratsky. 1995. IBMX induces calcium release from intracellular stores in rat sensory neurones. Cell Cal- cium 17: 197-206. Wagoner, P. K., and B. S. Pallotta. 1988. Modulation of acetylcho- line receptor desensitization by forskolin is independent of cAMP. Science 240: 1655-1657. Watanahe, K., and M. Gola. 1987. Forskolin interaction with volt- age-dependent K channels in Helix is not mediated by cyclic nucleo- tides. Neurosci. ten. 78: 21 1-216. White, M. M. 1988. Forskolin alters acetylcholine receptor gating by a mechanism independent of adenylate cyclase activation. Mol. Pharmacol. 34: 427-430. Wisely, B. 1958. The development and settling of a serpulid worm. Hvdroides nonegica Gunnerus (Polychaeta). Ansr. J. Mar. Freshwa- ter Res. 9: 351-361. Wodicka, L. M., and D. E. Morse. 1991. cDNA sequences reveal mRNAs for two G signal transducing proteins from larval cilia. Biol. Bull. 180: 318-327. Woollacott, R. M., and M. G. Hadfield. 1996. Induction of metamor- phosis in larvae of a sponge. Invertehr. Kiol. 115: 257-262. Yamamoto, H., A. Tachihana, K. Matsumura, and N. Fusetani. 1995. Protein kinase C (PKC) signal transduction system involved in larval metamorphosis of the barnacle, Balanus amphitrite. Zool. Sci. 12: 391-396. Reference: Biol. Bull 195: 30-42. (August. 1998) Transport Pathways in the Neotropical Sponge Aplysina SALLY P. LEYS 11 AND HENRY M. REISWIG 2 1 Department of Biology, University of Victoria, Victoria. British Columbia, Canada, V8W 3N5; 2 Redpath Museum and Biology Department, McGill University, 859 Sherbrooke St., Montreal, Quebec, Canada: and ' Bellairs Research Institute of McGill University; St. James. Barbados Abstract. Strands of cells distinct from the rest of the tissue were found running lengthwise through the endo- some of four species of the sponge Aplysina. Although the strands were more highly pigmented than the adjacent tissue and could be removed intact with forceps, ultra- structural studies revealed no obvious barrier separating the cells in the strands from the rest of the tissue. The strands consist of stretches of elongate cells tightly aligned along densely bundled collagen fibrils, and areas of other elongate cells that possess numerous filopodia. When sponges were fed fluorescent latex beads in situ. beads were taken up and transported specifically into the strands; eventually they were found at the tip of the sponge and further down the stalk, away from the site of feeding. Beads injected into endosomal strands were also transported upwards in the strands to the tip of the sponge. Video microscopy of cells in strands that had been ex- posed along a portion of their length showed no bulk movement of cells; but individual cells were seen mov- ing in both directions along the strands at 0.025- 0.04 pm-s '. The endosomal cell strands are suggestive of a primitive nutrient transport pathway in sponges. Introduction Sponges were the first multicellular animals to arise from unicellular protists during early metazoan evolution, at least 600 million years ago (Steiner etal., 1993; Reitner and Mehl, 1995), and their tissue organization, accord- ingly, is quite simple. Sponges are diploblastic metazoans that possess no organ systems; their tissues consist of a limited number of cell types that surround and penetrate a mesohyl of collagen. The entire animal is organi/ed Received 30 March 1998; accepted I June 1998. around canals and chambers through which water is chan- neled for feeding and respiration. Within the Porifera. however, some groups differ mark- edly from this fundamental plan. The Hexactinellida, for example, are mostly syncytial (Reiswig, 1979; Mackie and Singla. 1983), a condition that allows symplastic transport of food (Perez, 1996; Wyeth et a/.. 1996) and rapid conduction of action potentials that control the feed- ing current (Leys and Mackie, 1997). The Cladorhizidae, on the other hand, are deep-sea demosponges that are effectively carnivorous and lack flagellated chambers, which are otherwise a key poriferan character (Vacelet and Boury-Esnault, 1995). In a group as old as the Pori- fera, other exceptions to the basic sponge plan probably exist. One intriguing possibility lies within demosponges of the genus Aplvsimi. These sponges, which are known as bacteriosponges for the vast number of bacteria living within their mesohyl, belong to the Verongida, an order that contains numerous pharmaceutically interesting chemicals (Bergquist el ai, 1991; Ciminiello et a/., 1994, 1997). An unusual feature of some sponges in this genus is the presence of longitudinal strands of elongate cells in the endosome. The strands, up to lOO/um in diameter and particularly densely populated by bacteria, constitute a quasi-tissue that runs vertically through sponges of tube, chimney, and rope-form alike. Despite their prominence in these sponges and their marked absence in other members of the phylum studied to date, the strands have received only passing mention in the literature. Tsurumi and Reiswig (1997) suggested that the "endosomal" tissue strands in Aplysinu cauli- I'onnis (a rope-form sponge) serve as migration routes for rapid transport of cells for tip growth. Similar elongate cells, in the endosome and around the water canals in Verongia (synonymous with Aplysina). were suggested 30 SPONGE TRANSPORT PATHWAYS 31 to he contractile cells that act together to control water flow through the canal system (Vacelet, 1966). To have tested either of these hypotheses would have required cell labeling studies and, if possible, video-enhanced-contrast microscopy of live tissue, techniques that were not readily available until recently. We have now undertaken both approaches and present here a detailed description of the endosomal tissue strands and their role in transport of materials. Materials and Methods Description of the sponges and collection site Aplysina caulifonnis (probably synonymous to A. fiilva). a neotropical rope-form sponge, grows at the tip (Wulff, 1990), reaching about 1 cm in diameter, but up to several meters in length. It is pink or purple on the outside and yellow inside. At depths of 7- 10 m, A. cauli- fonnis is typically branched and grows only 20-30 cm in height. Aplvsiiw lacunosa is a mustard-yellow tube- form sponge. The tubes are 10-50 cm in length and can occur singly, but more often several of them bud from the same base. The tubes are 5-10 cm in diameter with an atrial diameter of 2-5 cm. Aplysina rigida, a branching stick-form sponge, is pink outside and yellow on the in- side, but is both thicker (up to 5 cm in diameter) and taller (up to 1 m in height in 7-m-deep waters) than A. caulifonnis. Aplysina fistularis is a robust, branching, stick-form sponge in Barbados, but it may be tube-form in other habitats in the Caribbean. A. fistularis is yellow outside and has an irregular outer surface, but it lacks the lacunae characterizing A. lacunosa', it is 3-5 cm in diameter and reaches 20-30 cm in height at depths of 7- 10m. Species of Aplysina were manipulated in situ and col- lected by scuba from coral rubble at a depth of 7- 10 m off the Bellairs Research Institute, St. James, Barbados (1310'N, 5940'W). Field experiments Latex bead feeding. Plastic bags (4x7 cm) open at both ends were slipped over the branches of A. caulifonnis in situ and, about 5- 10 cm from the top, were gently secured at both ends with plastic twist ties (Figs. 1 and 2a). A plastic syringe and needle were used to inject the bag with 5 ml of 1-^rn fluorescent latex beads (Molecular Probes, Eugene, OR) diluted 1 :5 with seawater. The beads were not treated with serum to prevent clumping, because trial experiments showed that clumping did not interfere with bead uptake. The bead-filled bags were left on the sponges for 4 h and then removed; the beads had been cleared from the bags during this period. Two branches were collected for fixation at 4 and 24 h, and 2, 3, 7, 1 1, and 1 3 days after feeding. The sponges were cut off at a bag with beads Tip Middle 5cm Lower i/WVM/ Figure 1. Diagram of a branch of Aplysinu caulifonnis showing the attachment of a plastic bag that is used to feed latex beads to the sponge. Also shown are the three segments (Tip. Middle, and Lower) that the branch was cut into; the segments were used in determining the fate of the beads. point 15 cm from the top, placed in a plastic bag, and taken to the laboratory, where each piece was cut into three segments: the top 5 cm (Tip), the middle 5 cm where the bag had been secured (Middle), and the bottom 5 cm (Lower) (Fig. 1). The segments were fixed in 2% para- formaldehyde in seawater and placed in the refrigerator at about 4C. After 4 hours the fixative was poured off and replaced with fresh, cold 2% paraformaldehyde. Insertion of latex beads. Whole branches of A. cauli- fonnis attached to loosely buried coral rubble were col- lected in plastic bags (with the coral) and brought into the laboratory; the sponges were never removed from seawater. The sponge branches were cut 7 cm from the tip, but not quite in two; thus a hinge of tissue still con- nected the tip segment to the main stem. The sponge was placed in a large bucket of seawater and, with the aid of a dissecting microscope poised over the bucket, l-um latex beads were injected with a plastic syringe and needle onto a severed cell strand and its immediate area. The wound was sewn together with dental floss and the sponges were returned to the field no more than 4 h after collection and reattached with plastic cable ties to other pieces of coral rubble. Preliminary experiments showed that, despite wave action, the wound healed and new tis- sue grew over the dental floss within 1 -2 days. Sponges treated in this way were recollected 1, 2, 3, 7. 11, and 13 days after wounding, cut into three segments as described above, and fixed in 2% paraformaldehyde in seawater. Laboratory analysis of field experiments Sponge pieces were transported while still in fixative to the University of Victoria. British Columbia, Canada, where 32 S. P. LEYS AND H. M. REISWIG they were rinsed in phosphate-buffered saline (PBS) for 4 h, infiltrated overnight in 1:1 PBS and Tissue Tek (O.C.T. Compound 4583, Sakura Finetechnical, Tokyo), and embed- ded in 100% Tissue Tek by freezing in liquid nitrogen. Blocks of embedded tissue were stored at 20C. Longitudi- nal sections (30 /um) from each segment (Tip, Middle, and Lower) of one of the two branches from each collection period were cut on a cryomicrotome. Sections were mounted on slides coated with poly-L-lysine and stored at 20C. Sections were viewed with a Leitz Aristoplan epifluores- cence microscope equipped with the F1TC filter (450-490- nm excitation) through which the tissue appeared green and orange and the latex beads bright yellow. Latex beads were counted in 10 sections chosen at random from each segment of the sponge for every time interval of collection. The location of the beads was scored as follows: i) in the tissue (i.e., in or near the flagellated chambers) of the sponge; ii) at the edges of water canals; and iii) in cell strands. Where beads were too numerous to count (for example in the pieces collected within the first 24 h after treatment), 100 beads were first counted, and estimates of every hundred beads visible thereafter were made. Selected sections were later stained in hematoxylin and eosin but not cleared in xylene, and then mounted in PBS-glycerol. This procedure allowed us to see the location of beads within the strands. Video microscopy Pieces of A. cunlifonnis were transferred to flow- through seawater tanks at the Bellairs Research Institute without removal from seawater. A 4-cm piece of a sponge branch was cut in two lengthwise with a shaip disposable scalpel. Where a strand was exposed without damage along a portion of its length, the piece of branch was tied with dental floss to a glass microscope slide with the cut surface against the slide. With the exposed strand facing upwards, the slide was rested on supports at either end in a How-through pertusion chamber so that the sponge did not touch the bottom of the dish. The pertusion cham- ber was set on the stage of a Zeiss compound microscope and the strands were filmed with a lOx objective lens for up to 5 h using a Panasonic CCD color video camera; the image was digitally enhanced with an Omnex real-time digital image processor (Imagen Corp.) and recorded on Fuji video tapes in real time with a Panasonic VCR. Pho- tographs were taken from the monitor of a Technitron television with a Nikon FG camera on TMAX 400 ASA Him. Bright field microscopy Pieces of A. cauliformis were cut in half lengthwise to expose the tissue strands and placed strand-side up, in a 5-cm diameter dish of seawater. on the stage of a Leitz Orthoplan microscope. The exposed surface was illumi- nated with two adjustable microscope lamps, and photo- graphs were taken on 100 ASA Fujicolor print film. Electron iiiu'roscop\ Tissue strands were dissected, with a scalpel and fine forceps, out of freshly collected sponge branches. Strands and whole pieces of sponge with exposed strands were fixed in a cocktail of cold 1% osmium tetroxide and 2% glutaraldehyde in 0.45 M sodium acetate buffer, pH 6.4, and 10% sucrose in the final volume. Pieces were kept in the refrigerator at 4C for 4 h, and the fixative was then changed. After 1 6 to 20 h, the fixative was poured off, and the tissue was rinsed in 0.45-/um Millipore-filtered seawater and stored in 70% ethanol for transport to the University of Victoria, British Columbia. Tissue strands and pieces of sponge were then further dehydrated to 100% ethanol, stained en Hoc in 0.5% uranyl acetate overnight at the 80% ethanol stage, rinsed three times in propylene oxide, infiltrated overnight in 1:1 Epon (Taab 812) and propylene oxide, and embedded in 100% Epon at 60C. Silver sections were cut on a Reichert ultramicrotome with a diamond knife, stained with lead citrate, and viewed in a Hitachi XOOO transmission electron microscope. H};iire 2. (a) Aplysina cauliformis attached to coral rubble, with plastic bags (arrowhead) tilled wilh latex beads attached to each branch. Latex beads were "fed" to the sponge branches in the plastic bags, and the tale of ingested materials was determined, (b-e) Live tissue strands of A. cauliformis. Bar: 1 mm. (f-h) Latex beads in longitudinal sections of preserved sponge branches. Bar: 50 //m. (b) A strand in a longitudinal section of a sponge. The strand (arrow) is darker than the adjacent tissue and divides to go around the spongin skeleton. The arrowhead indicates a piece of the broken spongin skeleton, (c) Two exposed strands (arrows) in a longitudinal section are similar in color to the tissue at the outer layer of the sponge, (d) A cross section of a sponge branch showing three strands (arrows), which are easily identified by their pink color and by their slight separation from the neighboring tissue. Polychaete worms (*) are an abundant macrosymbiont in these sponges and either use water canals or fashion their own pathways through the sponge, (e) A tissue strand can be lifted up with a dissecting needle, indicating the integrity and slight elasticity of the strands. ( f) Four hours after feeding, latex beads (arrowheads) were well distributed through- out the tissue but were not in the strands (s). (g) Two days after feeding, the beads (arrowheads) were predominantly in the tissues around the water canals (*): strands (s). (h) Eleven days after feeding, the beads (arrowheads) were only found in the tissue strands (s) of the tip segment of the branches. Very few were found in other tissues of the sponge at this time. SPONGE TRANSPORT PATHWAYS 33 Results Description of the strands The tissue of A. caitliformis contained longitudinal strands running from the base to the tip of the sponge (Fig. 2b). The strands were darker than the adjacent tissue. and were flecked here and there by white cells that ap- peared to be highly retractile. The strands were most similar in color to the tissue at the cortex (Fig. 2c) and the tissue at the tip of the sponge. Up to 12 strands could be identified in any one cross section of a sponge branch (Fig. 2d). Strands ranged in diameter from 20 to 100 pm. 34 S. P. LEYS AND H. M. REISW1G hut the diameter of any one strand was rather uniform for its entire length. Strands could be traced for at least 3 cm and often for more than 6 cm along a sponge branch. The strands tended to be straight, with only slight bends and turns where they coursed around the spongin skeleton (e.g.. Fig. 2b). and there were no cross-connections be- tween strands. Typically, strands branched or forked at the tip of the sponge, and if a strand was located along the cortex or edge of the sponge, a branch was often directed towards the cortex. The strands were sufficiently separate from the rest of the tissue in the sponge that they could be lifted by one end with forceps and gently pulled free from the sponge. The strands were slightly elastic, as can be seen by the ability of a dissecting needle to pull a portion of a strand away from the sponge tissue (Fig. 2e). However, more than a moderate pulling resulted in the breakage and slow curling of the excised ends. Occasionally, when they were pulled out of the sponge, the strands frayed at the edges and pieces at the edges were left behind. Other parts of the sponge lacked this integrity and could not be removed intact from the sponge. Examination of the live tissues of A. lacunosa, A. fistu- Uiris, and A. rigidu revealed tissue strands in all three species. Strands were always distinctly more pink than the surrounding tissue, up to 100/ym in diameter, and traceable for at least 3 cm. and often for the full length of the sponge. Sections 1-cm thick cut with a scalpel from top to bottom of these sponges revealed fewer strands at the base than at the middle or tip of the sponge, and none of the strands were especially close to the cortex. In one section of A. lacnnosu. 24 different strands were identi- fied. Most of the strands could be followed for 4 cm or more along the length of the sponge. Electron microscopy Ultrastructural examination revealed no clear barrier separating the tissue strands from the rest of the sponge tissue. Only in one instance was a string of connected cells 60-//m long found at the edge of the strand. More commonly the edges of strands were lined simply by collagen. Strands were identifiable in sections of the sponge tissue by the presence of an increased number of aligned elongate cells, vast numbers of bacteria, and a much denser collagenous mesohyl in which the collagen tibrils were often aligned in a single direction (Fig. 3A. B). Elongate cells (40-50-yum long and 2.5 I-/JITI wide) with numerous filopodia were separated laterally by about 10 fjm (Fig. 3A). These cells typically lacked conspicuous nucleoli and contained many small vesicles. Other cells in the strands (possibly spherulous cells) were stretched around large vesicles containing an electron-dense mate- rial (Fig. 3A). Along particularly dense bands of aligned collagen there were highly elongate cells with a well- developed cytoskeleton (Figs. 3B, 4A). Cells in the strand were often found in the process of phagocytosing bacteria (e.g.. Fig. 4B). The tissue outside of the strands consisted mostly of a comparatively loose collagenous mesohyl containing numerous bacteria, although fewer than in the strands; scattered amoeboid-like cells: and flagellated chambers. The sponge cortex consisted of many spherulous cells containing large inclusions, and some elongate cells em- bedded in collagen that was not aligned in any particular direction (Fig. 4C). Strands from other species of Aplysina were identical in ultrastructure to those in A. ccudiformis. Cells in strands that were preserved after they had been forced to recoil at the edges by cutting and tweezing appeared no different than those in strands that had been fixed while still within the sponge. In both cases, cells were elongate when lying beside aligned collagen fibrils and irregularly ovoid when not. Feeding and insertion of lute.x heads In the feeding experiment, the beads were first seen in the flagellated chambers and pinacoderm (Fig. 5a) and later in cells in the mesohyl (Figs. 5B and 20. Four hours after feeding, beads were found in all the sponge tissues of the middle segment except the cell strands (Fig. 2f). By 2 to 3 days after feeding the beads were concentrated in the tissues around the water canals (Fig. 2g). At this time only a few beads were found in the cell strands. However. 7 to 11 days after feeding, many beads were found in the tip segment, where they were specifically lodged in the cell strands (Fig. 2h); very few were found in adjacent tissues. Closer examination of selected stained sections showed that the beads were within the cells in these strands. At this level of resolution it appeared that the beads were in the elongate cells rather than in the spherulous cells, but the strands with beads were not ex- amined by electron microscopy to confirm this observa- tion. The relative proportion of beads in the tissue, at the edges of canals, and in the tissue strands over the course of 13 days after latex bead feeding and insertion is shown in Figures 6 and 7. In the feeding experiment (Figs. 6A and 7A) most latex beads in the middle segment were first found in the sponge tissues (triangles) and later at the edges of water canals (squares) during the first week after feeding. The number of beads in the middle segment declined during the second week after feeding. Although many beads were counted in the strands in the middle segment at day 2 (see Table I), these were far outnum- bered by those beads that were counted around the edges of the canals and in the tissue at this time. In the second week, 227c of the beads counted in all segments were in the strands in the tip segment 1 1 days after feeding (Fig. SPONGE TRANSPORT PATHWAYS 35 Figure 3. Longitudinal sections of tissue strands from ,4/>/vwmi um/i/c/n (transmission electron microscopy). (A) A section showing several elongate cells with numerous filopodia (arrows). These cells lie lengthwise throughout the strand within a bacteria-filled (b) collagenous mesohyl (co). Other cells in the strands contain large inclusions of electron-dense material (*). Bar: 10 pm. (B) Some cells in the tissue strands were highly elongate and were tightly aligned within dense bundles of collagen fibrils (arrows). Bar: 10 /jm. 6A: tip segment). However, 7 to 11 days after feeding, 95% of all the beads counted in the tip segment alone were in the strands (Fig. 7 A). Many beads were found around the water canals in the lower segment in the feed- ing experiment. A week after feeding, 63% of the beads counted in the lower segment were found in the strands (Fig. 6A: lower segment). Latex beads that were inserted into the sponge branches about 7 cm from the tip were transported predominantly upward to the tip 2 weeks after insertion (Fig. 6B: tip segment); very few beads reached the segment of the sponge below the insertion point. One day after their insertion, the beads were abundant in the general tissue of the sponge in the middle segment, much as was found in the feeding experiment. After 3 days a large proportion of the beads were found at the edges of the water canals, and a few (5%) were in the tissue strands in the middle segment. Whereas the number of beads counted in the 36 S. P. LEYS AND H. M. REISWIG ^. Figure -I. Aspects of the cell strands and the cortex of Aplyxinu cinilifonnix (transmission electron microscopy). (A) A higher magnification of cells among densely bundled collagen fibrils (co) in strands, showing bundles of electron-dense fibrils, which may be microtilaments ( arrowheads!, and of microtubules (arrows), in the cells. The niicrotubules were identified as such from their diameter in high magnification electron micrographs. Bar: 2 /jm. (B) An example of a cell within a strand in the process of engulfing a bacterium (b) Bar: 1 pm. (C) The cortex of the sponge. Just inside the dermal layer or pinacoderm (p) are numerous spherulous cells (sp) with large inclusions, and individual elongate cells (arrowheads) lying in a collagenous mesohyl (co). These elongate cells are far shorter than those in the strands and arc not in large tracts of aligned collagen fibrils. Bar: ? /jm. middle segments declined toward the end of the second week, their number in the tip segment increased at this time. After 7 days. 42% of the beads counted in the tip segment were in the strands: 13 days after feeding, how- ever, most beads in the tip segment were in the tissue (Fig. 7B). SPONGE TRANSPORT PATHWAYS 37 Figure 5. Latex bead uptake by ,4/j/v.w'm; fuiiliforniis (transmission electron microscopy). (A) Four hours after feeding, latex beads (Ib) were found in flagellated chambers (fc) and (B) in ameboid cells in the mesohyl. Bar (A): 5 pm: (B): 2 pm. Attempts to record movement of cells in the tissue strands of Aplysina were hampered by the opacity of the tissue strands; indeed, only cells at the edges of strands could be monitored by transmitted light microscopy. Con- sequently, strands that were exposed by cutting the sponge branch longitudinally were illuminated from above, and viewed at a low magnification with a 10X objective lens. The image was magnified by both the video camera and the digital image processor. With this system, highly refractile cells, which may correspond to the spherulous cells shown in Figure 3A, were the most conspicuous element in the strands; the movements of the smaller cells could not be monitored. The refractile cells showed no obvious movement as a group in either direc- tion along the strand. Nevertheless, more than 2 h after filming began on the sponge, several individual cells definitely moved along the strand in both directions at rates of 0.025 ^rn-s" 1 to0.04^m-s ' (/; = 8) (Fig. 8). Thus, the movement was probably not caused by contrac- tion of the strand after wounding. Discussion Endosomal tissue strands have not been described in sponges other than the genus Aplysina. nor is there any record that a specialized structure is involved in the distri- bution of particulates in sponges. We have shown here, however, that when latex beads are fed to A, cauliformis, they are taken up and transported into the tissue strands; eventually they end up at the tip of the sponge or further down the stalk. Sponges are well-known to have highly motile cells, whose rate of movement (2 to 21 mm -day"'; Bond. 1992) although slower than that of crawling by Amoeba is comparable to. and even faster than crawling by fibroblasts and neutrophils (Bray. 1992). Furthermore, at the growing edge of sponges, these cells (particularly those of a similar type) often become aligned in tracts (Bond and Harris, 1988). Similar tracts of cells, some- times referred to as cell "cords," have been described in various demosponges where their suggested role has been in growth (Brien, 1976), regeneration (Levi, 1960), and remodeling (Diaz, 1979). In each of these cases, the pri- mary role of the cell cords was thought to be in skeleto- genesis (Simpson, 1984), as demonstrated by Teragawa ( 1986) for the keratose sponge Dysidea etheria. However, none of these tracts or cords of cells is so permanent a structure within the sponge, nor so widespread and mor- phologically uniform within a genus, as the endosomal tissue strands in Aplvsina. Nor can any of those tracts be so readily extracted from the rest of the tissue as can Aplvsina' s tissue strands. The bead uptake experiments here show that the strands are involved in transport of materials taken in during feeding. Most of the beads that were fed or inserted into the sponges were excreted within a week; relatively few ended up in the strands, which suggests that most of the food taken in by flagellated chambers and the adjacent pinacoderm is probably distributed to cells locally, and wastes are probably expelled from the same area. How- ever, a role for the strands as transport pathways is sup- ported by the observation that a substantial proportion of beads were in the strands in the tip segment 2 weeks after 38 S. P. LEYS AND H. M. REISWIG 100 Tip 80 * Tissue 60 o Canal edge 40 * Strand 20 X^ -^- c9. . -S - -(. - - - I 2 3 7 11 Time after feeding (Days) 100 80 60 40 20 Tip 1 2 3 7 11 Time after bead insertion (Days) ro c 4hrs 0) a. 100 80 60 40 20 'U 8 0) CL 100 80 60 40 20 100 80 60 40 20 Middle Figure 6. Latex head experiments. Percent of the total number of beads counted in all sections of all three segments (Tip. Middle, and Lower) at each time interval, in the tissue (triangles), at the canal edges (squares), and in the strands (crosses): (A) feeding; (B) insertion. See text for explanation. the middle portion of the sponge was ted. The beads found in the strands 2 days after the sponges were fed were far outnumbered by the beads counted around the B T3 0) ro 0) .o O t: o I 100 80 60 40 20 100 80 60 40 20 *- Tissue o Canal edge "Strand 1 2 3 7 11 13 Time after feeding (Days) 1 2 3 7 11 13 Time after bead insertion (Days) Figure 7. Latex bead experiments. Percent of all the beads counted in the tip segments only: (A) feeding: (B) insertion. See text for explanation. canal edges or in the tissue (see Table I). The delayed appearance of the beads in the tip segment 7 days after feeding is best explained by the time required for the cells transporting the beads to crawl first to the strands and then along the strands to the tip, where they accumulated. That downward transport also occurs is suggested by the number of beads found in strands in the lower segments a week after feeding, and later around the water canals. The beads found around the canal edges in the lower segment may have been transported down and then moved to the canals for excretion, or they may have been rein- gested after excretion from the middle section. Aplysiiui fiilvu has been shown to grow rapidly, at an average rate of 2 cm month ' (minimum 0.2 cm month"', maximum 15 cm month '; calculated from Wulff, 1990). For this rate of growth, rapid translocation of materials to the tip would be necessary despite the ability of flagellated chambers at the tip to take in and distribute nutrients locally. Downward transport might supply regions of the sponge that possibly feed less and function primarily as a stalk. Video microscopy showed that individual cells moved within strands for distances of 30 pm or more at a rate of 2.2 to 3.5 mm day '. At this rate of movement, material could be transported 5 cm in 2 weeks, if trans- ported more or less in a straight line. The large numbers of bacteria in the strand might support the energy require- SPONGE TRANSPORT PATHWAYS 39 Table I The number of latex beads counted in the tissues, around the water canals, and in the strands of each segment of fed branches of Aplysina cauliformis Number of beads Collection period Segmenf 1 In each segment region h Total per segment Total per day (all segments) Percent of beads in each segment/day Tissue Canal edge Strand day 1 , 4 h Middle 20.000 9,400 7 29.407 29.407 day 1, 24 h Lower 572 3,765 20 4,357 9.792 44.5 Middle 3,348 1,060 6 4,414 45.1 Tip 876 144 1 1 .02 1 10.4 day 2 Lower 5,116 19.400 65 24,581 106.472 23.1 Middle 23.400 49.950 548 73.898 69.4 Tip 1.554 6.366 73 7.993 7.5 day 3 Lower 2.890 5.219 48 8.157 1 1 .536 70.7 Middle 319 3.042 (1 3.361 29.1 Tip 14 3 1 18 0.2 day 7 Lower 78 355 766 1.199 18.554 6.5 Middle 3,806 12.753 346 16,905 91.1 Tip 12 7 431 450 2.4 day 1 1 Lower 1,090 12,779 290 14,159 30.698 46.1 Middle 2,843 3,914 2.212 8.969 29.2 Tip 535 37 6,998 7.570 24.7 day 13 Lower 1,148 3,617 39 4.804 6.394 75.1 Middle 247 1,274 30 1 .55 1 24.3 Tip 5 5 29 39 0.6 J Lower. Middle, and Tip segments of each sponge branch fed latex beads. h Beads were counted in 10 sections from each segment of sponge (Lower. Middle. Tip) each day. and were noted as being in three regions of the segment: Tissue (flagellated chambers and associated tissues); Canal edge (cells specifically around the edges of water canals); and Strand (only within cell strands). ments of so many moving cells, and this notion is substan- tiated by the ample phagocytosis of bacteria seen in cells in the tissue strands. The presence of tissue strands in tube- and stick-forms of Aplysina suggests that the mechanism of transport is efficient for growth regardless of sponge form, and indeed has been maintained by at least four species in this genus. Whether more distantly related verongiid sponges also possess tissue strands or similar structures for nutrient transport or other functions is not known. Feeding in sponges has been well-documented, and with the exception of the two examples cited in the intro- duction, the Cladorhizidae and the Hexactinellida, particle uptake in sponges occurs at the choanocytes in the flagel- lated chambers or at the pinacoderm-lined incurrent ca- nals. Food is transferred via food vacuoles from the choa- nocytes to amebocytes. which deliver the nutrients locally to other cells, and wastes are excreted via the excurrent canals (Kilian, 1952; Weissenfels, 1976; Willenz, 1980; Imsieke, 1994). Directional translocation of cells and nu- trients occurs in gemmule formation (Rasmont and de Vos, 1974) and during oogenesis (Fell, 1983), but this transport occurs over very short distances. Hexactinellid sponges possess perhaps the longest transport pathway known in the Porifera, but translocation is intrasyncytial and occurs through a highly dynamic three-dimensional network of reticulated tissue, rather than along fixed path- ways (Wyeth et al.. 1996; Leys, 1998). Although sponges have not previously been shown to possess specific transport pathways for nutrients, all in- vertebrates have developed means of distributing nutri- ents. Many higher invertebrates such as ascidians. most echinoderms. and crustaceans have a well-developed fluid circulatory system with numerous types of hemocytes which, in addition to their many other functions, transport nutrients. Others invertebrates, however, move food around in a much slower manner: branching corals trans- locate nutrients to the tip of branches for tip growth (Buchsbaum Pearse and Muscatine, 1971); gorgonians have cells that travel through the stem canals and solenia. a collagenous filled mesohyl, to deliver nutrients to the other tissues of the animal (Murdoch, 1978); fixed t 40 S. P. LEYS AND H. M. REISWIG Figure 8. Cell movement in tissue strands (video microscopy). A highly retractile cell (arrowheads) was traced as it moved for 3(1 ^m along a tissue strand at 0.033 /ynvs' 1 . Time of frames: (A) (I min: (B) 5 min; (C) 15 min. Bar: 100 ^m. chymal cells in turbellarians are thought to act as a sort of stationary intraccllular circulatory system (Pedersen, 1961); and acid phosphatase staining has shown that amehocytes in the hemal lacuna of crinoids (again, a col- lagen-filled pathway) are involved in digestion and trans- port of nutrients (Hein/eller and Welsch, 1997). We cannot rule out a role for the tissue strands in contracting the ostia and water canals to control water flow as suggested for the elongate cells in Verongici ( Va- celet, 1966): nonetheless, there is little evidence to sup- port this hypothesis. Although some cells in the strands have what appears to be a highly developed actin cy- toskeleton, and the strands recoiled gently after being cut, none of the strands were wrapped around the water canals or ostia sufficiently to be able to reduce water flow if contracted. Furthermore, localized patches of elongate cells were present in valve- or sphincter-like structures around the water canal system, and these quite possibly function in controlling water flow through the sponge as suggested by Vacelet (1966). and by Reiswig (1971) for those sponges in which flow is not controlled by flagellar activity. The facts that video microscopy did not reveal all cells in the strands to be moving, and that not all cells in the strands are elongate, could also be taken as evidence against the transport hypothesis. But the thickness ( 100 jjm} of the strands allowed only a few cells at the edges of strands to be observed at high magnification by video microscopy. Despite the low magnification and use of epi-illumination. a few of the highly refractile cells could be clearly seen to travel for significant distances along the length of the strand. It could also be true, however, that not all the cells in the strand transport material. Thin sections of the strands showed that only some cells in the strands had a highly developed actin cytoskeleton and were aligned along densely bundled collagen fibrils. Other cells were clearly elongate with filopodia at either end, but were not associ- ated with the densely bundled collagen fibrils. Fibroblasts in culture orient themselves along grooves in culture dishes (Carter. 1967; Clark ct uL, 1980). and on collagen substrates they take on a bipolar spindle morphology with filopodia at either end (Elsdale and Bard. 1972). Further- more, Harris ( 1987) has demonstrated that the forces ex- erted by fibroblasts in culture are far greater than is neces- sary for cell crawling, and that the prime function of these cells is to bundle and align collagen to prepare the path for migratory cells (Stopak and Harris, 1982). An analogous situation could exist in Aplysina strands. We suggest that the role of the highly elongate cells with a well-developed actin cytoskeleton is primarily to bundle and align the collagen fibrils, creating a pathway that cells transporting nutrients can recognize and follow, thereby allowing the rapid transport of materials either to the tip or to the base of the sponge (Fig. 9). In this fashion, apoplastic nutrient transport by ameboid cells has become specialized in the genus Aplysina to the extent that these sponges have a differentiated structure that parallels fluid transport systems in other animals, but is a typically novel poriferan solution. Acknowledgments We thank the director and staff at the Bellairs Research Institute (BRI) of McGill University, in Barbados, for the use of facilities while we were conducting the field work SPONGE TRANSPORT PATHWAYS 41 Figure 9. Diagram of the proposed route taken by latex beads (Ib) that were fed to Aplysina caii/iformis. Beads are taken in via the incurrent canals (ic) to the flagellated chambers (fc), where they are ingested (1) by choanocytes. The beads are then transferred to wandering amoebocytic cells (2), which encounter strands of elongate cells (ec) and highly bundled collagen fibrils (co). The amoebocytic cells follow the direction of the collagen, thereby transporting the latex beads up and down the strand (3) as shown by the arrows. for this study. We also thank L. Verhegge. M. Tsurumi, and G. O. Mackie for comments on the manuscript. This research was supported by a Commander C. Bellairs Post- doctoral Fellowship from BRI and McGill University to SPL, and by a research grant from the Natural Science and Research Council of Canada (OGPOO 1427) to G.O.M. Literature Cited Bergquist, P. R., P. Karuso, R. C. Cambie, and D. J. Smith. 1991. Sterol composition and classification of the Porifera. Biocliem. Sysl. Ei-1000 Da. At the end of dialysis, concentrations of total Ca and free Ca were measured on samples withdrawn from inside the dialysis bags (dialyz- able + nondialyzable Ca) and from the external solution (dialyzable Ca). Nondialyzable Ca concentration was cal- culated by subtraction. Calcium speciation modeling Results from dialysis experiments were compared to theoretical estimates of Ca speciation by using the equilib- rium speciation model MINTEQA2/PRODEFA2 (Allison et al.. 1991). Values for the molar concentration of the major inorganic ions (Na + , K + , Ca 2+ , Mg 2+ , Cl. SO 4 2 , and total CO : ) in blood and EPF of M. mercenaria were obtained from Crenshaw ( 1972a). Since blood SO 4 : ~ and CO 2 concentrations were not available, the corresponding EPF values (Crenshaw, 1972a) were substituted as an approximation. The total CO 2 molarity provided by Cren- shaw (1972a) was entered in the model as CO 3 2 . MINTEQA2 was run using various permutations of the given concentrations to understand more clearly changes in Ca speciation with changing blood and EPF composi- tion. This included adding a hypothetical Ca-binding pro- tein to examine organic Ca speciation. The MINTEQA2 model provided percentage distributions of free, inor- ganic, and organic Ca species formed under the given conditions. Electrophoresis Plasma and EPF were subjected to discontinuous SDS- polyacrylamide gel electrophoresis (SDS-PAGE) using the method of Laemmli (1970), with a 3% stacking and 7.5% resolving gel. Protein concentrations in the samples [plasma = 1.34 0.15 ing/ml, n = 5; EPF = 0.97 0.17 mg/ml, n = 5; determined by Bradford's (1976) dye binding method] were adjusted by dilution such that the amount of protein applied per lane was 40-60 ^g. Stack- ing (100 V d.c.; 2 h) and resolving (200 V d.c.; 4 h) were carried out at 15C. Gels were stained either with 0.1% Coomassie brilliant blue R-250 (Hames and Rickwood, 1981) or with a periodic acid-silver stain for glycopro- teins, as described by Dubray and Bezard (1982). Protein molecular weights were determined by comparison with a standard mixture containing myosin (205 kDa). /3-galac- tosidase (116 kDa), phosphorylase B (97.4 kDa), BSA (66 kDa), ovalbumin (45 kDa), and carbonic anhydrase (29 kDa). Results Quahogs injected with 45 Ca :+ and 'H-tyrosine showed a progressive decrease in radioactive counts in the right shell EPF, the site of injection (Fig. la, b). Concentrations of both materials decreased 3 orders of magnitude over a period of 4 to 4.5 h. Samples of blood from the anterior and posterior adductor muscles and of EPF from the left shell exhibited a corresponding increase in '"Ca 2 ^ and 'H- tyrosine counts. Counts from the four sampling points converged at approximately 3 h, indicating complete mix- ing of the introduced radiolabel with the blood. The high molecular weight BSA, on the other hand, showed less tendency to redistribute into the circulatory system (Fig. Ic). Although a decline in BSA concentration was ob- served in the right EPF, a difference of 3 orders of magni- tude remained between counts at the point of injection and at the three other sampling sites at 4 h post-injection. 46 P. S. NAIR AND W. E. ROBINSON l&'+Vt = 1000000 | \ a re O .A 100000 i ~~ ' .--_ Tt a 10000 i 1000 ; ^ 100 i 1 n ( 112345 1UUUUU : >, b 0) 10000 i \ '3> ^"-~-. , 1000 i ' ~"T\^1^ =r-5 ^~ ro 100 i 10 i ? 1 ( 112345 1UUOUU 5 ^ C < 10000 '~' s m 1000 a 100 10 ._... A. I2H" y "v time (h) Figure 1. Radioactivity (cpm) in Mercenaria mercenaria extrapal- lial fluid (EPF) and blood following injection of (a) 45 Ca (MW = 40), (b) 3 H-tyrosine (MW = 181), and (c) I4 C-BSA (MW = 66 kDa) into the right shell EPF. Extrapallial fluid samples were drawn from the central zone of the right ( O ) and left ( ) shells; blood samples were drawn from the anterior (. . .A. . .) and posterior (----) adductor muscles. Lines were drawn by a logarithmic-fit function using SlideWrite Plus* V4.0. Furthermore, I4 C-BSA concentration in the left shell EPF appeared to remain constant over the 4-h period. The time required for the introduced BSA to become homogeneous with the blood was estimated to be at least 22 h. A confirmatory experiment using nonlabeled BSA was performed to determine whether the BSA remained solu- bilized in EPF or became adsorbed to the mantle tissue during the course of the experiment. SDS-polyacrylamide gels of EPF and blood plasma showed that the 66-kDa band of BSA was present only in the right shell EPF (the point of introduction) in samples taken 3 h and 6 h post- injection (results not shown). Staining intensities of the bands at 3 and 6 h were similar, indicating that the BSA was not bound to tissues. If BSA was present in the left shell EPF and in the plasma samples at 3 and 6 h. it was below the detection limits of SDS-PAGE. The shapes of the semilog plots in Figure 1 indicate a rapid decrease in injected counts during the first 30 min, followed by a longer phase of gradual decline and attain- ment of an apparent steady state over the next 4-4.5 h. This suggests that the release of radiolabel from the EPF may be biphasic. Apparent depuration constants (kjs) were calculated from the slopes of tangents to the initial and final portions of the plots. Depuration constants for the rapid phase (k d ,) were estimated as 9.7, 12.0, and 2.4 h~' for 45 Ca, 3 H-tyrosine. and I4 C-BSA. respectively, whereas constants for the slow phase (kj : ) were 0.1, 0.08, and 0.04 h"'. Biological half-lives calculated from the kj : values were 6.9 h for 45 Ca, 8.7 h for 'H-tyrosine. and 17.3 h for I4 C-BSA. The mean concentration of total Ca measured in the blood plasma of 10 quahogs was 10.9 1.3 mM. of which 0.4 0.1 mM was free Ca :+ . Bound Ca was estimated as 10.5 1.2 mM. Equilibrium dialysis of the plasma sam- ples revealed that the majority of bound Ca (7.2 mM; Table I) was nondialyzable (i.e., bound to ligands >1000 Da), but 1.1 mM was dialyzable. and 0.2 mM was free Ca. These concentrations reflect a 78% recovery of total Ca following dialysis; the remainder was presumably bound to the walls of the dialysis bags. Expressed as percentage of total Ca recovered, bound Ca (nondialyz- able and dialyzable) and free Ca :+ amounted to 84.7, 12.9, and 2.4%, respectively (Table I). Inorganic Ca speciation obtained by MINTEQA2 pre- dicted that a high proportion of plasma Ca should be present as free Ca 2+ (78.9%; Table II, permutation 1), and the remainder as bound Ca (primarily CaSO 4 , 20.2%). However, this prediction is inconsistent with the values of 2.4% free Ca 2+ and 12.9% bound-dialyzable Ca ob- tained by the dialysis experiments (Table I). If the SO 4 : concentration used in model permutation 1 (46.1 mM, Crenshaw, 1972a) was replaced by the seawater SO 4 : concentration (28.2 mM, Mantoura et 50 kDa) that were not detected in the plasma, indicating that the outer mantle epithelium contains an effective barrier to at least some high molecular weight proteins. These ob- servations are consistent with those of Bielefeld et al. (1993), who found the outer mantle epithelium of the Table II Theoretical speciation of calcium in Mercenaria mercenaria blood plasma and extrapallial fluid Model permutation* Sample CaSO 4 CaHCO, CaOIT Ca-Protein 1 Plasma 78.9 20.2 <0.9 trace EPF 85.9 19.3 <0.9 trace 2 Plasma 85.9 13.1 <0.9 trace EPF 86.5 12.5 <0.9 trace 3 Plasma 2.3 1.0 96.7 EPF 2.2 0.4 96.9 Note: Values expressed as percentage of total calcium. Results were obtained using the equilibrium speciation model MINTEQA2 (Allison et al.. 1991). * Premutations: (1) concentrations of inorganic ions taken from Crenshaw (1972a): (2) sulfate concentration decreased from 46.1 to 28.2 mM', (3) 60 mM of hypothetical Ca-binding protein (15:1 Caiprotein) with log,,, K a of 2.1 AT' added. 48 P. S. NAIR AND W. E. ROBINSON 100 100 mM protein Figure 2. Percentage distribution of free Ca (ionic Ca :+ ) in the blood plasma of Mercenaria mercenaria in the presence of a hypotheti- cal Ca-binding protein with 15 Ca-binding sites. The lines represent a range of binding affinity constants tested (log,,, K a = 2-6 A/~'). Data were generated using the equilibrium speciation model MINTEQA2 (Allison et a/.. 1991). snail Biomphalaria gluhrutii impermeable to injected horseradish peroxidase (MW = 40 kDa). An important factor governing the direction of Ca flux across the outer mantle epithelium is anaerobiosis. An- aerobiosis induced upon shell closure is accompanied by redissolution of CaCO, from the shell to maintain a more uniform plasma pH through the bicarbonate buffering sys- tem (Crenshaw and Neff, 1969; Hudson, 1992; Littlewood and Young, 1994). This leads to a net Ca flux away from the central zone EPF. On the other hand, injury to the shell causes remobilization of Ca into the EPF. Calcium for shell repair is derived predominantly from the calcium cells of the mantle and foot (see Watabe, 1983), by disso- lution of CaCOj and Ca,(PO 4 ) : spherules. Because the quahogs in our experiments had damaged shells and were kept out of water for as long as 4.5 h, the 3-h timeframe for Ca exchange in this study probably reflected opposing Ca fluxes resulting from anaerobiosis and attempts at shell repair. Other studies have found similar timeframes for Ca exchange in bivalves. 4S Ca introduced in the seawater took 2 h to reach a steady state with the mantle in Argopecten irmdians (Wheeler el inica. Mm: Biol. 114: 423-428. Dubray, G., and G. Bezard. 1982. A highly sensitive periodic acid- silver slam for 1 ,2-diol groups of glycoproteins and polysaccharides in polyacrylamide gels. Anal. Blot-hem. 199: 325-329. Falini, G., S. Albeck, S. Weiner, and L. Addadi. 1996. Control of aragonite orcalcite polymorphism by mollusc shell macromolecules. Science 271: 67-69. Feher, J. J., C. S. Fullmer, and R. H. Wasserman. 1992. Role of facilitated diffusion of calcium by culhindin in intestinal calcium absorption. Am. ./. Physioi. 262: C517-C526. Graf, F., and J. C. Meyran. 1983. Prcmolt calcium secretion in mid- gut posterior ceaca of the crustacean Orchestin: ultrastructure of the epithelium. J. Morphol. 177: 1-23. Greenaway, P. 1971. Calcium regulation in the freshwater mollusc Limmieii staginilis (L.) (Gastropoda: Pulmonata). II. Calcium move- ments between internal calcium compartments. J. Exp. Biol. 54: 609-620. Hames, B. I)., and I). Rickwood. 1981. Gel Electrophoresis of Pro- teins: A Pniclictil Approach. IRL Press, Oxford. 290 pp. Hudson, R. I,. 1992. Ion transport by the isolated mantle epithelium ol i he freshwater clam. Unio coiiiplimtiis. Am. ./. Physioi 263: R76- R83. Jones, R. G., and W. I,. Davis. 1982. Calcium-containing lysosomes in the outer mantle epithelial cells of Ainbleimi. a fresh-water mol- lusc. Anal. Rec. 203: 337-343. Karbach, U. 1992. Paracellular calcium transport across the small intestine. J. Nun: 122: 672-677. Kawaguchi, T., and N. Watabe. 1993. The organic matrices of the shell of the American oyster Crassostrea vi>\t>inicti Gmelin. J. Exp. Mar. Biol. Ecol. 170: 11-28. Keith, J., S. Stockwell, D. Ball, K. Remillard, D. Kaplan, T. Thann- hauser, and R. Sherwood. 1993. Comparative analysis of macro- molecules in mollusc shells. Comp. Biochem. Phvsiol. 105B: 487- 496. Kessler, G., and M. Wolfman. 1964. An automated procedure for the simultaneous determination of calcium and phosphorous. Clin. Chem. 10: 686-687. Laemmli, U. K. 1970. Cleavage of structural proteins during the as- sembly of the head of bacteriophage T4. Nature 227: 680-685. IJttlewood, D. T. J., and R. E. Young. 1994. The effect of air-gaping behaviour on extrapallial fluid pH in the tropical oyster Crassostrea rhizophorae. Comp. Biochem. Physioi. 107A: 1-6. Lucu, C. 1994. Calcium transport across isolated gill epithelium of Ciin-iiuis. J. Exp. Zoo/. 268: 339-346. Mantoura, R. F. C., A. Dickson, and J. P. Riley. 1978. The com- plexation ot metals with humic materials in natural waters. Estuarine Coastal Mar. Sci. 6: 387-408. Marsh, M. E., and R. L. Sass. 1983. Calcium-binding phosphopro- tem particles in the extrapallial fluid and innermost lamella of clams. 7. Exp. Zoo/. 226: 193-203. May, P. M., P. W. Linder, and D. R. Williams. 1977. Computer simulation of metal-ion equilibria in biofluids: models for the low- molecular-weight complex distribution of calcium (II), magnesium (II), manganese (II), iron (II), copper (II), zinc (II), and lead (II) ions in human blood plasma. ./. Chem. Soc: Da/ton Trans. 6: 588- 595. Meenakshi, V. R., P. L. Blackwelder, P. E. Hare, K. M. Wilbur, and N. Watabe. 1975. Studies on shell regeneration I. Matrix and mineral composition ot the normal and regenerated shell of Pomacea pahtdosa. Comp. Biochem. Phvsiol. SOA: 347-351. Misogianes, M. J., and N. D. Chasteen. 1979. Extrapallial fluid: a chemical and spectral characterization of the extrapallial fluid of Mytilus eilulis. Anal. Biochem. 100: 324-334. Miyamoto, H., T. Miyashita, M. Okushima, S. Nakano, T. Morita, and A. Matsushiro. 1996. A carbonic anhydrase from the nacre- ous layer in oyster pearls. Proc. Natl. Acad. Sci. USA 93: 9657- 9660. Neff, J. M. 1972. Ultrastructure of the outer epithelium of the mantle in the clam Mercenaria mercenaria in relation to calcification of the shell. Tissue Cell 4: 591-600. Putnam, F. W. 1975. The Plasma Proteins: Structure. Function, and Genetic Control. Vol. 2. Academic Press, New York. 423 pp. Robinson, W. E., and D. K. Ryan. 1988. Transport of cadmium and other metals in the blood of the bivalve mollusc Mercenaria merce- naria. Mar. Binl. 97: 101 - 109. Saha, A., T. K. Jana, and A. Choudhury. 1988. The extrapallial fluid of Macoma hirminica: an environment for calcium carbonate deposition. Philipp. ./. Sci. 117: 395-399. Saleuddin, A. S. M., and S. C. Kunigelis. 1984. Neuroendocrine con- trol mechanisms in shell formation. Am. Zoo/. 24: 91 1-916. Samata, T. 1990. Ca-binding glycoproteins in molluscan shells with different types of ultrastructure. Veliger 33: 190-201. Scott, B. J., and A. R. Bradwell. 1983. Identification of the serum binding proteins for iron, zinc, cadmium, nickel and calcium. Clin. Chem. 29: 629-633. Sorenson, A. K., D. S. Wood, and L. B. Kirschner. 1980. Electro- physiological properties of resting secretory membranes of lamelli- branch mantles. ./. Gen. I'hvsiol. 75: 21-37. Timmermans, L. P. 1973. Mantle activity following shell injury to the pond snail, Lymnaea stagnulis L. Malaco/ogia 14: 53-61. CALCIUM IN QUAHOG BLOOD AND EPF 51 Tripp, M. R., L. A. Bisignani, and M. T. Kenny. 1966. Oyster amoebocytes in vitro. J. Invert. Path. 8: 137-140. Wada, K., and T. Fujinuki. 1976. Biomineralization in bivalve mol- luscs with emphasis on the chemical composition of the extrapallial fluid. Pp. 175-190 in The Mechanisms of Mineralization in the Invertebrates and Plants. N. Watabe and K. M. Wilbur, eds. Univer- sity of South Carolina Press, Columbia. Watabe, N. 1983. Shell repair. Pp. 289-316 in The Mollusca. vol. 4, A. S. M. Saleuddin and K. M. Wilbur, eds. Academic Press, New York. Weiner, S. 1984. Organization of organic matrix components in min- eralized tissues. Am. Zoo/. 24: 945-951. Wheeler, A. P., and E. W. Harrison. 1982. Subcellular localization and characterization of HCOr-ATPase from the mantle of the fresh- water clam. Anodonta cataracta. Comp. Biochem. Physio/. 71B: 629-636. Wheeler, A. P., and C. S. Sikes. 1984. Regulation of carbonate calci- fication by organic matrix. Am. Zoo/. 24: 933-944. Wheeler, A. P., P. L. Blackwelder, and K. M. Wilbur. 1975. Shell growth in the scallop Argopecteii irnulians. I: Isotope incorporation with reference to diurnal growth. Biol. Hi///. 148: 472-482. Wheeler, A. P., K. W. Rusenko, D. M. Swift, and C. S. Sikes. 1988. Regulation of in vitm and in vivo CaCO 1 crystallization by fractions of oyster shell organic matrix. Mar. Biol. 98: 71-80. Wilbur, K. M. 1972. Shell formation in molluscs. Pp. 103-145 in Chemical Zoologv, vol. VII, M. Florkin and B. T. Scheer, eds. Aca- demic Press. New York. Wilbur, K. M., and A. M. Bernhardt. 1984. Effects of amino acids, magnesium, and molluscan extrapallial fluid on crystallization of calcium carbonate: i;i vitro experiments. Biol. Bull. 166: 251-259. Wilbur, K. M., and A. S. M. Saleuddin. 1983. Shell formation. Pp. 235-287 in The Molliixca. vol. 4, A. S. M. Saleuddin and K. M. Wilbur, eds. Academic Press, New York. Ziegler, A. 1996. Ultrastructural evidence for transepithelial calcium transport in the anterior sternal epithelium of the terrestrial isopod Porcellio scaber (Crustacea) during the formation and resorption of CaCO, deposits. Cell Tissue Res. 284: 459-466. Reference: Biol. Bull. 195: 52-59. (August. 1998) Induction of Extra Claws on the Chelipeds of a Crayfish, Procambarm clarkii ISAMU NAKATANI 1 *. YOSHINORI OKADA 2 . AND TAKUJI KITAHARA 1 1 Department of Biology. Faculty of Science. Yamagata University. Yainagata 990-8560, Japan; ~ Department of Biology, Faculty of Science. Okayama University, Okayama 700-8530, Japan Abstract. In a crayfish, Procambarus clarkii, growth of an extra claw was induced by making a V-shaped wound in the proximal end of the propodus of the second and third chelipeds. Two nerve bundles were damaged by the wounding. Some type of extra growth developed on the propodi of 13 of the damaged chelipeds: a pair of extra claws (2 chelipeds). a pair of extra dactyls (1 cheliped), a single extra dactyl (2 chelipeds), and a single slight projection (8 chelipeds). The extra claws and dactyls de- veloped from the peripheral side of the propodus away from the wound site. One of a pair of extra dactyls and the single extra dactyls could be moved only slightly, either manually or by the crayfish. The other extra dactyls could be moved by the crayfish. Muscles were associated with each of the extra and primary claws. The muscles attached to the double extra claw or dactyl were inner- vated by nerve bundles that were branches from the pri- mary thick nerve bundles. One possible explanation for these findings is that the severed nerve fibers in the thick nerve bundles regenerate, elongate into aberrant roots, and form extra claws or dactyls. Introduction The occurrence of double claws has been observed in the past. They have been reported to occur naturally on the cheliped of the American lobster, Homants ameri- canus, from the primary propodus near the dactyl (Faxon, 1881 ) and from the base of the primary propodus (Cole, 1910). In the crayfish. Procciinhunts clarkii, extra claws have been reported to occur naturally on the first and third chelipeds (Nakatani et a!., 1997). Both of these extra Received 1 August 1997; accepted 7 May 1998. * To whom correspondence should be addressed. E-mail: nakatani@ sci.kj.yamagata-u.ac.jp claws developed from the base of the primary propodus. These crayfish were not able to move the dactyls of the extra claws, although they could be moved manually. Nerve bundles were observed to branch from the primary thick nerve bundles into the extra claw on the third cheli- ped (Nakatani et ai, 1997). Lateral outgrowths on the first cheliped of crayfish can be induced by wounding. In one study (Murayama et al., 1994), outgrowths developed in about \0 c / f of wounded chelipeds. This percentage was increased to 61.9% by increasing the width and depth of the wound and remov- ing the tissue in the wounded area (Nakatani. 1996). Many of these outgrowths have serrations and hooks. However, there is no articulation between the primary propodus and the new structure. These structures, hereafter called "lateral outgrowths." may be distinguished from an extra claw or extra dactyl by the absence of an articulation. The objectives of the present study were to induce the growth of extra claws and to investigate the degree of movement and innervation of the extra growth. Materials and Methods The experimental crayfish Crayfish (Procambarus clarkii) of both sexes were col- lected, without regard to molting stage, from ponds in the suburbs of Yamagata City, Yamagata, Japan. The length of the carapace varied from 15.2 to 36.6 mm. We used both intact and eyestalk-ectomized specimens because eyestalk removal promotes growth and decreases molt interval (Fingerman and Fingerman, 1974; Nakatani and Otsu, 1979; 1981; Suzuki, 1980). Wounds The dorsal side of the base of the propodus of the second and third chelipeds was wounded using small scis- 52 EXTRA CLAW ON CHELIPED OF CRAYFISH 53 sors. The V-shaped incision was made to a depth of about half the width of the base of the propodus (Fig. 1 ). The ventro-median tissues were destroyed by inserting a nee- dle ( 1 mm in diameter) into the wound. In 34 crayfish, a total of 134 chelipeds were wounded. Both eyestalks of five of the crayfish were removed from the base 10 to 20 days post-operation. Rearing wounded crayfish From September 1996 to March 1997, the wounded crayfish were reared until they had molted two or three times. The animals were kept separately in individual containers (300 X 240 X 105 mm) under a natural photo- period at room temperature (18-26C). Prawn pellets (Super B; Nihon Nosan Industry, Japan) and dry persim- mon leaves were placed in the rearing containers so that the crayfish could feed on them ad libitum. Sensory neuron responses A cheliped with extra dactyls was cut off at the base of the merus and held on a glass slide by rubber bands. The exoskeleton at the base of the merus was removed and the nerve bundles were exposed. To detect responses from the sensory neurons, the point of a needle was used to push the extra dactyls in a closing direction. Action potentials from the exposed nerve bundles were recorded extracellularly, using suction electrodes. Mechanical force exerted by closure of dactyls Measurements were taken on the force exerted by clos- ing the dactyls. The cheliped with double extra dactyls was cut off at the base of the merus and prepared as described above. Nerve bundles were stimulated using a pair of tungsten hook electrodes. Repetitive pulses (single pulse, 10 ms in duration; 20 Hz) were applied for 0.5-2 s to hook electrodes connected to an electronic stimulator (SEN1 101; Nihon Kohden, Japan). The mechanical force exerted by closure of each of the two dactyls was recorded V-shaped wound Dactyl Carpus Propodus Pollex Figure 1. Diagram of the second cheliped of the crayfish showing V-shaped wound at the base of the propodus. simultaneously with a thermal array recorder (TRA1 100; Nihon Kohden) via a high-gain force-displacement trans- ducer (SB-1T-H; Nihon Kohden) coupled mechanically to the tip of the dactyl by a silk thread filament. The sensory nerve responses and mechanical force were recorded at room temperature (23-28C). Obsen>ations on nen'e bundles and muscles Observations were made on the nerve bundles and mus- cles of the propodus. The exoskeletons of the propodus and carpus of the above-mentioned chelipeds were re- moved, and nerve bundles and muscles were exposed and placed in physiological solution for crustaceans (van Harreveld, 1936). The nerve bundles were stained with a 0.005% solution of methylene blue dissolved in physio- logical solution and were observed under a dissecting microscope. The stained chelipeds were then fixed over- night at 0C in a 20% formalin solution acidified to pH 3.8 with an acetate buffer. The muscles were observed with a dissecting microscope and a polarizing light micro- scope (BHSP. Olympus, Japan). Results Development of extra structures Data on extra structure development are shown in Table I. Most of the chelipeds healed normally, but 13 (9.7%) formed extra structures. A pair of claws, a pair of dactyls, and single dactyls developed from the side away from the wound, and single, slight projections developed at the wound. These projections ranged in size from 0.3 to 1.8 mm at the second or third post-operation molt. Be- cause extra structures developed at such a low frequency, we cannot draw any conclusions about the effect of eye- stalk removal on their development. Morphology of extra claws A pair of extra claws that developed on one of the damaged chelipeds is shown in Figure 2. Here, the dactyl and pollex were torn off at the first post-operation molt. However, a pair of projections 0.6-mm long developed on the remaining untorn part (arrowheads in Fig. 2A). Two extra claws appeared on the propodus near the pri- mary dactyl, and the primary dactyl and pollex regener- ated at the second molt. The lengths of the three dactyls (Fig. 2B, a, b, c) and the single contralateral dactyl were 0.5. 1.9, 1.7, and 3.3 mm, respectively. The three claws elongated at the third molt, and the shape of each claw was normal. There was no boundary among the propodi of these claws. The lengths of these three dactyls (Fig. 2C, a, b, c) and the single contralateral dactyl were 2.9, 3.7, 3.7. and 4.4 mm. respectively. 54 I. NAKATANI ET AL Table I Summary of wounding experiment in the crayfish Procambarus clarkii (values represent number of chelipeds) Developed extra structures 1 Double Single Wounded Autotomized Claws Dactyls Dactyl Projection Healed normally 4 Eyestalks With 1 14 10 1118 93 Without 20 1 1010 17 1 A total of 134 chelipeds from 34 crayfish received a V-shaped incision. After 10 to 20 days, both eyestalks were removed from 5 of the animals (20 chelipeds). : Some animals autotomized chelipeds post-operatively. 3 Extra structures developed in 13 chelipeds. 4 Most of the chelipeds healed normally, with no outgrowths. Figure 2. Development of extra claws on the third chehped of the crayfish after a wound was made in the propodus. Both eyestalks were removed at the base on day 1 1 after wounding. (A) Two projections 0.6-mm long (arrowheads) developed on the propodus, but the dactyl and pollex were torn off at the first post-operation molt. (B) The projec- tions developed into two poor claws at the second molt, and the untoni remaining part regenerated the dactyl and pollex. (C) The claws devel- oped fully at the third molt. A and B. the castoff exuviae after the second and third molts, respectively, a, primary claw; b and c, extra claws. Bar = 5.0 mm. Morphology of a pair of extra dactyls Two dactyls appeared on one of the chelipeds of a crayfish with eyestalks. The V-shaped wound seemed to have healed normally at the first molt. However, extra dactyls appeared on the right second cheliped at the sec- ond molt (Fig. 3, a, b). The extra dactyls developed on the primary propodus near the primary dactyl, although the wound was made at the base of the propodus. The primary claw was angled about 45 to the left of the propodus near the primary dactyl. The angle of the extra dactyls and the primary claw was about 90. The extra dactyls a and b of Figure 3 are hereafter called "extra dactyl a" and "extra dactyl b," respectively. The lengths of the carapace, extra dactyls a and b, and primary and contralateral dactyls were 39.5, 2.9, 3.1, 5.1, and 3.1 mm, respectively. Morphology of single extra dactyl A single extra dactyl that developed on the propodus of the second right cheliped at the second molt is shown in Figure 4. The extra dactyl was located between the base of the primary propodus and the base of the primary dactyl. The extra and primary dactyls were 2.0- and 3.7- mm long, respectively. Movement of the extra and primary dactyls In the example shown in Figure 2, the crayfish could move the primary dactyl but not the two extra dactyls at the second molt. Extra claw b was usually in an open position, but opened readily when it was closed manually. After the third molt, the crayfish could move the extra and primary dactyls. It always moved these three dactyls asynchronously and often spontaneously moved the pri- mary dactyl synchronously with extra dactyl b. These EXTRA CLAW ON CHELIPED OF CRAYFISH 55 Figure 3. Two extra dactyls on the right second cheliped produced after the propodus was wounded. (A) Extra dactyls in the open state. (B) Extra dactyls in the closed state, a and b, extra dactyls; p, primary dactyl. Bar = 5.0 mm. three dactyls would move synchronously when one of them was touched. In the diagram shown in Figure 3, the crayfish could move the extra dactyls and the primary dactyl. When it was picked up. it moved the extra dactyl a more frequently than the primary dactyl. However, it moved these dactyls simultaneously if one of them was touched. The single extra dactyl in Figure 4 and extra dactyl b in Figure 3 could be moved only slightly, either manually or by the crayfish. Muscles and innen'iition Muscles were observed to correspond to each claw in the propodus. In view of their locations, these muscles could control the opening and closing of the three claws. The closer muscles are shown in Figure 5A. Four muscles in the propodus with two extra dactyls (Fig. 3) were dissected out as shown in Figure 6B. In this figure, muscles m 1 , m2, m3, and m4 are the closer muscle of the primary dactyl, the closer muscles of extra dactyls b and a, and the opener muscle of extra dactyl a, respec- tively. The mean ( standard error of the mean) lengths of 10 sarcomeres of each of these muscles after being fixed with formalin solution were 6.4 0.3, 6.2 0.3, 7.7 0.4, and 5.8 0.5 pm. There were two thick nerve bundles in the intact propo- dus of the second cheliped (Fig. 6A): one ran toward its dactyl along the boundary between the opener and closer muscles; the other ran to its pollex along the closer mus- cle. The extra dactyls shown in Figure 3 were innervated with nerve bundles from the primary nerve trunks. A thick nerve bundle ran toward the primary dactyl between the two closer muscles (Fig. 6B, m2 and m3) of the extra dactyls. At the proximal part of the propodus. the nerve bundle ran between the closer muscle of the primary dac- tyl (ml ) and the opener muscle of the extra dactyl a (m4) (Fig. 6B. C). In a propodus with two extra claws, the nerve bundles branched to each claw from the primary nerve trunk (Fig. 5B). Extracellular recording An extracellular recording was made from the nerve bundle at the merus of the cheliped with two extra dactyls. The neuron in the nerve bundle responded to the push toward closing of extra dactyl b (Fig. 7). However, the neuron did not generate action potentials when the dactyl moved in the opposite direction. Therefore this response might be the action potential from proprioceptors. Mechanical force generated by dactyl movement The mechanical force generated by the movement of the primary and extra dactyls in response to electrical Figure 4. A single extra dactyl that developed on the right second cheliped. Both eyestalks were removed on day 12 after the cheliped was damaged. The arrowhead shows an articulation. Bar = 5.0 mm. 56 I. NAKATANI ET AL stimulation of the nerve bundle in the merus was mea- sured. The force generated by the primary dactyl was about 8- and 3-fold greater than that generated by extra dactyls a and b, respectively (Fig. 8). The primary dactyl moved spontaneously and asynchronously with the extra dactyls just before the electrical stimuli were applied to the nerve bundles (Fig. 8A. A'). Discussion There have been previous reports of double extra claws developing from the carpus (Bateson, 1894) and coxa (Przibram, 1921) of the crayfish (Astacns fluvkitilis) and from the propodus of the cheliped of lobster (Homarus americanux) (Cole, 1910). Recently, two extra claws that Figure 5. Innervation and muscles in the propodus of a cheliped with two induced extra claws. The cheliped is the same as that shown in Figure 2C. (A) Muscles corresponding to each claw (dorsal view). ( B ) The nerve trunks in the primary propodus branch, and each branched nerve bundle innervates a claw (ventral view), a. primary claw; h and c. extra claws; ml, m2, and m3, closer muscles associated with (he primary claw (a) and extra claws (h. ci. respectively. Arrowhead shows (he nerve trunks. Bar = 1.0 mm. "-^^^^P" '^||- ; Figure 6. Innervation and muscles in a propodus with double extra dactyl and in a normal cheliped. The cheliped with extra dactyls is the same as that shown in Figure 3. (A) Two nerve bundles (arrowheads) in the propodus of the second cheliped. (B and C) Branched nerve bundles from the primary nerve trunks (arrowheads) innervate the extra dactvls. ml. closer muscle of the primary claw; m2 and mX closer muscles of the extra dactyls b and a, respectively; m4. opener muscle of the extra dactyl a. a, b. and p are extra dactyl a, extra dactyl b. and primary dactyl, respectively. Bar = 1.0 mm. developed naturally in P. chirkii were described (Nakatani ct al., 1997). These single claws were borne on the first and third chelipeds one on each propodus. The extra EXTRA CLAW ON CHELIPED OF CRAYFISH 57 Tjp. 0.5s Figure 7. Extracellular recording from nerve trunks in the merus of a cheliped with extra dactyls. Thick bar signifies manual closing of the extra dactyl shown in Figure 3. claws induced in the present study are essentially similar to those naturally occurring ones. The naturally occurring extra claws reported by Cole (1910) and Nakatani el at. ( 1997) developed from the base of the primary propodus. Lateral outgrowths from experimental incisions devel- oped from the wounded site on the outer surface of the propodus (Murayama el /., 1994; Nakatani, 1996). Thus, in the present study, the wound was made at the base of the propodus. However, all of the extra claws or extra dactyls developed on the distal side of the wound. This suggests that the extra claws reported by Cole (1910) and Nakatani el <;/. ( 1997) were caused by damage at a site more proximal to the extra claw for example, at the carpus. Although there were no extra propodi, the shape and location of the a pair of extra dactyls observed in this study were similar to those of the naturally occurring one reported by Faxon (1881) on the cheliped of H. ameri- canus. As in the present study, these extra claws devel- oped away from the base of the propodus. The propodi of the extra claws described by Faxon were about one- sixth as long as the extra dactyls, and each was separated from its primary propodus by an articulation. Lateral outgrowths on chelipeds have been assumed to result from the abnormal healing of a natural wound (Su- zuki and Odawara, 1971; Shelton el ai, 1981: Okamoto, 1991; Nakatani el ai, 1992). Many of those outgrowths consisted of a pair of projections without any articulation. The proximal and distal surfaces of the wound each devel- oped one projection, and they faced each other as mirror images (Nakatani, 1996). However, in the present study, the extra dactyls developed away from the damaged site. Different mechanisms may be involved in the develop- ment of a lateral outgrowth and of an extra claw or dactyl. In the propodus it was observed that one nerve trunk runs in the direction of the pollex and another to the dactyl. Primary dactyl Extra dactyl a Stimulation A' Primary dactyl Extra dactyl a Stimulation 1 s B 0.5g Primary dactyl 0.1g Extra dactyl b Stimulation Extra 0.5g dact V' b Extra 0.1g dactyl a Stimulation 1 s 0.5g 1 s MMilMI^ k^^Mk ,^iMkjtJ Uk^^jU^ "' ^WW* ^H^pj ^^R^PP 0.5g 0.1g 1 s Figure 8. Movements of the primary and extra dactyls. The movements generated by each of two of three dactyls were recorded simultaneously when repetitive pulses (single pulse, II) ms in duration; 20 Hz) were applied to the nerve trunks in the merus. which was monitored in each bottom trace. The same cheliped shown in Figure 3 was used. A', the time-scale for enclosed part with broken line in A was magnified 4-fold. 58 I. NAKATANI ET AL. Lateral outgrowths were induced by wound damage of the nerve trunk that runs toward the pollex (Nakatani. 1996). Meanwhile, in the present study, both nerve trunks may have been damaged by wounding, because of the depth of the wound and the location of the nerve trunks. The crayfish reported by Nakatani et til. (1997) could not move the dactyl of its natural extra claws, despite the presence of muscles and nerve bundles. In contrast, all of the extra dactyls in the present study could be moved by the crayfish. Further studies are needed to explain this difference, but some speculation is possible. Closer and opener muscles, and innervation of these muscles, are necessary for a crayfish to move its dactyls. The crayfish claw is moved by two muscles controlled by five efferent axons: one inhibitor and one excitor to the opener muscle, and one inhibitor and two excitors a "fast" and a "slow" to the closer muscle (van Harreveld and Wiersma, 1937: Wiens, 1976). Both the primary and extra claws can be functional if the axons are adequately branching and innervate the muscles of each claw. The primary and extra dactyls may move synchronously if the above-mentioned axons branch to each muscle. However, the crayfish observed in this study could move all three dactyls synchronously or asynehronously. Atwood ( 1973) reported that a wide variety of synaptic mechanisms and muscle fiber properties permit delicate control within a simple framework in the crayfish claw. The articulation of the dactyl also affects its mobility. An extra dactyl b could be moved only slightly by hand, even though it had well-developed closer and opener muscles, and innerva- tion (Fig. 6B, C). For an extra claw or dactyl to develop, some of the severed nerve fibers must separate from the primary nerve trunks. The present results suggest that severed nerve fibers regenerate; some of them elongate to form aberrant roots and extra claws or extra dactyls. Aberrant roots have been observed in the abdominal nerve cord of the crayfish P. simiilans by Bittner et al. (1974). They reported that neurons projected aberrant roots toward the periphery after the third abdominal ganglion was removed. Also, Goransson et al. ( 1988) showed that in crayfish P. clarkii. the regenerating neurons would orient themselves in the proper direction to make connections toward the preferred target area. Further, the neurons do not respond to posi- tional cues during the first week of regeneration. Other studies have shown that transplanted tissues can induce growth of new structures. Kao and Chang (1996) proved that dactyl, pollex, and ischium tissues of the crab claw all had claw-transforming activity if they were auto- transplanted into the autotomi/,ed stump of the fourth walking leg. Furthermore, by autotransplantation of a pol- lex into the eye sockets, a claw with complete proximal segments (ischium, merus, carpus and manus) developed (Kao and Chang, 1997). In the cockroach Blattella ger- nia/iica. graft/host junctions of the leg regenerated seg- mented structures consisting of two copies of all struc- tures distal to the point of the junction (French. 1976). French (1976) speculated that the graft and host do not heal together and interact, but rather regenerate autono- mously in mirror-image symmetry of the original graft and host levels. Further studies are needed to explore related issues in crayfish. Acknowledgments We are grateful to Dr. Toshiyuki Nishida, Professor Emeritus, University of Hawaii, for his valuable com- ments and for improving the manuscript. Literature Cited Atwood, H. L. 1973. An attempt to account for the diversity of crusta- cean muscles. Am. Zool. 13: 357-378. Bateson, \V. 1894. Materials for the Study of Variation, Treated With Especial Recant to Discontinuity in the Origin of Species: X\ l i. Mac- millan. London. New York. 598 pp. Bittner, G. D., M. L. Ballinger, and J. L. Larimer. 1974. Crayfish CNS: minimal degenerative-regenerative changes after lesioning. J. Exp. Zool. 189: 13-36. Cole, L. J. 191(1. Description of an abnormal lobster cheliped. Biol. Hull. 18: 252-268. Faxon, W. 1881. On some crustacean deformities. Bull. Mas. Comp. Zool. 8: 257-274. Fingerman, M., and S. W. Fingerman. 1974. The effects of limb removal on the rates of ecdysis of eyed and eyestalkless fiddler crabs. L'cu piixilator. Zool. Jb. Phvsiol. Bel. 78: 301-309. French, V. 1976. Leg regeneration in the cockroach, Blatel/a gcnnaii- ica II. Regeneration from a non-congruent tibial graft/host junction. J. Embiyol. Exp. Morphol. 35: 267-301. Goransson. L. G., W. P. Hunt, and S. J. Velez. 1988. Regeneration studies on a crayfish neuromuscular system. II. Effect of changing the nerve entry point into the muscle field on the gradient of innerva- tion. ./. Nciirobiol. 19: 141-152. Kao, H. VV., and E. S. Chang. 1996. Homeotic transformation of crab walking leg into claw by autotransplantation of claw tissue. Biol. Hull. 190: 313-321. Kao, H. W., and K. S. Chang. 1997. Limb regeneration in the eye sockets of crabs. Hiol. Hull. 193: 393-400. Murayama, O., I. Nakalani, and M. Nishita. 1994. Induction of lateral outgrowths on the chelae of the crayfish. Procambarus clarkii (Girard). Crust. Res. 23: 69-73. Nakatani. 1. 1996. Morphology of lateral outgrowths induced on che- lipeds of the crayfish. Procambarus clarkii (Girardl. Crust. Res. 25: 142-150. Nakatani, I., and T. Otsu. 1979. The effects of eyestalk. leg. and uropod removal on the molting and growth of young crayfish. Pro- cambarus clarkii. Biol. Bull. 157: 182-188. Nakatani. 1., and T. Otsu. 1981. Relation between the growth and the molt interval in the eyestalkless crayfish. Procambarus clarkii. Comp. Kiocliem. Physiot. 68A: 549-553. Nakatani, I., K. Vamauchi, and O. Murayama. 1992. Abnormalities found in the chela of the crayfish, Procambarus clarkii (Girard). Res. Cnist. 21: 207-209. Nakalani, I., Y. Okada, and T. Yamaguchi. 1997. An extra claw on the first and on the third cheliped of the crayfish. Procambarus clarkii (Decapoda, Cambaridae). Crit.itaccatia 70: 788-798. EXTRA CLAW ON CHELIPED OF CRAYFISH 59 Okamoto, K. 1991. Abnormality found in the cheliped of Geryim affini.i xrunuliitu.'i Sakai. Res. Crust. 20: 63-65. Przibram, H. 1921. Die Bruchdreifachbildung im Tierriche. Wilhelin Rt*i<.\ Arc/i. /inr-A/fcVi. Org. 48: 205-444. Shelton, P. M. J.. P. R. Truby, and R. G. J. Shelton. 1981. Naturally occurring abnormalities (Bruchdreitachbildungen) in the chelae of three species of Crustacea (Decapoda) and a possible explanation. J. Embtyol. Exp. Morphol. 63: 285-304. Suzuki, H., and T. Odawara. 1971. Malformation found in the cheli- peds of two edible crabs. Res. Crust. 4-5: 191-195. Suzuki, S. 1980. The effects of leg-removal to the molting of the crab. Sesarma (Holometopus) haematocheir. Res. Crust. 10: 61-68. van Harreveld, A. 1936. A physiological solution for freshwater crus- taceans. Proc. Sue. Exp. Biol. 34: 428-432. van Harreveld, A., and C. A. G. Wiersma. 1937. The triple innerva- tion of crayfish muscle and its function in contraction and inhibition. J. Exp. Biol. 14: 448-461. Wiens, T. J. 1976. Electrical and structural properties of crayfish claw motoneurons in an isolated claw-ganglion preparation. J. Comp. Phvsiol. 112: 213-233. Reference: Biol. Bull 195: 60-69. (August. 1998) Isolation and Characterization of Endostyle-Specific Genes in the Ascidian dona intestinalis MICHIO OGASAWARA* AND NORIYUKI SATOH Department of Zoology, Graduate School of Science, Kyoto University, Sakyn-ku, Kyoto 606-01, Japan Abstract. The endostyle is a special organ in the pharynx of Urochordata, Cephalochordata, and Cyclostomata. It may have arisen in the common ancestor of these taxa, along with a shift to internal feeding for extracting suspended food from the water. In addition, the endostyle has a functional homology to the vertebrate thyroid gland. The endostyle is therefore one of the structures key to the understanding of the origin and evolution of chordates. In the present study, we isolated and characterized cDNA clones for four endo- style-speciric genes, CiEiulsl, CiEntls2. CiEnds3. and CiEnds4, of the ascidian Cionu intestinalis. Although the predicted amino acid sequences of the gene products CiENDSl, CJENDS2, and CJENDS3 showed no similarity to known proteins, their mean hydropathy profiles suggest that they are secretory proteins. In addition, C1ENDS3 con- tained a unique repeat of 10 amino acids [R(QPCI)- (RRPC)!]. CiEndsl and CiEnds2 were expressed in zone 6, a protein-secreting glandular element of the endostyle, and CiEnds3 was expressed in zone 2, another secretory zone. CiEnds4, a cytoplasmic actin gene, was predominantly ex- pressed in zones 3 and 5, which are supporting elements of the endostyle. The amino acid sequences of CiENDS 1 and CJENDS2 resembled each other. In addition, they resembled a zone-6-specific gene product ( HrENDS2 ) of another ascid- ian, Halocvnthia roret~i. The results suggest that these genes are conserved among ascidian species, and therefore they (as well as CiEnds3 for the protein with a unique motif) may be useful probes for further analyses of molecular mechanisms involved in endostyle development. Introduction The endostyle is a specialized organ in the pharynx of tunicates, cephalochordates, cyclostomates, and certain Received 23 February 1998; accepted 20 May 1998. * To whom correspondence should he addressed. E-mail: ogasawara @ ascidian. /ool. kyoto-u.ac.jp prosobranchiates (Orton. 1912). The ascidian endostyle forms a trough-shaped structure in the ventral wall of the pharynx which extends from the fore-part of the pharynx to the esophagus (see Figs. 1A and 2D). In 1834, Lister first investigated how food particles in the feeding current are trapped in the pharynx of appendicularians. Fol ( 1876) found that the food is trapped by a mucous substance produced from the endostyle of tunicates. Thereafter, the ascidian endostyle has been intensively investigated by histological and ultrastructural observations (Olsson. 1963. 1965: Aros and Viragh. 1969; Fujita and Nanba, 1971); by the examination of ' ::> I incorporation (Thorpe et til.. 1972; Dunn. 1974); by histochemical detection of thyroperoxidase (Fujita and Sawano, 1979); and by partial purification of thyroperoxidase and its enzyme activity (Dunn. 1980). It is commonly considered that the endostyle of lower chordates may be a homolog and primitive antecedent of the vertebrate thyroid gland, mainly because the organ incorporates iodine (Harrington. 1957. 1958; Salvatore. 1969). However, the endostyle is a mucus-secreting and food-collecting organ, and the ability to concentrate io- dine is restricted to a small region of the organ (Olsson, 1963). The general organization of the ascidian endostyle is depicted in Fig. 1 B. The cells of this organ are differen- tiated into eight or nine strips, or zones, that run parallel to one another in longitudinal orientation. The cells of each zone are highly specialized in morphology and func- tion. The cells of zones 7. 8. and 9. like the thyroid cells of higher vertebrates, have an iodine-concentrating activity. The cells of zones 2, 4, and 6 have numerous secretory granules. These cells are believed to secrete the proteins or mucoprotein related to the digestion of food. The cells of zones 1. 3. and 5 are considered supporting elements and also as elements that might play a role in catching and transporting food. We are interested in molecular developmental mecha- 60 ASC1DIAN ENDOSTYLE-SPECIFIC GENES 61 B 8 I I supporting elements I I protein-secreting elements Illlllll iodine-concentrating elements 1 2 Zone Figure 1. Diagram of the ascidian endostyle. (A) Transverse section of the adult body showing the position of the endostyle (En) in the ventral wall of the pharynx (Ph). BWM, body wall muscle; Int. intestine; PhG, pharyngeal gill; Tu. tunic. (B) Enlargement of the endostyle show- ing compositional elements or zones of the endostyle. Zones 1. 3, and ? are supporting elements; zones 2. 4, and 6 are protein-secreting glandular elements; and zones 7. 8. and 9 are iodine-concentrating elements, equivalent to the thyroid gland of vertebrates. [Based on descriptions of Barrington (1957). Thorpe et nl. (1972). Fujita and Nanba (1471). and Dunn (1974)]. nisms that permitted or accelerated the advent of chordates. The phylum Chordata consists of the subphyla Urochordata (tunicates). Cephalochordata (amphioxus). and Vertebrata. Chordates are categorized as deutero- stomes, along with two other invertebrate groups, echino- derms and hemichordates, as was supported by molecular phylogenic studies (Wada and Satoh, 1994; Turbeville et /., 1994). Chordates share several characteristic features including a notochord. a dorsal hollow nerve cord, and pharyngeal gill slits. In addition, lower chordates (includ- ing tunicates, amphioxus, and lampreys) share an endo- style. We have emphasized that these hallmarks of the chordate body plan apparently evolved with the emer- gence of creatures resembling tadpole larvae (Satoh, 1995; Satoh and Jeffery. 1995). Therefore, investigations of the organization of these structures are of salient impor- tance in attempts to understand the origin of chordates. Coincidently with this change in the mode of larval locomotion, most of the primitive chordates or chordate ancestors may have shifted their feeding system to the use of pharyngeal gill slits for extracting suspended food from the water and of an endostyle for secreting mucus to catch the food particles (e.g., Brusca and Brusca, 1990). This possibility suggests that, in addition to the notochord and nerve cord, the pharyngeal gill and endostyle are key organs that can be used to explore molecular mechanisms involved in the emergence of chordates. In other words, we believe that the origin and evolution of chordates may be approached by isolating genes specific to the pharyn- geal gill or endostyle and by analyzing how these genes are organized during evolution in various deuterostomes. In previous studies, we isolated cDNA clones for pharyn- geal gill-specific genes (HrPliGI and HrPhG2: Tanaka et ul.. 1996) and for endostyle-specific genes (HrEnd.il and HrEnds2; Ogasawara et ai. 1996) from the ascidian Halocvnthia roretzi. which belongs to the order Pleuro- gona. Both endostyle-specific genes are expressed in zone 6 and encode secreted proteins. In the present study, we attempted the isolation of cDNA clones for endostyle- specific genes from Cionu intestinalis, which belongs to the order Enterogona. If both ascidian species conserve endostyle-specific structural genes, they may serve as models for the investigation of the molecular mechanisms involved in the organization of other chordate groups. Materials and Methods Biological materials Adults of dona intestinalis. C. savignyi. and Snela clava were collected near the Marine BioSource Educa- tion Center of Tohoku University, Onagawa, Miyagi and Otsuchi Marine Research Center, Ocean Research Insti- tute. University of Tokyo. Iwate, Japan. After the dissec- tion of adult specimens, tissues and organs were quickly frozen in liquid nitrogen, and kept at -80C until use. Isolation of RNAs and construction of cDNA libraries Total RNA was extracted from the endostyle and pha- ryngeal gill of C. intestinalis by the AGPC method (Chomczynski and Sacchi, 1987). Poly(A) + RNA was pu- rified with oligotex dT30 beads (Roche Japan. Tokyo). Complementary DNA was synthesized and cDNA librar- ies were constructed as described in a previous report (Ogasawara et al, 1996). An endostyle cDNA library was constructed using a uni-ZAP-II vector (Stratagene, La Jolla. CA). Isolation and sequencing of cDNA clones for endostyle- specific genes The endostyle cDNA libraries were screened differen- tially. Duplicate filters of the library were made; one was hybridized with a [ 32 P] -labeled total cDNA probe pre- pared from 5 fj,g of poly(A) + RNA of endostyle under high-stringency conditions, and the other was hybridized with a [ 3: P]-labeled total cDNA probe of pharyngeal gill under the same conditions. Plaques that showed positive hybridization with the endostyle probe but were negative for the pharyngeal-gill probe were selected and isolated by two rounds of screening. The specificity of the clones positive for the endostyle was confirmed by a Northern blot analysis. The clones were prepared for sequencing by controlled nested deletion from either the T3 or T7 62 M. OGASAWARA AND NORIYUKI SATOH side and sequenced using the ABI PRISM 377 DNA Se- quencer (Perkin Elmer, Norwalk. CT). Northern blot analvsis The Northern blot hybridization was carried out by the standard procedure (Sambrook et ai, 1989), and the filters were washed under high-stringency conditions. DNA probes for blot hybridizations were labeled with ["Pj- dCTP using a random primed labeling kit (Boehringer Mannheim, Heidelberg, Germany). In situ hybridization Juveniles of C. intestinalis, C. savignyi, and Styela clava were fixed in 4% paraformaldehyde in 0.5 M NaCl, 0.1 M MOPS buffer at 4C for 12 h. Probes were synthe- sized by following the instructions from the supplier of the kit (DIG RNA Labeling kit; Boehringer Mannheim). The /;/ situ hybridization of whole-mount specimens was earned out basically as described previously (Ogasawara et til., 1996). For the in situ hybridization of sectioned specimens, samples were dehydrated with a graded series of alcohol, embedded in polyester wax (BDH), and sec- tioned at 6 fj.m. Results As in the case of the isolation of cDNA clones for the endostyle-specific genes of H. roretzi (Ogasawara et al., 1996), differential screenings of a C. intestinalis endostyle cDNA library with total cDNA probes for the endostyle and pharyngeal gill yielded several cDNA clones specific to or enriched in the endostyle library. The preliminary in situ hybridization analysis of sectioned specimens dem- onstrated that transcripts of four cDNA clones were spe- cific to the endostyle. We named the corresponding genes CiEndsl (Ciona intestinalis endostyle gene J_), CiEnds2, CiEndsS, and CiEnds4. During the screening procedures, we noticed that the CiEndsl transcript was abundant in the library, representing nearly 10% of the library clones. When 1 /jg of poly(A) 1 RNA of the endostyle was elec- trophoresed, we detected the transcript as a band stained with 0.5 /yg/ml ethidium bromide (data not shown). Characterization of cDNA clone for CiEndsl As shown in Figure 2C, the Northern blot analysis of the CiEndsl transcript in various tissues and organs of a C. intestinalis adult detected the transcript of about 2.3 kb only in the endostyle. Hybridization signals were not de- tected in the pharyngeal gill, body-wall muscle, intestine, or gonad. The nucleotide sequence of the cDNA clone for CiEndsl will appear under the accession number of ABO 10895 in the DDBJ, EMBL, and GenBank nucleotide sequence databases. The insert of the cDNA clone con- sisted of 2265 nucleotides, including 17 adenyl residues at the 3' end. The clone contained a single open reading frame (ORF) of 1950 nucleotides, which predicted a poly- peptide of 650 amino acids (Fig. 2A). The calculated molecular mass (Mr) of the CiEndsl -encoded protein (CiENDSl) was 75.5k. CiENDSl did not show any sequence motifs shared by transcriptional factors, a transmembrane domain, nuclear localization signals, or motifs found in growth factor pro- teins. However, as shown in Figure 2B, the mean hydrop- athy profiles of CiENDSl showed that the N-terminus was highly hydrophobic. This region had a typical signal peptide sequence that consisted of a positively charged residue (amino acid position 2; K, Lys), a hydrophobic (3-13) region of 10-15 residues, a charged residue (posi- tion 15; S, Ser), and a residue containing the short side chain (position 16: A. Ala). A predicted cleavage site of the signal peptide was evident behind the Ala (position 16). This sequence motif strongly suggests that CiENDSl is a secretory protein, with a probability of 82% deter- mined by using the PSORT Program (Online. PSORT World Wide Web Server: Available: http://psort.nibb. ac.jp). In addition, CiENDSl contained four putative N- linked glycosylation sites (Fig. 2 A). The //; situ hybridization of whole-mount specimens demonstrated that the signals were restricted to the endo- style (Fig. 2D). In addition, the in situ hybridization of sectioned specimens demonstrated that the signals were not distributed over the entire regions of the endostyle but rather were restricted to zone 6 (Fig. 2E). No signals above background level were found in the control speci- men hybridized with the sense probe (data not shown). Characterization of cDNA clone for CiEnds2 The nucleotide sequence of cDNA clone for CiEnds2 will appear under the accession number of ABO 1 0896 in the DDBJ. EMBL, and GenBank nucleotide sequence databases. The insert of the cDNA clone consisted of 2107 nucleotides, including 25 adenyl residues at the 3' end. The occurrence of a 2.3-kb-long CiEnds2 transcript only in the endostyle (Fig. 3C) suggested that the cDNA was close to full-length. The clone contained a single ORF of 1950 nucleotides, which also predicted a polypeptide of 650 amino acids (Fig. 3A). The calculated Mr of the C/m/\2-encoded protein (CJENDS2) was 77.3 k. C1ENDS2 may also be a secretory protein. As shown in Figure 3B, the mean hydropathy profiles of C1ENDS2 showed that the N-terminus was highly hydrophobic. This region had a typical signal peptide sequence that consisted of a positively charged residue (amino acid position 2; K, Lys), a hydrophobic (3-13) region of 10-15 residues, a charged residue (position 15; N, Asp), and a residue ASCIDIAN ENDOSTYLE-SPECIFIC GENES 63 B MKVLL I LLAF I AAASAFSYGNGYGYGYNKCY6SYKGYSSGCYSYGYRK CYVYPKSQVFCYN I PYKKSWCSYKYYEPVLHVYPGCDCGTEGWTEKTV ADLE I EMTNLLKEALLK I TTEMNNCKTTFVEQLKSS I EQYKLNVKNKL FNYYAYY I QSAKTDEERENL I KKRDDA I KEYNEELDKKRDDV I LKCEE DVADKLKC I ADYHTKLVENGVECLKTRLTK I VDYTTTLTAKCVOYVKN YVACHMS I LEQKKSYYRSFLHKVHGSSEWEKVTVOAV I QLYHOQEVAK I TALATEYATKLATWKLKL I MNYSCAYRCYMSNGC I RFYKKRYYSTCK RYGCWYKYKTRYCFVRYCLQPFKFCFNPTKYTGLKTCVFPAVVRDGAT I IKEHCEKLEKAILEYETQFGEWKLKWTTYHTEYCTKYDEI IKARHDW Y I EYLRSQY I CANNSTELTDEQKAKLAEVQKECDEKRTAAVEAYKLKL AE I LLECATKFSTS 1 1 DYRTKAKSY I QS I GDNFDACQKKRSED I KAYK EKLEAKG I SAKQSLFDSMNKAKKSHLTKYLD I LKLHHDOFVVNGDVCD GPTD I VTMANKYSEKLSEYCAAVLLECDKYWED 1 1 RTGRTSSLGWYST IPAATSARNRSASCPDSAGWITHGA 65033 D 3.00 0.00 -3.00 i 141 281 421 561 701 Si Q. g - m s. o 2.3kb Figure 2. Characterization of the CiEmlsl gene. (A) The predicted sequence of 650 amino acids of CiENDSl . The predicted signal peptide sequence is shown by red capitals, and putative N-linked glycosyla- tion sites by green capitals. The nucleotide sequence for CiEmlsl will appear under the accession number of ABO 10895 in the DDBJ, EMBL, and GenBunk nucleotide sequence databases. IB) Mean hydropathy index of the CiENDSl calculated across a window of 19 residues according to the method of Kyle and Doolittle (1982). The N-terminus of the protein is characterized by a 16-amino-acid-long hydrophobic region that contains the predicted signal peptide sequence (see text for details). This suggests that CiENDSl is a secretory protein. (C) Distribution of CiEndsl transcript in tissues and organs of the adult. Northern blots of poly(A) + RNA prepared from the endostyle (En), pharyngeal gill (PhG). body-wall muscle (BWM). intestine (Int), and gonad (Gonad) were hybridized with the random-primed ["P]-labeled DNA probes, and the membrane was washed under high-stringency conditions. The CiEndsl transcript of about 2.3 kb in length was detected only in the endostyle. Each lane was loaded with 8 /j.g of poly(A) + RNA. (D, E) Localization of CiEndsl transcript, as revealed by in situ hybridization. IS. incurrent siphon; OS, outcurrent siphon. (D) A whole-mount specimen of a 1 -month-old young adult and (E) a cross-section of an adult showing that the signal is restricted to the endostyle (D. arrow) and to zone 6 of the endostyle (E, arrows). containing the short side chain (position 16; A, Ala). A predicted cleavage site of the signal peptide was evident behind the Ala (position 16). However, the results of the PSORT Program calculation showed that CJENDS2 seems to have an uncleaved N-terminal signal sequence. In addition, CiENDS2 contained a putative /V-linked gly- cosylation site (Fig. 3A). CJENDS2 showed 48.3% iden- tity at the amino acid level with CiENDSl, a rinding that will be discussed later. The //i situ hybridization of whole-mount specimens demonstrated that the signals were restricted to the endo- style (Fig. 3D), whereas that for sectioned specimens demonstrated that the signal was restricted to zone 6 (Fig. 3E). No signals above the background level were found in the other zones. Characterization of cDNA clone for CiEnds3 The nucleotide sequence of cDNA clone for CiEndsS will appear under the accession number of ABO 10897 in the DDBJ, EMBL. and GenBank nucleotide se- quence databases. The insert of the CiEndsS cDNA was 64 M. OGASAWARA AND NOR1YUKI SATOH AMK I LFVLLAALAVTNAYSYTHCGCKRFLVYPKPQAFCYH I QYKSCWCS YKYYEPVLHVYPGKDCGTLGWTETP I EOLQTEMENLLKESLLE I TKKM MLRKWFVRQLTKTADEYKEEYKKN I TRYYAYY I ASADS I LRKEHL I K ERNEL I EEYNEQLDQKVTAAVEKCTTO I LAK I K I I AAYHEKLVKSAVG CLETREQK I EEY I TTLENKCKLYVGRFVSKHLA I LEQNKKYYRATLAK VHGSALWQKAKVDAV I EVYHEQEAKK I SVLAKEYAOTLNTCKAKL I TN YRCAYKCYMSNSCLRFYKKTYYSYCRNLGCWYKYTPSYCVVRKTLCPF YYPYKP I SFSCLRTCVVPAVVRNGAT I I KELEEKLEKA I KE YVV I FTA WKTKWTQYHTEYCDKYNE I LKERHEWY I RRV I ARFVVENNSTELTALQ KAE I AKLKKELNERRTEAVLTYKKKLLNLLLECVAKFNEN I EEYKQKA LDL I KG I AKAFDACLTKRKAO I LAYRFKM I KHSFSEKE I MRKDM I KSK LVHLRHYKEMLKTYHDGEEFPEEVNAMI EAYEDKLENYCED I LEECTK DWSAA I PKLTHHYACSYTCREGRFCMPKFCFGGYFRf VVKYPTAKCYT YYYRCYRTTCNYRVFYQKGCYRYGCH gen gg B 3.00 OS c C C. > * O 111 n m C en 2.3kb Figure 3. Characterization of the CiEnds2 gene. (A) The predicted sequence of 650 amino acids of CiENDS2. The predicted signal peptide sequence is shown by red capitals, and a putative N-linked glycosyla- tion site by green capitals. The nucleotide sequence for CiEiulx2 will appear under the accession number of ABO 1 0896 in the DDBJ. EMBL. and GenBank nucleotide sequence databases. (Bl Mean hydropathy index of the OENDS2. calculated as in the case of CiENDSl. (C) Distribution of CiEnds2 transcript in tissues and organs of the adult. The CiEnds2 transcript of about 23 kb in length was detected only in the endostyle. Each lane was loaded with 8 fjg of poly(A)* RNA. (D, E) Localization of CiEnds2 transcript, as revealed by HI situ hybridization. IS, incurrent siphon. (D) A whole-mount specimen of a 1-month-old young adult and (E) a cross-section of an adult showing that the signal is restricted to the endostyle (D, arrow) and to zone 6 of the endostyle (E, arrows). 1319 nucleotides, including 17 adenyl residues at the 3' end. The Northern blot analysis demonstrated the occurrence of a 1 .4-kb long CiEnds3 transcript, which was detected only in the endostyle (Fig. 4C). This sug- gested that the cDNA was close to full-length. The clone contained a single ORF of 945 nucleotides, which predicted a polypeptide of 315 amino acids (Fig. 4A). The calculated Mr of the CiEnds3-encoded protein (CiENDS3) was 35.3 k. CiENDS3 may also be a secretory protein. As shown in Figure 4B, the mean hydropathy profiles of CiENDS3 showed that the N-terminus was highly hydrophobic. This region had a typical signal peptide sequence that consisted of a positively charged residue (amino acid position 2; R), a hydrophobic (3-13) region of 10-15 residues, a charged residue (position 15; T), and a residue containing the short side chain (position 16; C). A predicted cleavage site of the signal peptide was evident behind the Cys (position 16). This sequence motif strongly suggests that C1ENDS3 is a secretory protein, with a probability of 58% determined by the PSORT Program. In addition, CiENDS3 contained a unique repeat of 10 amino acids (Fig. 4A). The repeat consisted of R(QPCI)(RRPC)I. This type of repeat has not been re- ported to date in the PDB, SWISSPROT. and PIR data- bases surveyed. The //; situ hybridization of whole-mount specimens demonstrated that the signals were restricted to the endostyle (Fig. 4D). In addition, the ;/; ,v//i/ hybridiza- tion of sectioned specimens demonstrated that the sig- nal was restricted to zone 2 (Fig. 4E). No signals above the backuroimd level were found in the other zones. ASCIDIAN ENDOSTYLE-SPECIFIC GENES 65 B MRFL I LLCLLFTTATCQE I SVSASAETLANE I LOG I TLDD I ELPE I NLPSQP I SVTQ I PSVQR I RQPSIR QP 1 1 RQPC I RQPCQPC I RKPC I RQPC I RRPCPPCV RKPC I RRPCQPC I RRPGPPC I RKPCQPC I RRPCPP C I RKPC I RRPCQPC I RRPCPPC I RKPC I RQPGQPC I RKPCQPC I RRPGQPC I RRPCQPC I RRPCQPCVRQ PCQPC I RRPCQPC I RRPCQPC I RRPCLPLKRPC I T PTPT I EPP I TEPAPEVDGTTTASATATVTTDGTTT VTETTTDGNAEAVAE I AEALSGVTRQ I FLTSFK I L 315aa RQPCI RRPC I 3.00 0.00 -3.00 65 129 193 257 321 c j > - o ui a. m JE ' -.. 1.4kb Figure 4. Characterization of the CiEnds3 gene. (A) The predicted sequence of 315 amino acids of C1ENDS3. The predicted signal peptide sequence is shown by red capitals. The sequence contained a characteristic repeat [R(QPCI)(RRPC)I]. The underlined amino acids are identical to those of the consensus sequences. The nucleotide sequence for CiEnds3 will appear under the accession number of ABO 10897 in the DDBJ, EMBL. and GenBank nucleotide sequence databases. (B) Mean hydropathy index of the CiENDS3 calculated as in the case of CiENDSl . (C) Distribution of CiEndsS transcript in tissues and organs of the adult. Northern blots of poly(A)* RNA prepared from the endostyle (En), pharyngeal gill (PhG). body-wall muscle (BWM). intestine (Int), and gonad (Gonad) were hybridized with the random-primed ["P]-labeled DNA probes, and the membrane was washed under high-stringency conditions. The CiEnds3 transcript of about 1 .4 kh in length was detected only in the endostyle. Each lane was loaded with 8 //g of poly( A)* RNA. (D, E) Localization of CiEnds3 transcripts, as revealed by in situ hybridization. IS. incurrent siphon; OS, outcurrent siphon. (D) A whole-mount specimen of a 1 -month-old young adult (scale bar is 1 mm) and (E) a cross-section of an adult (scale bar is 100 /urn) showing that the signal is restricted to the endostyle (D, arrow) and to zone 2 of the endostyle (E. arrows). Characterization of cDNA clone for CiEnds4 The in xitu hybridization to isolate endostyle-specific cDNA clones demonstrated that the transcript of another clone (CiEnds4) was restricted to the zones 3, 5, and 7 (Fig. 5A). Strong signals were evident in zones 3 and 5, supporting elements of the endostyle; weak signals were also evident in zone 7, a putative iodine-concentrating element (Fig. 5 A). The insert of the CiEnds4 cDNA was 1541 nucleotides, including 19 adenyl residues in the 3' end. The clone contained a single ORF of 1 125 nucleotides. which pre- dicted a polypeptide of 375 amino acids (data not shown). A cDNA clone for C. intestinalis cytoplasmic actin has been isolated and characterized in the laboratory of Dr. Takahito Nishikata of Konan University. Kobe, Japan (pers. comm.). The determination of the partial nucleotide sequence of CiEnds4 cDNA clone revealed that CiEnds4 encodes a cytoplasmic actin. The above-mentioned results suggested that the gene or genes for one or more cytoplasmic actin are actively 66 M. OGASAWARA AND NORIYUKI SATOH C. savignyi B 1.6kb o_ m . I o C. intestinalis Figure 5. Characterization of the CiEnds4 gene. The predicted amino acid sequence of CiENDS4 suggested that CiENDS4 is a cytoplasmic actin. (A) Localization of CiEnds4 transcripts, as revealed by in MIII hybridization. A cross-section of a young adult showing signals in zones 3, 5. and 7 of the endostyle. (B) Distribution of CiEnds4 transcript in tissues and organs of the adult. Northern blots of poly(A)* RNA prepared from the endostyle (En), pharyngeal gill (PhG). body-wall muscle (BWM), intestine (Int). and gonad (Gonad) were hybridized with the random-primed |' : P]-laheled DNA probes, and the membrane was washed under high-stringency conditions. The CiEmls4 transcript of about 1. 6kb in length, or closely related molecules, was detected in the organs examined. Each lane was loaded with 8 /ug of polylAT RNA. (C, D) Cross-reactivity of CiEnds4 probe with zone 3 (arrows) of the endostyle of (C) C. savignyi and (D) Styela clava. as revealed by in .\itii hybridization (scale bars in A, C, and D are 100 pm). expressed in the supporting elements of the endostyle. We therefore examined, using the CiEmls4 probe, whether the cytoplasmic actin gene is expressed in the supporting elements of the endostyle of two other ascidian species, C. savignyi and Styela clcmi. As shown in Figure 5C and D, the CiEmls4 probe detected cytoplasmic actin tran- scripts in zone 3 of both species. In addition, the Northern blot analysis demonstrated the occurrence of a 1 .6-kb-long CiEmls4 transcript not only in the endostyle but also in other organs including the pharyn- geal gill, body-wall muscle, intestine, and gonad (Fig. 5B). Discussion In the present study, we isolated cDNA clones for four uenes (CiEndsl, CiEmls2, CiEndsJ, and CiEnds4), which are expressed in different zones of the ascidian endostyle. CiEndsl and CiEnds2 are expressed in zone 6 and encode peptides with similar amino acid sequences. CiEndsB is expressed in zone 2 and encodes a polypeptide with a novel repeat of 10 amino acids, tentatively called "ends- repeat." CiEnds4 encodes a cytoplasmic actin that is ex- pressed mainly in zones 3 and 5. In a previous study, we characterized cDNA clones for two endostyle-specific genes. HrEndxl and HrEnds2. from the ascidian H. roretzi (Ogasawara el til.. 1996). Both genes are expressed in zone 6 and encode secreted proteins. Transcripts of both genes are abundant in the endostyle library; each represents about 10% of the cDNA clones of the library. As revealed by the present study. CiEndsl is also expressed in zone 6 and encodes a se- creted protein. This transcript also represents the most ASCIDIAN ENDOSTYLE-SPECIFIC GENES 67 abundant species in the library. These results strongly suggest that zone 6 plays a major role in the secretion of mucus by the endostyle and has functions different from those of zones 2 and 4. This notion is consistent with a previous ultrastructural observation that the size and structure of the secretory granules in zone 6 differ from those of the other glandular zones 2 and 4 (Aros and Viragh, 1969). Zone 6 occupies the largest area in the endostyle and is characterized as the most developed glan- dular zone, containing abundant endoplasmic reticulum (Fujita and Nanba. 1971). Aros and Viragh (1969) and Fujita and Nanba (1971) reported that zone 6 contains at least two types of secretory granule a large electron- dense granule and a smaller granule. Directly facing the pharynx in zone 6 is a wide exit for secretion; in contrast, zones 2 and 4 have only a very limited exit for secretion (Thorpe et aL 1972). When the amino acid sequences are compared for CiENDSl and CJENDS2 (Fig. 6) and for CiENDSl, HrENDS2, and C1ENDS2 (Fig. 7), these three polypep- tides closely resemble each other: the sequence identity was 48.3% between CiEndsl and CiEnds2, 22.2% be- tween CiENDSl and HrENDS2 (similarity 43.7%). and 22.8% between CiENDS2 and HrENDS2 (similarity 42.6%). Our previous genomic Southern blotting analysis of HrEnds2 suggested that there are some other genes in the H. mretzi genome that contain a sequence similar to that of HrEnds2. In the present study, we isolated two cDNA clones from C. intestinalis that contained a se- quence similar to that of HrEmls2. It is therefore likely that CiENDSl, CiENDS2, and HrENDS2 are members of the same protein family, and it is possible that these genes were derived from a common ancestral gene. The first aim of our studies is to isolate genes that are expressed in certain zones of secretory function, these genes being common in different species that be- long to different orders of uscidians. In the present study, we first attempted the screening of C. intestinalis homologs of HrEndsl and HrEnds2 with low-strin- gency hybridization conditions, using these genes as probes. Unfortunately, we could not isolate any homo- logs with the conditions we adopted. Therefore, we next tried the differential screening we used in a previ- ous study (Ogasawara et al., 1996), because we thought that if the nature of the endostyle is the same between C. intestinalis and H. roret-i, we might isolate homolo- gous genes easily using this method. In the present screening, we were able to isolate an endostyle-specific cDNA clone for CiEndsl; this clone was highly ex- pressed in zone 6 and contained a sequence similar to that of HrEnds2. In the further screening, we isolated CiEnds2, which was also expressed in zone 6 and con- tained sequences similar to that of HrEnds2. Ciendsl, CiEnds2, and HrEnds2 may be related to each other and play an important role in the ascidian endostyle, and therefore they are good candidates for future stud- ies. Interestingly, the CiEnds3 encodes a polypeptide with a novel repeat of 10 amino acids [the core repeat is 8 amino acids (QPCI)(RRPC)|. We tentatively call this the "ends-repeat." Because of its uniqueness and conserva- CiEnds! CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 CiEndsl CiEnds2 WYSTIPAATSA-RNRSASOSDSAG YTCREERFCMPKFCFGGYFRWVVKYGTAKC 559 Dl 537 E 644 VVITHGA 623 YTYYYRCYRTTCNYRVFYQKGCYRYGCH Figure 6. Alignment of the amino acid sequences of CiENDSl and CiENDS2. Identical amino acids are enclosed by boxes. Gaps were induced to obtain the maximal similarity. 93 71 186 164 279 257 372 350 465 443 558 536 643 622 650 650 68 M. OGASAWARA AND NORIYUKI SATOH CiEndsl aa HrEnds2 aa CiEnds2 aa CiEndsl aa HrEnds2 aa CiEnds2 aa CiEndsl aa HrEnds2 aa CiEnds2 aa CiEndsl aa HrEnds2 aa CiEnds2 aa CiEndsl aa HrEnds2 aa CiEnds2 aa 315 CiEndsl aa HrEnds2 aa CiEnds2 aa CiEndsl aa HrEnds2 aa MKVLLILLAFIAAASAFSY6N MKILIVLLSCLAVASAFGYGGYSRYSRYGRViNKYPSYGGYSKYSSYGSYKY-PSye MKILFVLLAALAVTNAYSYTHCGC-KlFLvl -RKYKQ 60 61 NIPYKKSCSYKYYEPVLHVYP;GCDCGTEG1TEKTVADLE|EMTNI gg KGLFKNE|CTTAFYSDVLSYYPpDAKNlLP-KANATDQVV|---A| ft HIQYKSCISCSYKYYEPVLHVYPGKDCGTLGWTETPIEDLQtlEMENi nm fragum, I', iiucilo), and an azooxanthellate cardiid clam (Vasticardium flavum) were collected in the Repub- lic of Palau, Western Caroline Islands, during the cruises of the R/V Sohgen-mani in 1993-1994. T. gigas, T. de- rasa, and H. porcellanus were purchased from the PMDC (Palau Mariculture Demonstration Center). Live speci- mens of a zooxanthellate cardiid, Corciiliim cardissa, which had been collected from mariculture tanks for tri- dacnid clams, were gifts from PMDC. A live specimen of Fragnin fragum collected in Okinawa, Japan, was a gift from Mr. Osumi of Ryukyus University. An azooxan- thellate cardiid clam. Fulvia tnutica, was purchased from a local fishery market in Shimizu, Japan. The shells were saved (except for those of T. gigas, which were lost during transportation) and were kindly identified by Prof. Oku- tani of Nihon University. Extraction of DNA from the clams Live clams were dissected, and the organs were sepa- rated and frozen at -20C until used. The white part of the gill or foot tissue (0. 1-0.2 g), which was devoid of zooxanthellae, was homogenized in a glass homogenizer in 2-3 ml TE buffer ( 10 mM Tris-HCl, 100 mM EDTA, pH 8.0) containing 0.5% sodium dodecyl sulfate, and digested with proteinase K (100/ug/ml) at SOT for 3 h (Wada et ai, 1992). The DNA was then extracted with TE buffer-saturated phenol and washed twice with a mix- ture of chloroform and isoamyl alcohol (24:1). An equal volume of 5 M ammonium acetate was then added, and the DNA was precipitated with cold ethanol, washed once with cold 70% ethanol, and dried. Amplification of 18S rDNA and DNA sequence determination A polymerase chain reaction (PCR) kit (Takara Shuzo, Kyoto, Japan) was used to amplify the 18S rDNA from 1 pg of genomic DNA. The manufacturer's instructions were followed; i.e.. 30 cycles comprising 93C for 1 .5 min. 58C for 1 .5 min, and 72C for 2 min. The exten- sion reaction at 72C in the final cycle was prolonged to 10 min, and the PCR products were then frozen at -20C until used. The universal eukaryotic primers used in the amplification of the 18S-rDNA were 5'-GGTTGAT- CCTGCCAGTAGTCATATGCTTG-3' (ss5) and 5'- GATCCTTCCGCAGGTTCACCTACGGAAACC-3' (ss3). These sequences were reported to be located four nucleotides from the 5' and 3' ends, respectively, of the 18S-rRNA of Prorocentrum micans, a dinoflagellate (Herzog and Maroteaux, 1986; Rowan and Powers, 1992). Amplified DNAs were cloned in pT7 plasmids (Nova- gene. USA) with E. coli JM 109 as the host. At least three cloned DNAs amplified from the genomic DNA of an individual clam were sequenced for each clam species, using custom-made 20-b DNA primers (Japan BioSer- vice, Saitama), a Takara PCR sequence kit with an ABI- type 373A DNA sequencer. Analysis of the sequence data The 18S rDNA sequences were aligned using Clustal W (Thompson et ai, 1994) and gap and insertion regions removed with the software program RMINSDEL in MOLPHY-2.2 (Adachi and Hasegawa, 1994). The aligned sequences were then analyzed by the neighbor- joining (NJ) method (Saitou and Nei, 1987) using the DNADIST and NEIGHBOR programs (using the Kimura 2-parameter model; Kimura, 1980) in PHYLIP 3.57c (Felsenstein, 1995). The statistical significance of each cluster in the tree was evaluated with 1000 iterations of bootstrap resamplings and tree reconstructions (Felsenstein, 1985) using DNADIST, NEIGHBOR, SEQ- BOOT, and CONSENSE in PHYLIP 3.57c. The same sequences were analyzed by the maximum parsimony 72 T. MARUYAMA ET AL (MP) method using the DNAPARS program in PHYLIP 3.57c. Decay analysis (Bremer, 1988, 1994; Winnepen- ninckx el al., 1996) was also undertaken by using PHY- LIP 3.57c. The sequence data of nine selected clams were further analyzed by the maximum likelihood (ML) method using the NUCML program adopting the Alpha/ Beta ratio 2.0 model (Hasegawa et al., 1985 ) in MOLPH Y 2.2 (Adachi and Hasegawa, 1994). The tree with minimal AIC (Akaike Information Criterion), which is defined as -2 (log-likelihood) + 2 (number of free parameters) (Akaike, 1974), was considered the most appropriate tree. The statistical significance of each cluster in the ML tree was evaluated by the bootstrap analysis with 1000 itera- tions. Phylogenetic trees were drawn with TreeView (Page, 1996) on a Macintosh computer. Results The PCR products Lengths of the amplified 18S rDNAs (between the primers ss5 and ss3) from the zooxanthellate and azoo- xanthellate cardiid clams were 1780-1875 bp longer than the corresponding lengths in the other clams exam- ined (1745-1772 bp) (Table I). Those sequences were deposited in the DDBJ (DNA data base of Japan) under the accession numbers shown in Table I. The clam shells (except T. gigas) were deposited in The National Science Museum in Tokyo with the specimen numbers listed in Table I. Phvlogenetic trees The alignment of 18S rDNA sequences of the clams is available on the web site of this journal at the following address: http://www.mbl.edu/html/BB/home.BB.html After removing the gaps and insertions, we analyzed 1698 sites within the sequences by the NJ and MP methods. Figure 1 shows a phylogenetic tree constructed by the NJ method. The 18S rDNA sequence of LiinicoUirici kambeitl (Gastropoda, Pulmonata. Achatinidae) was chosen as the outgroup. We also analyzed the data with the 18S rDNA Table I Length of amplified IXS-rDNAs Length of the PCR Accession Specimen Species Place of collection products' number number' Zooxanthellate bivalves (present work) Tridacna gigus Palau 1780 bp D84189 n.d. Tridacna derasa Palau 1803 D84658 NSMT-Mo70918 Tridacna crocea Okinawa ( Japan I 1803 D88908 NSMT-Mo70917 Tridacna maxima Palau 1805 D84659 NSMT-Mo70919 Tridacna sqiiamosa Palau 1804 D84190 NSMT-Mo70916 Hippopus hippopus Palau 1813 D84660 NSMT-Mo70914 Hippopus pOTcellanus Palau 1813 D84661 NSMT-Mo70915 Fragum fragum Palau 1865 D84662 NSMT-Mo70921 Fragum fragum Okinawa (Japan) 1865 D84663 NSMT-Mo70920 Fragum unedo Palau 1865 D84664 NSMT-Mo70922 Corcuhim cardissa Palau 1875 D88909 NSMT-Mo70923 Azooxanthellate Cardiidae bivalves (present work) Vasticai'diiim fiavuni Palau 1852 DS8910 NSMT-Mo70913 Fulvia nuitica Mikawa-bay (Japan) 1840 D88911 NSMT-Mo70912 Other bivalves (from database) Spisula solidissima 1772 LI 1270 Tresu* nuttali 1772 LI 1269 Placii/wti'n magellanicus 1745 X53899 Chlamvs islandiCQ 1746 LI 1232 Mytilux edulis 1757 L24489 Gastropod (from database) Limicolariu kamhcul 1770 X66374 1 Length of the amplified DNA belween but excluding primers ss5 and ss3. ; Accession number of the DNA sequence data in DNA Data Base of Japan. ' Deposit number of the shell specimen in National Science Museum, Tokyo; n.d., not deposited. PHYLOGENY OF CLAMS WITH ZOOXANTHELLAE 73 0.98 r Spisuk 1.000 L Tresus 0.870 1.000 solidissima nuttali 1.000 ^Vasticardium flavum \Fulvia mutica |- 1.000 JfHippopus hippopus 1.000 |1 1}IW ^H 1.0001 1.000 0.993 0.677 1 Hippopus porcellanus ! Tridacna gigas Tridacna derasa Tridacna crocea Tridacna maxima \ Tridacna squamosa 000 r\ Corculum cardissa 1 Fragum fragum UKI Fragum unedo 0.1 s |" Placopecten magellanicus .000 L Chlamys islandica Figure 1. A phylogenetic tree of zooxanthellate and azooxanthellale clams, calculated by the neighbor-joining method adopting the Kimuru 2-parameter model. Numbers at the nodes are the bootstrap values for the clades in 1000 replications. Box. zooxanthellate clam. Bar, 0.10 substitutions per site. sequence of Symbiodinium corculorum, the symbiotic alga of Corcitlwn cardissa (accession number LI 37 17) as the outgroup; the results were the same as those with L. kambeul (data not shown). Clams belonging to the Pteriomorphia (Mytilus edulis in Mytilidae, and Placopec- ten magellanicus and Chlamys islandica in Pectinidae) made a clade that was clearly separated from those of the Heterodonta. In the heterodont branch, a clade including the families Tridacnidae and Cardiidae and comprising all of the zooxanthellate clams examined, as well as two azooxanthellate clams, V. flavuin and F. mntica was clearly separated from another clade containing two spe- cies in the family Mactridae (Spisula solidissima and Tre- at m nuttali). This clade was further resolved into three lineages: Corculum-Fragum, Vasticardium-Fulvia, and Tridacna-Hippopus. The NJ tree indicates that the zoo- xanthellate cardiids and azooxanthellate cardiids diverged before the zooxanthellate tridacnids. The bootstrap value at the node uniting the Vasticardium-Fulvia and Tri- dacna-Hippopus clades was 0.870, and that at the node uniting the Vasticardium-Fulvia-Tridacna-Hippopus and Corculum-Fragum clade was 1.000. This indicates that tridacnid clams are more likely to form a clade with clams of Vasticardium-Fulvia than with those of Fragum-Cor- culiim. Three tridacnid clams. T. maxima, T. squamosa. and T. crocea, are evidently very closely related, whereas T. gigas and T. derasa are relatively distant from them. Fragum fragum and F. unedo are very closely related, the identity between their 18S rDNA sequences being 99.8% (1861 out of 1865 bp). The maximum parsimony (MP) method gave three equally most parsimonious trees of similar, but distinct, topologies (Fig. 2, A-C). The lengths of the trees were Limicolaria kambeul Mytilus edulis Chlamys islandica Placopecten magellanicus Tresus nuttali Spisula solidissima Fulvia mutica Vasticardium flavum Tridacna crocea 0.957] i JTridacna maxima 0.326 [ Tridacna squamosa 1 .000 ^IHippopus porceWanusI fiippopus hippopus [ [CorcuJum cardissa \ unedo 0.999 fragum 1.000 Tridacna gigas Tridacna derasa Tridacna maxima Tridacna crocea \ Tridacna squamosa 1.000 0.326 Tridacna gigas Tridacna derasa Tridacna squamosa Tridacna maxima Tridacna crocea Figure 2. Phylogenetic trees of zooxanthellate and azooxanthellate clams, calculated by the maximum parsimony method. Three topologies were selected as the best tree; the differences were found in Triilucim .ii/iiainnsa, T. crocea, and T. maxima. (A) One of the best trees. (B and C) Parts of the two other trees showing the dissimilarities to A. Numbers at the nodes are the bootstrap values tor the clades in 1000 replications. *, decay index. Box, zooxanthellate clam. Consistency index, 0.76: retention index, 0.88. 74 T. MARUYAMA ET AL. 1019 steps. Consistency and retention indices of the three trees were 0.76 and 0.88, respectively. These trees had essentially the same topology as that obtained by the NJ method, except for the positions of the three closely re- lated clams in the genus Tridacna (T. squamosa, T. cro- cea, and T. maxima). These differences suggest that the three species are closely related. In this tree, bootstrap value for the clade containing the tridacnid clams and Vasticardium-Fulvia was high (0.901). The decay index, which is the number of extra steps required before this clade collapsed, was calculated to be 6. These data sug- gest that the azooxanthellate curdiid lineage is more likely to form a clade with the zooxanthellate tridacnid clam lineage than with the zooxanthellate cardiid clam lineage. Lineages Tridacna- Hippopus- Vasticardium-Fulvia and Corculum-Fragum formed a clade, but the bootstrap value was relatively low (0.519). Due to the computational limitations, we selected nine clam species for a maximum likelihood analysis; one from subclass Pteriomorphia (Mytilus edulis) as an outgroup. two from zooxanthellate Cardiidae (Coirnlum cardissa and Fragum fragum), two from azooxanthellate Cardiidae (Vasticardium flavum and Fulviu niutica), two from the genus Tridacna (Tridacna gigas and T. derasa), and two from Hippopus (H. hippopus and H. porcellanus). A total of 1713 sites were used for the ML analysis. The tree topology (Fig. 3), was essentially the same as that ob- tained by the NJ method. The bootstrap value at the node ofTridacna-Hippopus and Vasticardium-Fulvia was 0.76, and that of Tridacna-Hippopus-Vasticardium-Fulvia and Corculum-Fragum was 1 .00. The differences between the minimal AIC (Akaike Information Criterion, as defined in the Materials and Methods) and the AICs of the other possible trees were greater than 15.5 (1.0), which suggests that the tree topology shown in Figure 3 is statis- tically significant. To examine the genetic difference between individuals from two distant locations, 18S rDNA sequences of F. fragum collected from Okinawa and Palau were com- pared. The identity between them, 99.9% (1864 out of 1865bp). indicated no significant difference between them. Insertions and gaps Table II shows the major gaps and insertions found in 18S rDNAs of cardiids and tridacnids when aligned with those of other clams, especially Limicolaria kambeiil, listed in Table I. Designations of the regions of the 18S rRNA are shown according to Winnepennickx ct al. ( 1992). The table includes any insertions and gaps longer than 10 b in any of the cardiids and tridacnids. Corre- sponding smaller gaps and insertions in other zooxanthel- Mytilus edulis 1.00 L Corculum cardissa Fragum fragum i.oo Hippoppus hippopus 0.76 i.oo Hippopus porcellanus 1.00 Tridacna derasa Tridacna gigas 1 .00 L Vasticardium flavum Fulvia mutica 0.1 Figure 3. A phylogenetic tree of nine selected bivalves, including zooxanthellate and azooxanthellate clams, calculated by the maximum likelihood method. Bar indicates 0.1 base substitutions per site. Numbers at the nodes are the bootstrap values for the clades in 1000 replications. Box. zooxanthellate clam. late clams are also shown. Insertions in the El 0-1 region are restricted to the azooxanthellate cardiid calms, V. fla- vum and F. mutica. On the other hand, zooxanthellate clams have gaps in this region. Some of those insertions were aligned and are shown in Figure 4. The identity between El 0-1 insertions of V. flavum and F. mutica is low (70%). While the inserts in Corculum and Fragum are similar to each other, those in Hippopus are different from those of Corculum and Fragum. Inserts in region 47 are restricted to Tridacna, but T. gigas lacks this insertion. Those of T. squamosa and T. maxima are the same: one base different from that of T. crocea, and two bases differ- ent from that of T. derasa. Discussion The tree topologies based on 18S rDNA sequences (Figs. 1-2) are consistent with the morphological taxon- omy of Mytilidae, Pectinidae, and Mactridae (Morton. 1996). The family Tridacnidae comprises two genera. Tri- dacna and Hippopus (Rosewater. 1965, 1982). The genus PHYLOGENY OF CLAMS WITH ZOOXANTHELLAE 75 Table II Major insertions and gaps in 1SS rDNAs in tridacnid and cardiid clams in comparison with those of Limicolaria kambeul Regions* Positions in the alignmentt Size (b) Species:): Insertions E10-1 247-269 23 Vf 22 Fm E21-2 768-782 15 Cc 12 Ff, Fu 11 Hh, Hp 4 Vf. Fm 3 Tg 4! 1484-1513 30 Ff. Fu 27 Cc 4 Vf. Fm. Hh. Hp. Tg. Td. Tc. Ts. Tm 47 1879-1910 32 Ts. Tc. Tm 31 Td 8 Cc 6 Fm 5 Fu. Ff, Vf, Tg Gaps E10-1 232-281 25 Hh, Hp. Tg. Td, Tc, Ts, Tm 232-246 14 Cc, Fu, Ff Gaps and insertions were detected in the aligned 1 8S rDNAs listed in Table I. * Regions in 18S rRNA of Limicolaria kambeul designated in Winnepennickx el al. (1992). t The alignment is available at the following URL: http://www.mbl.edu/htrnl/BB/home.BB.html t Cc, Corcitlum cardissa: Fu, Fragitm unedo; Ff, Fragum fragitm; Vf, Vasticardiinn flavum; Fm, Fulvia inutica; Hh, Hippopits hippopus; Hp, Hippopus porcelhmus; Tg, Tridacna gigas; Td, Tridacita derasa: Tc, Tridacna crocea: Ts, Tridacna squamosa; Tm, Tridacna maxima. S Very close gaps or insertions were combined. Tridacna is further divided into three subgenera, Tri- dacna, Persikima, and Chametrachea. Tridacna gigas be- longs to subgenus Tridacna; T. derasa to Persikima; and three species, T. squamosa, T. maxima, and T. crocea, belong to Chametrachea. This classical taxonomy also agrees well with the present tree topologies (Figs. 1-3). However, the present data are also in marked disagree- ment with the classical taxonomy: i.e., the tridacnid clams, genera Tridacna and Hippopus, are more closely related to the cardiids Vasticardium flavum and Fulvia mulica than these clams of Fragum and Corculum, sug- gesting either that the family Cardiidae is paraphyletic or that tridacnids belong to the family Cardiidae. The correct interpretation -either the traditional taxonomy of Tridac- nidae and Cardiidae or the present molecular view of these groups-is obscure. Molecular phylogenetic analy- ses using some other genes are underway and may answer this question. The present results indicate that the divergence between the Corculum-Fragum lineage and that of Vasticardium- Fulvia is deeper than that between Corculum-Fragum and Tridacna-Hippopus. Because there is no evidence of sym- biosis in clams of Vasticardium and Fulvia, clams in the tridacnid and Corculum-Fragum lineages might have ac- quired symbiotic relationships with zooxanthellae inde- pendently after their divergence from the lineage of Vasti- cardiiim-Fulvia. An alternative explanation is that the clam ancestral to the three lineages acquired the symbiotic relationship with zooxanthellae, and only the lineage of azooxanthellate Cardiidae lost this characteristic. If the latter explanation is correct, there may be some traces of symbiosis, not yet reported, in some of the azooxanthel- late Cardiidae. Further studies are necessary to distinguish between these two hypotheses. Masuda et al. (1994) reported that extracts of mantle homogenates from tridacnid clams (Tridacna derasa, T. maxima, T. crocea, and H. hippopus) are much more stimulatory to the excretion of photosynthate by T. derasa zooxanthellae than are tissue extracts of zooxanthellate Fragum clams, other azooxanthellate bivalves, or gastro- pods. These may reflect the relative phylogenetic distance between tridacnids and zooxanthellate cardiids, although the active substance in the mantle homogenate is still not known. Tridacna crocea larvae develop through several mor- phological stages: straight-hinge and prodissoconch. 76 T. MARUYAMA ET AL. Inserts in E10-1 region 247 269 1 : T-TG-TTTCAGCGTCCGCAGG-GCGG 1 : TCCGCTCTC-GGGT-GGC-GGCCCCG Inserts in E21-2 region 768 782 C. rardi aga . unedo H. hi pnopu.q 1 : TCCT-TACTAG-CTCCG 1 : -CC CTTGCCTCCG 1:-CC CTTGCCTCCG 1: -CCTGTA G-C-AAA H. porrpl laniig 1 : -CCTGTA G-C-AAA Inserts in 41 region 1484 1513 I 1 : GGGACCCAGG GACCGCTCT-CGAGCGGA 1 : GGGACCTAGGTCCGTCCGCTCTGCGGG-GGA 1 : GGGACCTAGGTCCGTCCGCTCTGCGGG-GGA Inserts in 47 region 1879 1910 I 1 : GTTG-GTTTCATCTCCTCGCGGGGTGTGCCTT 1 : GTTGCGCTCCATCTCCTCGCGGGGTGGGCCTT 1 : GTTGCGCTTCATCTCCTCGCGGGGTGGGCCTT 1 : GTTGCGCTTCATCTCCTCGCGGGGTGGGCCTT Figure 4. Inserts in I8S rDNAs of zooxanthellate tridacnids and cardiids, as well as in azooxanthellate cardiids, when aligned with 18S rDNA sequences ot Limicolaria kambeiil, Mytilus ethilis. Placo^ci ten magellanicus, Clilumys ixluinlifii. S/'ixii/a xolUlixxiimi, and 7Vr,v\ nut- itili. *, identical base sequence. Numbers at both ends of the sequence indicate the position of the base in the aligned sequences, which are available at the following URL: http://www.mbl.edu/html/BB/home.BB.htmJ cardiid, proto-tridacnid, and pre-hippopus, before reach- ing the final tridacnid stage (Kawaguti. 1983b). This sug- gests that the tridacnid clams evolved from the ancestral cardiid clam. The present phylogenetic topologies (Figs. 1-3) are consistent with this idea. Insertions in E10-1 in azooxanthellate cardiids are found in the variable region. The identities between these insertion sequences are relatively low, but they might appear after divergence from both of the zooxanthellate cardiids and tridacnids. In region 47, the insertions are restricted to tridacnid clams, except T. gigas. This inser- tion might appear after divergence between T. t,'M,'650 nm) light. The evoked depolarization consisted of a transient component, followed by a steady plateau in which the amplitude of the depolarization was well correlated with the log of the stimulus intensity. The evoked depolariza- tions induced action potentials in the photoreceptor cells, with the frequency of firing being well correlated with the stimulus intensity. The morphologies of individual photoreceptor cells were visualized by intracellular injec- tions of the fluorescent dye Lucifer yellow, and the path of the epistellar nerve across the stellate ganglion, into the pallial nerve, toward the brain was traced using the lipophilic dye Di-I. This pathway was confirmed physio- logically by recording light-evoked responses from the cut end of the pallial nerve. Introduction Most cephalopods have, in addition to their retinal pho- toreceptor system, extraocular photoreceptors or photo- Received 24 December 1997; accepted 21 April 1998. * To whom correspondence should be addressed. sensitive vesicles (PSVs). The PSVs of octopods such as Eledone moschata are located inside the mantle sac and appear as a small pigmented vesicle on the ventral poste- rior margin of the stellate ganglion (Bauer, 1909; Young, 1936, 1971; reviewed in Mauro, 1977; Fig. 1). These PSVs have also been termed the 'epistellar bodies' (Young, 1929, 1936). The ultrastructure of the epistellar body of E. moschata shows packed arrays of photorecep- tor cells with microvilli, reminiscent of rhabdomeres, but without dioptric apparatus (Nishioka et al.. 1962). to- gether with an epistellar nerve running into the stellate ganglion (Young. 1936; Perrelet and Mauro. 1972). The epistellar photoreceptor system in E. moschata has been shown to contain the visual pigment rhodopsin. with a maximum absorption wavelength of 475 nm; this is very close to 470 nm, which is the maximum absorption wave- length of retinal rhodopsin in this octopod species (Nishi- oka et til.. 1966; Mauro, 1977). This wavelength is also close to electrophysiological estimates of spectral sensi- tivity in E. moschata (Nishioka et al., 1966; Mauro, 1977). The accessory pigment retinochrome is also pres- ent in cephalopod PSVs (Hara and Hara. 1980; Ozaki et nl.. 1983). Although preliminary evidence has indicated that the photoreceptor cells within the epistellar body of octopus give depolarizing responses to light (Mauro and Bamnann, 1968; reviewed in Mauro. 1977), there has been no detailed study of these responses. Previously we have shown that extracellular, light-in- duced generator potentials in the epistellar bodies of Ele- done cirrhosa can be produced by light flashes of different intensity and duration, transmitted through the mantle sac wall of the octopus (Cobb et al., 1995a,b). In addition, the normal circadian behavioral rhythm entrained by a 78 OCTOPUS PHOTORECEPTOR ELECTROPHYSIOLOGY AND INNERVATION 79 Figure 1. Diagrams showing the position of the photosensitive vesi- cle or epistellar body in an octopus (not to scale). Top left: dorsal right view of octopus showing position of stellate ganglion (arrow) attached to the inner surface of the dorsal mantle sac wall. Bottom right: ventral view of left stellate ganglion in the same orientation as top left, with epistellar body (E), stellate ganglion (SG), stellar nerve (SN), and mantle connective or pallial nerve (MC) indicated. light-dark cycle is maintained after removal of the epistel- lar body in sighted octopus (Cobb el ai, 1995a,b). In the present study, we examine the capacity of the octopod extraocular photoreceptor system for light detection by characterizing the extracellular and intracellular responses to light flashes of increasing intensity and duration in the octopus Eledone cirrhosa. Lucifer yellow was used to stain iontophoretically and identify the photoreceptor cells in the octopus epistellar body; there was no evidence for dye coupling between these extraocular photoreceptor cells. It is also established that the epistellar body sends afferent nerves that pass across the stellate ganglion and connect with the central nervous system, via the mantle connective (pallial) nerve. Materials and Methods Collection and maintenance of experimental animals The lesser or northern octopus, Eledone cirrhosa (dor- sal mantle length 5-14 cm) used in this study were caught offshore near the coast of Plymouth, UK, at a depth of 10-15 m and transported to laboratory holding tanks with flow-through, aerated seawater at 12-18C. To prevent the octopus from escaping, the side walls of the tank were lined with 1-cm-thick sheets of plastic foam (Boyle, 1981). Octopus were supplied with live crab (Carcinus maenits) food ad libitum and were maintained in the tank system for up to 9 months until required. Electrophysiology For electrophysiological recordings, an octopus was anesthetized with 3% ethanol in seawater and decapitated. The paired stellate ganglia and attached epistellar bodies were then removed from the dorsal mantle wall. The epi- stellar body and attached stellate ganglion were washed in fresh seawater and pinned out, ventral side up. in a Sylgard-lined recording dish filled with filtered artificial seawater (ASW: NaCl 470 mM, KC1 lOmM, MgCl : 50 mM, CaCl : 10 mM, MOPS (3-[N-Morpholine]pro- pane-sulfonic acid) 10 mM, pH adjusted to 7.8, osmolarity 1010 mmol/kg). All subsequent electrophysiological re- cordings were made at room temperature (between 18 and 23C). Illumination stimuli were provided by aSchott KL1500 cold light source, with a quartz halogen bulb (Thorn EMI 15 V. 150 W). The light was passed through a standard heat filter into a glass fiber light guide and then, via an electronically controlled shutter and a second fiber light guide, to the ventral side of the stellate ganglion and epistellar body. The duration of the light flash was set by the electronic shutter (Uniblitz TI32. Optilas Ltd, UK) to between 10 ms and 10 s. The light intensity was varied by inserting neutral density filters into the light path and was measured at the preparation level using a portable calibrated radiometer (Haling Electro-optics, UK: model 27-5479). An additional, uncalibrated photocell was used to record the precise timing and duration of the stimulus light flashes. In some experiments, a red filter (Kodak 1A, UK, wavelength >650nm) was introduced into the light path. Suction electrodes were used to take extracellu- lar recordings from the mantle connective (pallial) nerve of the stellate ganglion in response to flashes of white or red light that were applied to the epistellar body of the preparation from two octopus. The extracellular receptor potentials were recorded by inserting a low-resistance microelectrode (2 M?2 resis- tance when filled with ASW) into the epistellar body wall. Intracellular recordings from photoreceptor cells in isolated intact epistellar bodies were made using high- resistance microelectrodes of borosilicate glass capillaries with inner filaments (Clark Electrochemical, UK, GC- 150F, 1.5 mm OD X 0.86 ID), filled with 3 M KC1 and having tip resistances of 30-150MS1 A conventional 80 C. S. COBB AND R. WILLIAMSON microelectrode amplifier (AxoClamp 2B amplifier, Axon Instruments, Inc. USA) was used for recording resting, venerator, and receptor potentials and for injecting current pulses through the intracellular microelectrode. In some experiments, octopus photoreceptor cells were injected iontophoretically with the fluorescent napthalimide dye Lucifer yellow CH (Stewart. 1978; Sigma, UK). For this the microelectrodes were back-filled with 3% Lucifer yel- low in 1 M LiCl and had resistances of 150-200 MO. Injections of hyperpolari/ing current ( 1.0 nA at 1 Hz) re- sulted in rapid movement of Lucifer yellow into the cells, and dye filling was considered complete after 30 min. Resting potential was recorded during cell impalement and generator potentials were recorded in response to lieht flashes, before switching to current injection in these experiments. Immediately after dye injection, the prepara- tion was photographed in whole mount under an epiflu- orescent microscope, using color film (400 ASA). For normal recordings, the signal from the microelec- trodes was amplified and, together with the signal from the photocell monitor, was passed to a computer-con- trolled signal averager (CED 1401 computer interface running Sigavg software, Cambridge Electronic Design, UK). Typically, 5-10 responses were averaged to im- prove signal-to-noise ratios, and repeated light flash stim- uli were separated by at least 20 s. The octopus prepara- tions remained viable for at least 6 h. Illumination of the experimental 'darkroom' was provided by red safelight (>650nm). Illumination of 0.1 ^W/cnr with this red light was found not to cause a decrement in photore- sponse. Epistellar body iniiervutio/i Innervation of the epistellar body was studied by ortho- dromic filling from the epistellar body using the lipophilic dye Di-I (Honig and Hume, 1989) in 10 animals. With the aid of a dissecting microscope, the ASW was removed from the dish and a crystal of the fluorescent carbocyanine lipophilic dye Di-I ( l,l'-dioctadecyl-3,3,3',3' tetramethyl- indocarbocyanine perchlorate, D-282 (Di-I C18(3)). Mo- lecular Probes, Inc. USA) was placed inside the epistellar body of each stellate ganglion through an incision made with a tine steel pin or razor blade. The tissues were then fixed in 2.5% paraformaldehyde in 0.1 M phosphate buffer. A microscope with epitluorescence and a rhoda- mine filter (Nikon DM580) was used to map the progress of the orange Di-I fluorescence from the epistellar body, across the stellate ganglion, through axons and into the pallial nerve over several days. Tissues were cleared in 60% (w/v) meglumine lothalamate with 0.01% (w/v) so- dium calciumedetate and 0.01% (w/v) sodium acid phos- phate (product: Conray 280. May & Baker Ltd (Rhone- nc Group). UK; Zill el cii. 1993), before the Di-I stained axons were photographed (400 ASA Kodak black- and-white film), under the epifluorescence microscope with rhodamine filter set. Additional white light illumina- tion was sometimes used to observe and photograph the relative positions of fluorescent Di-I-filled axons and the stellate ganglion tissue. Results Ancitonn- mid innervation of the octopus epistellar body Figure 2 shows the size, shape, and position of the epistellar body that contains the extraocular photorecep- tors on the stellate ganglion of Eledone cirrhosa. The epistellar body is located at the ventral posterior margin of the ganglion and is commonly spherical (Fig 2A). al- though it sometimes appears divided into two or more compartments (Fig 2B). It is confined within the common capsule of connective tissue surrounding the ganglion, but is separated from the ganglionic neuronal cell bodies by one or more sheath layers. The epistellar body is usu- ally orange in freshly dissected preparations, but the color may fade with time after dissection or be absent in some animals, particularly larger, older specimens. In the course of the intracellular recordings described below, some photoreceptor cells (/; = 4) were injected Ki(juri' 2. Epistellar body (arrows), stellate ganglion (SGl, and stel- lar nerves (SN) of the octopus Eledone cin-linxii. The epistellar body is commonly spherical (A), but sometimes appears to be divided into two compartments (B). Scale bars are 1 mm. OCTOPUS PHOTORECFPTOR ELECTROPHYSIOLOGY AND INNERVAT1ON 81 with Lucifer yellow, a fluorescent dye (Fig. 3). The dye fills showed that the photoreceptors have a cell soma (50 fj.m) that is located in the wall of the epistellar body and gives rise to a single axon, extending towards the periphery of the epistellar body. The axons from separate photoreceptor cells converge outside the epistellar body to form the 'epistellar' nerve. In addition, a process that may be as long as lOO/um projects towards the center of the epistellar body, sometimes dividing or branching toward its distal end. No dye coupling between photore- ceptors was observed in any of the cells that were well filled with dye. Figure 4A shows an epistellar body, the attached stel- late ganglion, and the connection of the mantle connective (pallial) nerve to the ganglion. For all cases (n = 8) in which the epistellar body and attached stellate ganglion preparations were filled with fluorescent Di-1 (Fig. 4B, C, D), orthodromic fills indicated that the epistellar body gave off an 'epistellar' nerve (Fig. 4B, C, D) that passed across the stellate ganglion, into the mantle connective (pallial) nerve (Fig. 4B, C), and then, presumably, to the brain. The nerve bundle was more shallow in its path across the stellate ganglion in some preparations (n = 2) than in others. In these preparations, the epistellar nerve remained intact and branched, sometimes into three sepa- rate fiber bundles (Fig. 4B), only before entry to the man- tle connective (pallial) nerve. In addition, these prepara- tions showed a single nerve branch, which occurred close SO^im Figure 3. lontophoretic microinjection of Lucifer yellow into an epistellar body photoreceptor cell. Detail of photoreceptor cell filled with Lucifer yellow showing cell soma (1) and axon process (2). Note long, branched process (arrows) that extends toward the center of ihc epistellar body. Figure 4. Innervation of the epistellar body. (A) Epistellar body and stellate ganglion of the octopus Eledone cirrhosa. The ovoid epistel- lar body (E) lies at the posterior end of the stellate ganglion near the origin of several stellar nerves (SN), opposite the mantle connective (pallial) nerve (MC). Scale bar is 1 mm. (B) Whole mount of epistellar body and stellate ganglion with orthodromic fluorescent Di-I-filled 'epi- stellar' nerve (arrow), epistellar body (E), stellar nerves (SN). and mantle connective (pallial) nerve (MC). Scale bar is 1 mm. (C) Whole mount of stellate ganglion with orthodromic fluorescent Di-I-filled epistellar body photoreceptor cell axons (arrows) connecting the epistellar body (E) and the mantle connective (pallial) nerve (MC). Scale bar is 1 mm. (D) Whole mount as in (C) under rhodamine-filtered light only, showing detail of Di-I-filled axons. Scale bar is I mm. to the epistellar body. This branch separated from the epistellar nerve and passed deep into the stellate ganglion, to an unidentified destination (Fig. 4B). However, in most preparations examined (n = 8), the Di-I-filled epistellar nerve branched close to the epistellar body, forming nu- merous nerve bundles that passed into the mantle connec- tive nerve and then toward the central nervous system (Fig. 4C. D). In addition to the epistellar nerve, these preparations showed two Di-I-filled nerves running from the side of the epistellar body in parallel to and on each side of the epistellar nerve (Fig. 4C. D). These parallel 82 C. S. COBB AND R WILLIAMSON nerve bundles then appeared to focus on the central path of the epistellar nerve bundles and pass into the center of the stellate ganglion (Fig. 4C, D). perhaps to the neuropil. The destination remains conjecture, however, because none of the preparations examined showed conclusive evidence of interaction between a Di-I-filled nerve fiber and the neuropil of the stellate ganglion. Recordings from the mantle connective (pallial) nerve To test whether light-evoked activity from the epistellar body could be observed in the mantle connective (pallial) nerve, suction electrode recordings were made from the cut end of the nerve as controlled light flashes were ap- plied to the epistellar body. When white light was flashed, a compound action potential was observed, mainly com- prising a large downward trough (Fig. 5). To test for artifacts and to determine the spectral sensitivity of the suction electrode response, a flash stimulus of red light was produced by using a >650 nm filter. Whereas a flash of white light evoked an extracellular voltage response in the mantle connective (pallial) nerve, a flash of red light, of the same duration, evoked no response (Fig. 5). The relatively long delay between stimulus and response does not permit any conclusions to be drawn about whether this is a direct or post-synaptic response; clearly the former would be expected. Electrophysiology of the epistellar body photoreceptors Generator potential responses induced by light flashes of constant duration but increasing intensity were re- corded extracellularly from the epistellar body photore- ceptors of the octopus, Eledone cirrhosa (Fig. 6A). The generator potential appeared as a short-latency downward deflection of the voltage trace, and the amplitude of the evoked response increased with increasing flash intensity. A graph of response against stimulus intensity (Fig. 6B) shows that the amplitude of the evoked response was well correlated to the log of light intensity across the entire range examined (Fig. 6B). The latency between the start Red Hash 400ms White flash Figure 5. Light-evoked extracellular responses in the mantle con- nective (pallial) nerve of the stellate ganglion preparation recorded by suction electrode in response to flashes of white and red light. Duration of the light Hash applied to the epistellar body is indicated in the bottom 250ms 50 uV B 150 -i 3- ioo - Si I 50H Flash intensity (u\V/cm ) I 300 1 200 ' ^ 100 - 100 " 1 50 I U 25 2 80 Flash intensity (uW/cm ) Figure 6. Epistellar body responses to change in intensity of the light stimulus. (A) Upper traces show generator potentials recorded extracellularly from photoreceptor cells in the epistellar body in response to a series of 250-ms-duration light flashes, as indicated in the lower trace, of increasing intensity. (B) Graph of log light-flash intensity against extracellular response amplitude. (C) Latency and rise time of the extracellular response with increasing stimulus intensity. of the stimulus and the start of the photoreceptor response was between 40 and 25 ms, and this latency decreased with increased stimulus intensity (Fig. 6C). Increasing stimulus intensity was also linked with a decrease in the time from the start of the stimulus to the peak of the response or rise time (Fig. 6C). The epistellar photorecep- tor cells needed some time to recover from each light flash: the extracellular response generated by a second flash given 2.5 s after the first was reduced by almost 80% (Fig. 7A). This decrease in response was evident up to 30 s after the first light flash. A plot of the response amplitude to the second flash (Fig. 7B) showed that the recovery had an exponential time curve with a time con- stant of almost 6 s. Similarly, the size of the extracellular response to a single light flash was affected by the inten- sity level of ambient illumination (Fig. 8). Figure 8A shows the extracellular responses to constant light flashes given after at least a I -min exposure to three different OCTOPUS PHOTORECEPTOR ELECTROPHYSIOLOGY AND INNERVATION 83 50 mV _n B 5. 100 i 80 - 60 - I I U 40 - 20 - T = 5.9s 10 15 20 25 30 Delay between flashes (s) 35 Figure 7. Adaptation or recovery responses between light flashes to the epistellar body. (A) Upper traces show examples of photoreceptor cell generator potentials recorded extracellularly from the octopus epi- stellar body in response to a series of 200-ms-duration light flashes with increased delay between flashes, but constant intensity (34 /vW/cnr), as indicated in the lower trace. (B) Amplitude of the extracellular response with increasing delay between flashes. intensity levels of 'background' illumination. It can be seen that the greater the level of background illumination, then the smaller the extracellular response to the light flash; Figure 8B shows that the relationship is more-or- less exponential. Intracellular recordings from individual photoreceptor cells demonstrated that, in the dark, these had membrane resting potentials of about 49 7 mV (mean SD, n = 43); such recordings could be maintained for up to 45 min. Epistellar body photoreceptors responded to a short flash of white light with a depolarization that often re- sulted in the firing of a burst of action potentials (Fig. 9). Although not studied in detail, this response was depen- dent on the wavelength of the light stimulus, for stimuli at wavelengths greater than 650 nm evoked no response (Fig. 9). This also provided confirmation that the response was not artifactual or related to other stimuli, such as the noise of the mechanical shutter. The amplitude of the intracellular receptor potential response varied with the intensity of the light flash (Fig. 10A). A graph of these responses (Fig. 10B), demonstrated that the amplitude of the depolarization was well correlated to the log of the light-flash intensity across the intensity range examined. Similarly, Figure 10B illustrates that the peak firing fre- 250ms SuV _n B v 1 I o. i 10- 20 Background illumination 40 Figure 8. Effect of background illumination on the response of the epistellar body to light-flash stimuli. (A) Upper traces show examples of photoreceptor cell generator potentials recorded extracellularly from the octopus epistellar body in response to a 200-ms-duration light flash (intensity 167 //W/cnr), as indicated in the lower trace, while subjecting the epistellar body to increased background illumination. (B) Amplitude of the extracellular response with increasing background illumination. In all cases each point represents the mean SEM, n = 10 separate receptor responses. quency of the resulting action potentials was well corre- lated to the log of the stimulus intensity and the conver- sion factor from receptor potential to action potentials or 'spikes' is about 2 spikes/mV depolarization; that value is within the range of other invertebrate spike encoding photoreceptor cells for example 1 spike/mV in eccen- Figure 9. Intracellular receptor potential's response to white and red light (>650 nm) stimuli recorded from the same photoreceptor cell in an epistellar body. The lower trace indicates the time and duration of 250-ms light flashes. Light-flash intensity was 167 ^iW/cnr and resting potential was 53 mV in both cases. 10 mV C. S. COBB AND R. WILLIAMSON Flash intensity (uW/cm) Figure 10. Typical intracellular receptor potentials recorded in a photoreceptor cell from an epistellar body, in response to change in light- Hash stimulus intensity. Resting potential was \2 mV to 47 mV. (A) The upper traces show the photoreceptor potentials in response to the series of 250-ms-duration light flashes, as indicated in the lower trace in response to increased intensity. (B) Graph of light-flash intensity against response receptor potential amplitude and peak firing frequency. (C) Rise time and latency of receptor potentials against increased light stimulus intensity. In all cases each point represents the mean SEM. n = 10 separate receptor potential responses to light-flash stimuli from a single photoreceptor cell. In some cases the SEM is too small to be visible. slowly decaying back to the normal resting potential for the photoreceptor cell (Fig. 11); this probably indicates that the action potentials do not actively invade the cell soma and are thus attenuated or absent in some soma recordings. Discussion A number of morphological and biochemical studies have shown that octopod and decapod cephalopods have extraocular photoreceptors (reviewed in Messenger, 1991). Young (1936) first described epistellar structures in several octopod genera: from histological considera- tions, he hypothesized that these might have a neurosecre- tory function. The epistellar body of Octopus vulgaris was later re-investigated in detail using electron micros- copy and found to contain cells that have microvilli very like those found in the rhabdomeres of the extraocular photoreceptors seen in other molluscs and arthropods (Nishioka et al.. 1962). This work was later extended to include the epistellar bodies of Eledone moschata (Nishi- oka et al., 1966); in it. electron microscopy gave further evidence for a rhabdomeric ultrastructure, and biochemi- cal evidence indicated the presence of the photopigment rhodopsin. These studies thus implied that the neuronlike cells of the epistellar body were not neurosecretory. but photoreceptor cells. This agreed with comparative studies showing the presence of extraocular photosensitive cells in the central nervous systems of many other inverte- brates for example, in the caudal ganglion of the cray- fish (Kennedy, 1963) and the molluscs Aplysiu californica (Arvanitaki and Chalazonitis, 1961) and Onchidiwn ver- ruculatiim (Hisano et til.. 1972). Preliminary electrophysi- ological evidence demonstrating a light-evoked response from cephalopod epistellar body photoreceptor cells was first obtained by Mauro and Baumann ( 1968) in an octo- pus, Eledone moschatu. Decapod cephalopods such as the trie cells from Linuilns polypheniiis (Behrens and Wulff. 1965). As already seen for the extracellular compound generator potentials, the time-to-peak depolarization for individual photoreceptor cells decreased with increasing stimulus intensity (Fig. IOC). However, stimulus-to-re- sponse latency for individual photoreceptor cells was fairly constant at around 25 ms for flashes of intensity greater than 5 /jW/cm 2 , but longer for very weak flashes (Fig. IOC). The effect of long-duration (5 s and l()s) but constant-intensity stimuli on the intracellular photore- sponse was examined. There was an increase in the dura- tion of the receptor potential as the light stimulus was lengthened from 5 s to 10s, with the long-lasting plateau maintained for the duration of the stimulus and then 5s SmV Figure 11. Intracellular receptor potentials recorded in a single pho- toreceptor cell from an epistellar body in response to constant intensity (77 ^/W/cnr) and increased long-duration light stimuli. The receptor potentials recorded from a 5-s and 10-s light flash are shown as indicated by the stimulus marker in the bottom trace. Resting potential was -50 mV. OCTOPUS PHOTORECEPTOR ELECTROPHYSIOLOGY AND INNERVATION 85 squid Loligo vulgaris and the cuttlefish Sepia officinalis also have PS Vs. or 'parolfactory vesicles': these are not located on the stellate ganglia, but lie underneath the cranial cartilage casing on and below the optic tract (Young, 1936; Perrelet and Mauro, 1972). These PSVs also have neuronlike cells with a rhabdomeric ultrastruc- ture containing rhodopsin (Nishioka e! al., 1966; Bau- mann ct til.. 1970) and their responses to light stimuli have been briefly reported for Todarodes sagittatus (Mauro and Sten-Knudsen, 1972; Mauro. 1977). Loligo pealei and Loligo forbesi (Sperling et al., 1973; Cobb and William- son. 1998). Electrophysiology of extraocular photoreceptor cells Using extracellular and intracellular recordings from extraocular photoreceptors in Eledone cirrhosa, the pres- ent study extends our understanding of the electrophysiol- ogy of epistellar body photoreceptors in cephalopods. The results show that the photoreceptors have cell resting po- tentials of about -49 7 mV and respond to a flash of light with a depolarization consisting of a transient component, often accompanied by a burst of action poten- tials, followed by a steady-state or plateau depolarization. The amplitude of the evoked depolarization and the peak firing frequency of the cell were directly correlated with the intensity of the light flash. No 'quantum bumps' were observed in these recordings, although these have been reported in recordings from the extraocular photorecep- tors of Eledone moschata (Mauro and Baumann. 1968). as well as from other invertebrates (Lisman and Brown, 1975). The extracellular generator potentials recorded from the epistellar body of Eledone cirrhosti showed adapta- tional changes when the level of ambient illumination was increased in this study. This agrees well with the adaptational changes seen in retinal photoreceptors for cephalopods (Weeks and Duncan, 1974) and other inver- tebrates (Laughlin. 1989). It should be noted that there are no reports of screening pigments in cephalopod extra- retinal photoreceptors, so the mechanism of screening pigment migration cannot be invoked here. The observed decrease in the stimulus-to-response latency of the evoked intracellular depolarization (from about 40 ms to 25 ms) with increasing intensity is similar to that seen in photore- ceptor latency measurements from the epistellar bodies of both Eledone moschata and Eledone cirrhosti (Mauro, 1977; Cobb et al., 1995b) and also from the retina of Sepiola atlantica, a sepiolid cephalopod (Duncan and Weeks, 1973). The small differences between the latency values reported in this study (40 ms) and in the others (30 and 23 ms, respectively) are probably due to the different stimulus intensities used, for as shown above, the onset delay decreases with stimulus intensity. When two extraocular photoreceptor cells from the same cluster in a parolfactory vesicle from a squid (Loligo pealei) were impaled simultaneously, partial electrical coupling was recorded in 1 out of 4 pairs of cells (Sperling et al.. 1973). In the present study, which used extraocular photoreceptor cells from the epistellar body of an octopus (Eledone cirrhosa), no evidence of Lucifer yellow dye coupling was observed. The lack of dye coupling may perhaps indicate a lack of electrical coupling, for the two are often well correlated, as, for example, in the horizontal cells of the turtle retina (Stewart, 1978) and in the pineal gland photoreceptors of teleost fish, where numerous gap junctions are present (Omura, 1984), and Lucifer yellow dye coupling has been observed (Nakamura et al.. 1986). However, electrical coupling is not always associated with dye coupling, particularly in molluscan preparations (Williamson. 1989; Ewadinger et al., 1994). Light-flash stimuli transmitted through the red filter (Kodak 1 A, wavelength >650 nm) neither stimulated the PSVs from Eledone cirrhosa nor evoked an afferent re- sponse in the mantle connective nerve; therefore, this wavelength of red light O650 nm), at the intensities used in this study, did not stimulate the photoreceptors. This outcome was perhaps to be expected, for the photorecep- tors in both the retinal and extra-retinal systems of Ele- done moschata have a rhodopsin absorption maximum around 470 nm (Nishioka et al.. 1966; Hamdorf et al., 1968), and the red light stimuli are therefore likely to be outside the detection range of the photoreceptors, at the stimulus intensities employed here. Innen'ation of the epistellar body Orthodromic fills of the photoreceptor axons with fluo- rescent Di-I from the epistellar body of Eledone cirrhosa have shown nerve bundles passing from the epistellar body into the mantle connective (pallial) nerve. Microana- tomical studies, silver-staining techniques, and electron microscopy were used in studies of the octopuses Eledone moschata and Octopus vulgaris to trace the 'epistellar nerve' from the photoreceptor cells of the epistellar body, through the stellate ganglion, to the mantle connective (pallial) nerve (Young, 1936; Cazal and Bogoraze, 1944; Nishioka et al., 1966; Perrelet and Mauro, 1972). In addi- tion, degeneration of nerve fibers running to the epistellar body was shown in E. moschata after the mantle connec- tive nerve was cut (Young, 1936). The mantle connective (pallial) nerve is known to contain about 16,000 nerve fibers (Young, 1965), with two groups of efferent, and possibly two groups of afferent fibers (Young, 1971). In E. moschata, the epistellar nerve contains about 1500 afferent nerves (Perrelet and Mauro, 1972). Therefore, the selective nerve tracing in this study (using Di-I and electrophysiological recordings from the mantle connec- 86 C. S. COBB AND R. WILLIAMSON live (pallial) nerve) presents new evidence supporting the view that the epistellar nerve contains the afferent axons of extraocular photoreceptor cells, and that these axons conduct photoreceptive information to the central nervous system in the octopus. Whether such sensory information passes to the palliovisceral lobe of the octopus brain from nerves earned in the mantle connective (pallial) nerve remains to be determined. In summary, this study provides further evidence for the presence of an active extraocular photoreceptor sys- tem in adult cephalopod molluscs. However, the func- tional role of the epistellar body and parolfactory vesicles in octopus and squid still remains enigmatic. The lack of a structurally organized retina or optical apparatus such as a lens, and the relatively deep location of all these PSVs argues against any function in visual image forma- tion. Rather, it is generally assumed that in both octopods and decapods the extraocular photoreceptors play a role in monitoring ambient light levels over lengthy periods; that is, they function as a photometer (Mauro. 1977; Houck, 1977a.b), perhaps connected with seasonal repro- ductive activity (Baumann et ui. 1970), control of circa- dian activity rhythms in octopods (Houck, 1977ab, 1981, 1982), diel vertical migration in squid (Palmer and O'Dor. 1978). or even the feedback control of ventral photo- phores in midwater squid such as Abralia trigoniiru (Young, 1972, 1973; Young et . Eye. Res. 17: 183-192. Kwadinger, N. W., N. I. Syed, K. Lukowiak, and A. G. M. Bulloch. 1994. Differential tracer coupling between pairs of identified neu- rones of the mollusc Lvmnaea stagnalis. J. E.\p. Biol. 192: 291 297. Hamdorf, K., J. Schwemer, and U. Taiiber. 1968. Der Sehfarbstoff, die Absorption der Rezeptoren und die spektrale Empfindlichkeit der Retina von Eledone moschata. Z. Vergl. Physio/. 60: 375 H5. Hara, T., and R. Hara. 1980. Retinochrome and rhodopsin in the extraocular photoreceptor of the squid. Todarodes. J. Gen. Physiol. 75: 1-19. Hisano, N., H. Tateda, and M. Kmvabara. 1972. Photosensitive neurons in the marine pulmonate mollusc, Onchidium verruculatum. J. E.\p. Biol. 57: 651-660. Honig, M. G., and R. L. Hume. 1989. Oil and DiO: versatile fluo- rescent dyes of neuronal labelling and pathway tracing. Trends Neu- rosci. 12: 333-341. Houck, B. A. 1977a. A morphological and behavioral study of an extra-ocular photoreceptor in octopods. Ph.D. Thesis, University of Honolulu, Hawaii. Houck, B. A. 1977b. Photoreception in octopods: The role of ocular and extra-ocular photoreception in maintenance of locomotor activ- ity rhythms [Abstract], Am. Zool. 17: 969. Houck, B. A. 1981. Locomotor activity and home selection in three species of Hawaiian octopods [Abstract]. Am. Zool. 21: 967. Houck, B. A. 1982. Temporal spacing in the activity patterns of three Hawaiian shallow-water octopods. Nautilus 96: 152-156. Kennedy, D. 1963. Physiology of photoreceptor neurons in the ab- dominal nerve cord of the crayfish. 7. Gen. Phyxinl. 46: 551-572. Laughlin. S. B. 1989. The roles of sensory adaptation in the retina. J. E.\p. Binl. 146: 39-62. I isman, J. E., and J. B. Brown. 1975. Light-induced changes of sen- sitivity in Limulus ventral photoreceptors. 7. Gen. Physiol. 66: 473- 488. Mauro, A. 1977. Extra-ocular photoreceptors in cephalopods. Symp. Zool. Soc. Ltiiul. 38: 287-308. Mauro, A., and F. Baumann. 1968. Electrophysiological evidence of photoreceptors in the epistellar body of Eledone moschata. Nature 220: 1332-1334. Mauro, A., and (). Sten-Knudsen. 1972. Light-evoked impulses OCTOPUS PHOTORECEPTOR ELECTROPHYSIOLOGY AND INNERVATION 87 from extra-ocular photoreceptors in the squid Todarodes. Nature 237: 342-343. Messenger, J. B. 1991. Photoreception and vision in molluscs. Pp. 364-397 in Evolution of the Eye and Visual System. J. R. Cronly- Dillon and R. L. Gregory, eds. Macmillan, London. Nakamura, T., G. Thiele, and H. Meissl, H. 1986. Intnicellular re- sponses from the photosensitive pineal organ of the teleost. Phoxintis phoxinux. J. Comp. Physio/. A. 159: 325-330. Nishioka, R. S., I. Yasumasu, and H. A. Bern. 1962. Ultrastructure of the epistellar body of the octopus. Z. Zellforsch. Mikrosk. Anat. 57: 406-421. Nishioka, R. S., I. Yasumasu, A. Packard, H. A. Bern, and J. Z. Y'oung. 1966. Nature of vesicles associated with nervous system of cephalopods. Z. Zellforsch. Mikrosk. Anat. 75: 301-316. Omura, Y. 1984. Pattern of synaptic connections in the pineal organ of the ayu. Plecog/ossus altivelis (Teleostei). Cell Tissue Res. 236: 611-617. Ozaki, K., R. Hara, and T. Mara. 1983. Histochemical localization of retinochrome and rhodopsin studied by fluorescence microscopy. Cell Tissue Res. 233: 335-345. Palmer, B. W., and R. K. O'Dor. 1978. Changes in vertical migration patterns of captive lllex illecebrosus in varying light regimes and salinity gradients. Tech. Rep. Fish. Mar. Sci. Canada 833: 23.1- 23.12. Perrelet, A., and A. Mauro. 1972. Ultrastructure of nerves associated with the epistellar body of the octopod Eledone moschata and the parolfactory vesicles of the squid Todarodes sagittatus. Brain Res. 37: 161-171. Sperling, L., J. E. Lisman, and A. Godfrey. 1973. Light-evoked responses from the ventral parolfactory vesicles of Lot/go pealei [Abstract]. Biol. Bull. 145: 456. Stewart, W. W. 1978. Functional connections between cells as re- vealed by dye-coupling with highly fluorescent naphtalimide tracer. Cell 14: 741-754. Sundermann, G. 1990. Development and hatching state of ectodermal vesicle organs in the head of Sepia officinalis, Loligo vulgaris and Loligo forbesi (Cephalopoda. Decabrachia). Zoomorphologv 1(19: 343-352. Weeks, F. I., and G. Duncan. 1974. Photoreception by a cephalopod retina: response dynamics. Exp. Eye Res. 19: 493-509. Williamson, R. 1989. Electrical coupling between secondary hair cells in the statocyst of the squid Alloteuthis sitbiiluta. Brain Res. 486: 67-72. Young, J. Z. 1929. Sopra un nuovo organo dei cefalopodi. Boll. Soc. Iinl. Biol. Sper. 4: 1022-1024. Y'oung, J. Z. 1936. The giant nerve fibres and epistellar body of ceph- alopods. Q. Jl. Microsc. Sci. 78: 367-386. Y'oung, J. Z. 1965. The diameters of fibres of the peripheral nerves of Octopus. Proc. R. Soc. Land. B. 162: 47-79. Young, J. Z. 1971. The Anatomy of the Nen'ous system of Octopus vulgaris. Clarendon Press. Cambridge. Young, R. E. 1972. Function of extra-ocular photoreceptors in bathy- pelagic cephalopods. Deep-Sea Res. 19: 651-660. Young, R. E. 1973. Information feedback from photoreceptors and ventral countershading in mid-water squid. Pac. Sci. 27: 1-7. Young, R. E., C. F. E. Roper, and J. F. Walters. 1979. Eyes and extraocular photoreceptors in midwater cephalopods and fishes: Their roles in detecting downwelling light for counterillumination. Mar. Biol. 51: 371-380. Zill, S. N., S. F. Frazer, D. L. Maclarland, and S. E. Fish. 1993. Characterization of insect sense organs and optical clearing of whole- mount preparations using Dil in fixed tissues. J. Exp. Biol. 175: 299-303. Marine Biological Laboratory Woods Hole Massachusetts One Hundreth Report for the Year 1997 One-Hundred and Ninth Year Officers of the Corporation Sheldon J. Segal, Chairman of the Board of Trustees Frederick Bay, Co-Vice Chair Mary J. Greer, Co-Vice Chair James D. Ebert, President of the Corporation John E. Burris, Director and Chief Executive Officer Mary B. Conrad, Treasurer Neil Jacobs, Clerk of the Corporation Contents Report of the Director and CEO Rl Report of the Treasurer R7 Financial Statements R8 Report of the Library Director R19 Educational Programs Summer Courses R22 Short Courses R26 Other Programs R31 Summer Research Programs Principal Investigators R33 Other Research Personnel R34 Library Readers R36 Institutions Represented R37 Year-Round Research Programs R42 Honors R53 Board of Trustees and Committees R59 Administrative Support Staff R62 Members of the Corporation Life Members R65 Members R66 Associate Members R76 Certificate of Organization R79 vi tides of Amendment R79 .Bylaws . R79 Photo credits: W. Amos R19 Beth Armstrong R2 (top), R34, R37, R43 Ken Foreman R5 (bottom), R31 Linda Colder R59, R65 Roger Hanlon R3 (top), R62 Richard Howard R4 (top), R5 (top), R22, R26, R30, R78 Tom Kleindinst R45 Alan Kuzirian R53 Beth Liles, R49 George Lower R4 (bottom), R7 Chris Neill, R42 Bruce Peterson R47 Janice Reed Rl Report of the Director and Chief Executive Officer The Marine Biological Laboratory had another exciting and successful year in 1997. We set our sights high and achieved a number of major goals for the Laboratory. We launched the first Semester in Environmental Sciences program for undergraduates from liberal arts colleges; recruited new investigators for our resident research programs; planned for and established new advanced summer courses; and found new and expanded ways to capitalize on the flexibility and sophistication of the Marine Resources Center. These endeavors were supported by the tangible and intangible resources and energies generated by the laboratory's new fundraising campaign. Discovery: The Campaign for Science at the Marine Biological Laboratorv, which was launched with a public celebration and street fair on August 8. 1997. In my report last year. I discussed the concerted planning process that set the stage for this campaign. By focusing our research and educational directions and priorities, we were able to announce last August a campaign goal of $25 million to be raised by December 31, 2000. I am pleased to say that, as of the spring of 1998, we have already raised $14.5 million towards that goal. I have every reason to believe that we will meet, if not exceed our goal within the timeframe of the campaign. Research at the MBL We made great strides in the MBL's resident research centers and programs during 1997. The avenues for independent research broadened and diversified in each of MBL's programmatic areas, while our investigators found new ways to work together in various interdisciplinary pursuits. The Josephine Bay Paul Center Dr. Mitchell Sogin and his colleagues in the Josephine Bay Paul Center for Comparative Molecular Biology and Evolution continued their important studies on the evolution of eukaryotes, most recently focusing on those found in extreme environments. An exciting development is a major grant from the National Institutes of Health to sequence the genome of Giurdia lamblia. a water-borne human pathogen. Giardiasis is a major contributor to the enormous burden of human diarrhea! diseases, which are second only to respiratory infections as causes of mortality and morbidity worldwide. While Giardia can be a deadly organism, it is manageable in terms of genome analysis because it has a relatively modest genome size of 12 million base pairs distributed onto five chromosomes. Dr. Michael Cummings was recruited to the Bay Paul Center in 1997. He comes to the Center to study Mycobacteria as a means of understanding more clearly the evolution of pathogenicity and how various strains of tuberculosis develop drug resistance. Another member of the Bay Paul Center, Dr. Monica Riley, continues to travel world-wide sharing her expertise on the E. coli genome. She was a co-author in September on a paper published in the journal Science which reports the successful sequencing of the genome of the K-12 strain of the bacterium. In 1997, Center scientist Dr. Neal Cornell and his co-workers conducted a molecular phylogenetic analysis for 5-aminolevulinate synthase, the first enzyme of heme biosynthesis. The analysis supports the suggestion that alpha-purple bacteria are the closest contemporary eubacterial relatives of eukaryotic mitochondria. The study also indicates that the massive gene duplication required for the evolution of true vertebrates had occurred when non-vertebrate jawless fish appeared, 400 to 500 million years ago. The Ecosystems Center The MBL's Ecosystems Center, co-directed by Drs. John Hobbie and Jerry Melillo, continues its leadership in environmental research on local and global scales. Rl R2 Annual Report For more than a decade, MBL scientists have led a Long-Term Ecological Research (LTER) project on the arctic tundra around Toolik Lake in Alaska. In 1997. scientists at the Center secured another LTER project grant of nearly $3.4 million over six years from the National Science Foundation to study the Plum Island Sound system in northeastern Massachusetts. An LTER grant is important scientifically and institutionally because it allows long-term data collection and analysis in the same location and provides a solid base of funding with which to recruit scientists and research technicians for the project. We are proud of the Ecosystems Center team because they were the only group that was funded out of 2 1 proposal submissions nationwide. In the southern hemisphere, our scientists have also been compiling data for more than a decade on the effects of deforestation on soil fertility and trace gas emissions in the rainforests in Brazil. Their work continues as they evaluate what effects changing and more aggressive agricultural practices are having on the Amazon Basin and the world. In 1997. the MBL received a challenge grant of $1 million from The Clowes Fund. Inc.. of Indiana, to provide support for new and expanded facilities for The Ecosystems Center. Thanks to this grant, the MBL is now in the early stages of planning renovations and expansion that will nearly double the Center's existing laboratory. Held staging, and office space. The plans also call for the creation of a new computer teaching laboratory in the Center. The Marine Resources Center Dr. Roger Hanlon has advanced his research agenda in the Marine Resources Center (MRC) with the introduction of the Program in Sensory Biology and Neuroethology and the Program in Scientific Aquaculture. An MRC Advisory Committee has been constituted to review research directions and scientist recruitment strategies. Members of the Committee include Drs. Gerald Fischbach, Harvard Medical School; Irwin Levitan, Brandeis University; and Vilayanur Ramachandran, University of California, San Diego. In the program on neuroethology. Dr. Hanlon and his colleagues study vision, balance, chemical sensing, and signaling to gain a better understanding of the output of neural activity, i.e., the resultant behavior of an organism that has evolved through natural selection. Towards that end. Dr. Hanlon has recently been awarded research grants that will enable molecular and cellular studies on squid and cuttlefish. These marine organisms possess very large neurons and have evolved dramatic and rapidly changing skin patterning displays as evidence of aggressive and reproductive behaviors. Last summer, Hanlon and his colleagues performed a series of experiments in the MRC in which they discovered that squids are able to detect transparent prey at greater distances thanks to a specialized way of seeing known as polarization vision. The result is enhanced predution and feeding by the squid. Hanlon' s findings appeared in the journal Nature in May 1998. The second MRC initiative, the Program in Scientific Aquaculture. seeks to select organisms with defined genetic lines or at precise developmental stages for use as biomedical models, to attract industry and Report of the Director and CEO R3 commercial ventures to develop and test products in the MRC. and to study molecular and cellular mechanisms of disease processes and prevention. Architectural Dvnamics in Living Cells Program MBL scientists continue as leaders in the design and use of advanced imaging systems to study the molecular definition and physical forces that affect the structure and function in living cells. Dr. Shinya Inoue, MBL Distinguished Scientist, received additional honors in 1997 for his contributions to light microscopy and was awarded the Ernst Abbe Award for Microscopy by the New York Microscopical Society. Inoue and his colleagues announced at the 1997 meeting of the American Society of Cell Biology the development of the Centrifuge Polarizing Microscope. This new microscope allows researchers to view architectural changes in cells at the subcellular level when exposed to high gravitational forces. This technical feat was the fruit of collaboration between scientists at the MBL, Olympus Optical Co., and Hamamatsu Photonics Corp. Dr. Rudolf Oldenbourg and his associate Kaoru Kato, together with Dr. Peter Smith and Katherine Hammar of the BioCurrents Research Center, reported in 1997 on their observations of the dynamic behavior of actin molecules at the tip of growing nerve cells. These observations were made possible by the exceptionally high sensitivity and sharp image quality of Oldenbourg' s Pol-Scope, which was developed in collaboration with Cambridge Research and Instrumentation, Inc. Laboratoty for Reproductive Medicine Dr. David Keefe, of the Laboratory for Reproductive Medicine of Brown University and the Women and Infants Hospital in Providence, R.I., established a resident laboratory at the MBL in 1997. Here he investigates the mechanisms underlying age-related female infertility. Keefe uses technologies developed at the MBL to seek out clues that indicate whether an egg has been damaged with age and is therefore not viable. With the ion selective probe, developed in Peter Smith's BioCurrents Research Center, Keefe can detect the flow of certain ions in and out of the cell. This flow may reflect an egg's potential to develop successfully. He is also using Oldenbourg's Pol-Scope to study the egg's spindle; flaws in the spindle may be responsible for the faulty transfer of genetic material during cell division. Laboratory of Aquatic Animal Health and Pathology In 1997, scientists from the University of Pennsylvania's Laboratory of Aquatic Animal Health and Pathology, which is based at the MBL, continued their efforts to identify and treat diseases affecting marine organisms. They identified in toadfish an abcess of the abdominal organs caused by Pseudomonas sp. bacteria and a bacterial pericarditis disease caused by Edwardsiella sp. Dr. Roxanna Smolowitz continued her effort to culture QPX and describe the pathogenesis of the disease in susceptible strains of clams, while Dr. Robert Bullis and colleagues worked on ways to detect illegal chemical scrubbing of eggs in female lobsters. Boston University Marine Program (BUMP) BUMP Postdoctoral Fellow Frank Grasso organized and co-hosted an international workshop on "Plume Tracing" last year. Twenty-five scientists presented and debated the state-of-the-art in our knowledge of animals R4 Annual Report following odor plumes to their source of release. Lobsters are animals that locate odor sources (for example, baited traps) under water without visual or auditory signals. Scientists in Dr. Jelle Atema's laboratory have developed a tool known as Robo-Lobster to perform the plume tracing task and to learn about the sensory input and signal processing used by lobsters. Summer Research The MBL's laboratories were again tilled with scientists from all over the world during the summer of 1997. We welcomed back investigators with whom we have worked for a number of seasons and were able to share in the enthusiasm of scientists working here for the first time, either as fellows or as new principal investigators. More than 400 scientists came to the MBL last summer to do their research, including 23 research fellows who received more than $171,000 in fellowship awards. The summer research season is framed by the Poster Session in June and is capped off by the General Scientific Meetings in August. These events attest to the breadth and diversity of the multidisciplinary work at the MBL where presentations range from papers in ecology and population biology to cellular and molecular biology to vision and biophysics. One of the highlights of the summer research season was Dr. Stephen Highstein's (Washington University) continuing work on the vestibular system of toadfish. The toadfish is an excellent marine model for studying how changes in pressure affect balance and equilibrium, because its vestibular system is similar to humans. Such studies teach us about motion sickness and dizziness, which are often signs of vestibular organ dysfunction. Highstein and his colleagues are focusing on the problems of nausea and dizziness experienced by astronauts when they get out beyond the earth's gravity, a malady known as "space adaptation syndrome." As I write, four toadfish are passengers aboard the space shuttle Columbia, participants in an experiment designed to help scientists understand what happens in the vestibular system during an extended period in space. Another area of research that is becoming increasingly important at the MBL is evolution and development or "evo-devo." Researchers from all over the world come to the MBL to learn more about how the dividing cells in an embryo decide which physiological role they will assume as the organism grows. For example, Drs. Mark Q. Martindale (University of Chicago), Jonathan Q. Henry (University of Illinois), and Barbara Boyer (Union College) take advantage of the wide variety of marine organisms available at the MBL to compare the developmental patterns of different kinds of embryos. They are studying relatively primitive animals like ctenophores whose development is simple enough that scientists can trace how a change in one cell or set of cells affects the whole organism as it grows. Ultimately, they hope to understand more clearly when and how organisms have diverged from their evolutionary ancestors. The MBL's General Scientific Meetings are held annually in mid-August. This three-day meeting is an opportunity for summer and resident scientists to report on their most recent research results. One highlight from the 1997 meeting is the finding from the laboratory of Robert Barlow (SUNY Health Science Center) that the lateral eyes of two species of horseshoe crabs the Woods Hole species Liiuuhis polvpliennis and its Japanese counterpart Tachypleus triilcnuitus are similar, but not alike. Both show the same nighttime circadian rhythms, but Barlow and co- workers found that the mechanism responsible for those rhythms differs between the two. The eye of Limn/its is larger and contains more ommatidia than that of Tuchvpleiis. As a result, the Limulus lateral eye Report of the Director and CEO R5 becomes more sensitive at night by capturing more photons. The smaller lateral eye of Tachypleus is nearly as sensitive, but due to its size is not able to catch as many photons as its Woods Hole relative. Instead, its photoreceptors make up for this deficiency by increasing gain, or the amount of information absorbed by the photoreceptors. Education at the MBL Summer Courses The MBL's educational program in biology and biomedicine continues to be preeminent, with almost 400 students and 400 faculty coming to the MBL each summer for a superb hands-on scientific experience. The MBL now offers 20 courses each summer, three of which were developed in 1997. One of these, Molecular Mycology: Current Approaches to Fungal Pathogenesis, was taught for the first time during the summer of 1997. The others. Frontiers in Reproduction: Molecular and Cellular Concepts anil Applications, and Neural Development and Genetics of Zebrafish, will be offered for the first time during the summer of 1998. Competition for admission to these and all MBL courses is intense. The 368 students enrolled during the summer of 1997 were selected from 833 applications that were culled from 1521 inquiries. Students continue to come to MBL courses from throughout the world, with about 35% to 40% of them from foreign countries. The popularity of the courses is one indication of their excellence. Another is the peer review process that suiTounds the awarding of many of the funds that support the courses. For the first time in my and many of my colleagues' recollections, one of our summer courses, Neural Systems and Behavior, received a perfect score of 100 in its N1H review; Neurobiology also received near perfect marks from the NIH. This attests to the vision of the course directors and faculty and the quality of the curriculum that is offered at the MBL. The generous support of a number of foundations, as well as the Federal government and a large cadre of individual donors is crucial to our ability to offer the best courses to the best students. Admission to many of the courses continues to be on a need-blind basis because of the scholarships we provide from these gifts and grants and from our endowment income. Five course directors retired in 1997, including David Kleinfeld, University of California, San Diego, and David Tank, Lucent Technologies (Methods in Computational Neuroscience); Steve Hajduk, University of Alabama, Birmingham (Biology of Parasitism); Mark Mooseker, Yale University (Physiology); and Pierre Drapeau. Montreal General Hospital (Neurobiology and Development of the Leech). They will be replaced by William Bialek and Rob de Ruyter, NEC Research Institute (Methods in Computational Neuroscience); Ed Pearce, Cornell University (Biology of Parasitism); Kerry Bloom, UNC, Chapel Hill (Physiology); and Christie Sahley, Purdue University (Neurobiology and Development of the Leech). I thank all retiring course directors for their efforts on behalf of the MBL, and welcome our newest faculty members to the Laboratory. Semester in Environmental Sciences As I mentioned above, we are buoyed by the success of our first semester-long program in environmental science for undergraduates from liberal arts colleges. We welcomed 1 6 students for the inaugural semester which was held in the fall of 1997. Twenty-four liberal arts colleges and universities now participate in the consortium of institutions from which these students are drawn. The Ecosystems Center scientists who served as R6 Annual Report faculty were impressed by the dedication and intellect of the group. I attended the student presentations before a large audience at semester-end and was impressed by the range and creativity of the research projects that they undertook. The semester proved a satisfying experience for all participants in this innovative new educational program, and it attests to the flexibility and commitment of the MBL in designing new and creative ways to further education in the life sciences. MBL/WHOI Library The Library is constantly responding to the challenges of straddling the traditional and the electronic modes of information transfer. This requires both technical achievement and renewal as well as the ongoing commitment to provide access to the best information essential to scientific research and teaching in the diverse Woods Hole scientific community. This charge translates into maintaining the unique, internationally acclaimed archival collection while, at the same time, being responsive to the fast-paced, complex, and demanding forces shaping electronic information delivery. So adept is the MBL/WHOI Library at this synthesis, that its director, Catherine Norton, is serving as President of the Boston Library Consortium, a group of academic, research, and hospital libraries. Facilities Our physical plant received significant attention during 1997. Major grants from the National Science Foundation and Colonial Gas Company provided nearly $600,000 to renovate classrooms and laboratories in the Loeb building. Additional monies were spent to reconfigure and modernize research laboratories to accommodate newly appointed MBL scientists and expanded research needs. MBL Trustees At their November 1997 meeting, the Board of Trustees of the Marine Biological Laboratory elected Dr. John E. Dowling to serve as President of the MBL Corporation. Dowling succeeds Dr. James D. Ebert, who retired after serving as President for seven years. In recognition of his "outstanding service to the development of American science in general and his long-standing dedication to the Marine Biological Laboratory," the Board has designated Ebert an Honorary Member of the Board of Trustees. Dr. Dowling is the Maria Moors Cabot Professor of Natural Science at Harvard University, a long-time MBL summer investigator, and former MBL Trustee. His tenure as President of the MBL Corporation began with the March 1998 meeting of the Board. Two New Trustees Join Board At that same meeting, Mr. Sydney M. Cone, III, was elected a member of the Class of 2002. He is a partner with the law firm of Cleary, Gottlieb, Steen & Hamilton in New York City. He is also Starr Professor of Law of International Trade and Finance and the Director of the Center for International Law at New York Law School. He has served as a member of the MBL's Council of Visitors since 1996. Mrs. Robert W. (Jean) Pierce was elected as a member of the Class of 1999. She is a resident of Wellesley, Woods Hole, and Boca Grande. Florida. She is an incorporator of the Heritage Plantation in Sandwich, and serves on the boards of the Women's Club of Boca Grande and the Penzance Point Road Trust. Her late husband. Robert, served as a member of the MBL's Board of Trustees from 1990 to 1993. Five Board Members Reappointed Mr. John R. Lakian, Chairman of the Board of the Fort Hill Group. Inc.; Dr. Joan Ruderman. Marion V. Nelson Professor of Cell Biology at Harvard Medical School; Dr. Sheldon J. Segal. Distinguished Scientist at the Population Council; Dr. William T. Speck. President and CEO of Columbia-Presbyterian Medical Center; and Mr. Alfred Zeien. Chairman and CEO of Gillette, were also elected in 1997 as members of the Class of 2002. Council of Scientific Advisors I owe a debt of thanks to my colleagues here in Woods Hole and to the international network of scientific, philanthropic, and business colleagues who share their time, interest, leadership, and support to keep the Marine Biological Laboratory at the forefront of biology. During 1997. I was especially fortunate to receive input and advice from the Council of Scientific Advisors, a newly constituted group of eminent scientists. They include: Drs. Bruce Alberts. Thomas Eisner. Walter Gilbert. Eville Gorham, Marc Kirschner. and Carla Shatz. With their counsel and that from our Board of Trustees, the MBL Science Council, and all of you, I look forward to a bright and dynamic future for the Marine Biological Laboratory. John E. Bums Report of the Treasurer The year 1997 was a successful one for the Marine Biological Laboratory. The net assets of the Laboratory increased by $6.2 M in 1997. from $54.5 M to $60.7 M. This increase was due to operations, the efforts of the capital campaign, and the success of our investment management activities. In accordance with generally accepted accounting principles, the Laboratory reported the results of operations after depreciation of $1.5 M, resulting in unrestricted net assets decreasing by $.8 M. This practice has masked a comparatively strong and healthy result of operations for the year. The capital campaign was responsible for a $3.1 M increase in net assets and permanently restricted net assets through increased gifts; and our long-term investments increased $4.9 M, resulting in a total return of 18.2% on our portfolio. The 1997 balance sheet revealed an increase in liquidity as cash and short-term investments increased by $.6 M to $5.0 M and account for 50% of current net assets. Long-term investments increased by $5.8 M. This increase was the result of $2.1 M of new gifts, market value increase of $4.9 M, and a withdrawal of $1.2 M to pay for scholarships and fellowships in accordance with the donors' instructions. Land, buildings, and equipment, net of accumulated depreciation, decreased by $.7 M as a result of the depreciation expense exceeding capital expenditures. This resulted in our inability to recover depreciation expense on federally funded capital projects. The Laboratory continues to demonstrate the ability to attract funds from the federal government, from foundations, and from individuals. We are in the process of significantly upgrading the physical plant, making the Laboratory an even more attractive facility at which to do science. Our housing budget continues to generate surplus cash. With long-term investment growth and the strength of our capital campaign, the Marine Biological Laboratory will continue to provide a productive scientific experience for its community. Mary B. Conrad R7 Financial Statements Coopers & Lybrand L.L.P. &Lybrand a professional services firm REPORT OF INDEPENDENT ACCOUNTANTS To the Board of Trustees of Marine Biological Laboratory Woods Hole, Massachusetts We have audited the accompanying balance sheet of Marine Biological Laboratory (the "Laboratory") as of December 31, 1997 and the related statements of activities and cash flows for the year then ended. These financial statements are the responsibility of the Laboratory's management. Our responsibility is to express an opinion on these financial statements based on our audit. We conducted our audit in accordance with generally accepted auditing standards. Those standards require that we plan and perform the audit to obtain reasonable assurance about whether the financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures in the financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management, as well as evaluating the overall financial statement presentation. We believe that our audit provides a reasonable basis for our opinion. In our opinion, the financial statements referred to above present fairly, in all material respects, the financial position of Marine Biological Laboratory as of December 31, 1997, and the changes in its net assets and its cash flows for the year then ended in conformity with generally accepted accounting principles. Our audit was conducted for the purpose of forming an opinion on the basic financial statements taken as a whole. The supplemental schedule of functional expenses as of December 31, 1997 is presented for the puipose of additional analysis and is not a required part of the basic financial statements. Such information has been subjected to the auditing procedures applied in the audit of the basic financial statements and, in our opinion, is fairly stated, in all material respects, in relation to the basic financial statements taken as a whole. Boston, Massachusetts f^ . ~7 .r~-~J / / . fi March 27, 1998 ** ^ ...... ~7 .r~-~J &*** Coopers & Lybrand L.L.P. is a member of Coopers & Lybrand International, a limited liability association incorporated in Switzerland. R8 MARINE BIOLOGICAL LABORATORY BALANCE SHEETS December 31, 1 997 (with comparative totals as of December 31, 1996) ASSETS Cash and cash equivalents Short-term investments, at market (Note Cl Accounts receivable, net of allowance for doubtful accounts of $36,782 in 1997 and $15,000 in 1996 Current portion of pledges receivable (Note H) Receivables due for costs incurred on grants and contracts Other assets 1997 $ 560,801 4,408.046 1,221,781 2,219,056 1,157,165 560,269 1996 $ 391,528 3,918,194 762,860 1,826.494 1,114.082 482,992 Total current assets 10.127,118 8,496.150 Long-term investments, at market (Notes C and D) Pledges receivable, net of current portion (Note H) Plant assets, net (Notes E and F) 35,614,151 2,238,826 20,026,580 29.763,495 2.406,350 20,695,624 Total long-term assets Total assets 57,879,557 $68,006,675 52.865,469 $61,361,619 LIABILITIES AND NET ASSETS Current portion of long-term debt and capital leases (Note E) Accounts payable and accrued expenses Deferred income and advances on contracts 229,657 1,494,948 384,258 204.108 1,536,015 321,998 Total current liabilities 2.108.863 2.062.121 Annuities and unitrusts payable Long-term debt and capital leases, net of current portion (Note E) Advances on contracts 1.213,583 2.567.370 1.433,208 959,513 2,591,973 1,210,950 Total long-term liabilities Total liabilities 5.214,161 7.323.024 4.762,436 6,824.557 Commitments and contingencies (Notes F and H) Net assets: Unrestricted Temporarily restricted Permanently restricted 18,729,311 25,596,656 16,357.684 19,371.978 21,484,748 13.680,336 Total net assets (Note B) 60,683,65 1 54,537,062 Total liabilities and net assets $68,006,675 $61,361.619 The accompanying notes are an integral part of the financial statements. R9 MARINE BIOLOGICAL LABORATORY STATEMENT OF ACTIVITIES for the year ended December 31, 1997 Operating support and revenues: Government grants Private contracts Laboratory rental income Tuition Fees for conferences and services Contributions Investment income Miscellaneous revenue Present value adjustment to annuities Net assets released from restrictions Total operating support and revenues Expenses: Research Instruction Conferences and services Other programs (Note B) Total expenses Change in net assets before nonoperating activity Nonoperating revenue: Total investment income and earnings Less: investment earnings used for operations Reinvested investment earnings Total change in net assets Net assets, beginning of year Net assets, end of year Temporarily Permanently 1997 Unrestricted Restricted Restricted Total $ 9,986.800 $ 9,986,800 1.178,192 1.178,192 1.478,757 1.478.757 399.703 399,703 3,085.616 3,085,616 615,882 $4.434,938 $1,390,609 (,,441,429 471.832 1.238.151 1 .709,983 322.667 322.667 (165.504) 1.057 (164.447) 3.811.922 (3.871,922) 60,000 21.351.371 1,635,663 1.451.666 24,438.700 11, 03 1, 1 ) 14 11.031.914 4,144,508 4.144,508 1,487.705 1.487.705 5.440.808 5.440,808 22.104.935 22,l()4. t )35 (753.564) 1,635.663 1.451.666 2.333.765 152.543 3,490,810 1.225.682 4.869,035 (41.646) (1.014.565) (1.056.211) 110,897 2.476.245 1.225.682 3.812.824 (642,667) 4.1 1 1.408 2.677.348 6.146,589 19,371.978 21.484.748 13,680,336 54,537,062 $18.729.311 $25,596.656 $16,357,684 $60,683,651 Tin' notes an- tin intci;nil part of the financial statements. Kill MARINE BIOLOGICAL LABORATORY STATEMENTS OF CASH FLOWS for the year ended December 31. 1997 (with comparative totals for the year ended December 31, 1996) Cash flows from operating activities: Change in net assets Adjustments to reconcile change in net assets to net cash (provided by) from operating activities: Depreciation Unrealized (gain) loss on investments Realized (gain) loss on investments Present value adjustment to annuities payable Contributions restricted for long-term investment and annuities Provision for bad debt Provision for uncollectible pledges Change in certain balance sheet accounts: Accounts receivable Pledges receivable Grants and contracts receivable Other assets Accounts payable and accrued expenses Deferred income and advances on contracts Annuities and unitrusts payable Advances on contracts Net cash provided by operating activities Cash flows from investing activities: Purchase of property and equipment Proceeds from sale of investments Purchase of investments Net cash used in investing activities Cash flows from financing activities: Payments on annuities and unitrusts payable Receipt of permanently restricted gifts Annuity and unitrusts donations received Loan proceeds Payments on long-term debt and capital leases Net cash provided by financing activities Net increase (decrease) in cash and cash equivalents Cash and cash equivalents at beginning of year Cash and cash equivalents at end of year 1997 6,146,589 1,483,203 (1,740,501) (1,728.792) 164.447 (1,390.609) 21,781 89.620 (480,702) (314.658) (43.083) (77,277) (71,564) 62,260 120,052 222.258 2,463,024 (814,159) 23,450,218 (26.321.432) (3.685.373) (30.430) 1,321,302 69,307 250.000 (218.5571 1,391.622 169,273 391.528 $ 560,801 1996 6,375,604 1.473.878 (1,765,532) (932,530) (118,076) (1,042,945) 5.000 (25,489) (1,726,570) 502,596 (15,940) (409,962) (59,539) 757.875 3.018,370 (730,966) 7,293,473 (11.347.575) (4.785,068) (23.369) 954.739 88.206 500,000 (205.878) 1,313.698 (453,000) 844.528 $ 391,528 The uc nii nines are an integral pun of the financial statements. Rll R12 Annual Report Marine Biological Laboratory Notes to Financial Statements A. Background: The Marine Biological Laboratory (the "Laboratory") is a private, independent not-for-profit research and educational institution dedicated to establishing and maintaining a laboratory or station for scientific study and investigation, and a school for instruction in biology and natural history. The Laboratory was founded in 1888 and is located in Woods Hole, Massachusetts. B. Significant Accoiintinx Policies: Basis of Presentation The accompanying financial statements have been prepared on the accrual basis of accounting and in accordance with the principles outlined in the American Institute of Certified Public Accountants' Audit Guide. "Not-For-Profit Organizations." The financial statements include certain prior-year summarized comparative information in total but not by net asset class. Such information does not include sufficient detail to constitute a presentation in conformity with generally accepted accounting principles. Accordingly, such information should be read in conjunction with the Laboratory's financial statements for the year ended December 31, 1996. from which the summarized information was derived. The Laboratory classifies net assets, revenues, and realized and unrealized gains and losses based on the existence or absence of donor-imposed restrictions and legal restrictions imposed under Massachusetts State law. Accordingly, net assets and changes therein are classified as follows: Unrestricted Unrestricted net assets are not subject to donor-imposed restrictions of a more specific nature than the furtherance of the Laboratory's mission. Revenues from sources other than contributions are generally reported as increases in unrestricted net assets. Expenses are reported as decreases in unrestricted net assets. Gains and losses on investments and other assets or liabilities are reported as increases or decreases in unrestricted net assets unless their use is restricted by explicit donor stipulations or law. Expirations of temporary restrictions on net assets, that is, the donor-imposed stipulated purpose has been accomplished and/or the stipulated time period has elapsed, are reported as reclassifications between the applicable classes of net assets. Temporarily Restricted Temporarily restricted net assets are subject to legal or donor-imposed stipulations that will be satisifed either by the actions of the Laboratory, the passage of time, or both. These assets include gifts plus monies for which the specific, donor-imposed restrictions have not been met and pledges, annuities, and unitrusts for which the ultimate purpose of the proceeds is not permanently restricted. As the restrictions are met, the assets are released to unrestricted net assets. Also, realized/unrealized gains/losses associated with permanently restricted gifts which are not required to be added to principal by the donor are classified as temporarily restricted but maintain the donor requirements for expenditure. Permanently Restricted Permanently restricted net assets are subject to donor-imposed stipulations that they be invested to provide a permanent source of income to the Laboratory. These assets include gifts, pledges and trusts which require that the corpus be invested in perpetuity and only the income be made available for program operations in accordance with donor restrictions. Nonoperating revenues include realized and unrealized gains on investments during the year as well as investment income on the master pooled investments. Investment income from short-term investments and investments held in trust by others is included in operating support and revenues. To the extent that nonoperating investment income and gains are used for operations as determined by the Laboratory's total return utilization policy (see below), they are reclassified from nonoperating to operating on the statement of activities as "Investment earnings used for operations." All other activity is classified as operating revenue. The Laboratory recorded net realized gains of $1,728,792, net unrealized gains of $1,740,501 and dividend and interest income of $2,053,514 in 1997. Cash and Cash Equivalents Cash equivalents consist of resources invested in overnight repurchase agreements and other highly liquid investments with original maturities of three months or less. Financial instruments which potentially subject the Laboratory to concentrations of risk consist primarily of cash and investments. The Laboratory maintains cash accounts with one banking institution. Investments are maintained primarily with two institutions. Investments Investments purchased by the Laboratory are carried at market value. Donated investments are recorded at fair market value at the date of the gift. For determination of gain or loss upon disposal of investments, cost is determined based on the first-in, first-out method. Investments with an original maturity of three months to one year are classified as short-term. All other investments are considered long-term. In 1924, the Laboratory became the beneficiary of certain investments, included in permanently restricted net assets, which are held in trust by others. The Laboratory has the continuing rights to the income produced by these funds in perpetuity, subject to the contractual restrictions on the use of such funds. Accordingly, the trust has established a process to conduct a review every ten years by an independent committee to ensure the Laboratory continues to perform valuable services in biological research in accordance with the restrictions placed on the funds by the agreement. The committee met in 1994 and determined that the Laboratory has continued to meet the contractual requirements. The market Financial Statements R13 values of such investments are $7,440.158 and $6,214,477 at December 31, 1997 and 1996, respectively. The dividend and interest income on these investments totaled $254,898 and $224.324 in 1997 and 1996. respectively. Investment Income and Distribution For the master pooled investments, the Laboratory employs a total return utilization policy that establishes the amount of the investment return made available for spending each year. The Finance and Investment Committee has approved a standing policy that the withdrawal will be based on a percentage of the latest three-year average ending market values of funds. The market value includes the principal plus reinvested income, realized and unrealized gains and losses. Spending rates in excess of 5%, but not exceeding 1%, can be utilized if approved in advance by the Finance and Investment Committee of the Board of Trustees. For fiscal 1997, the Laboratory obtained approval to expend 6% of the latest three-year average ending market values of the investments. The 6% includes a l'A% administration fee for endowments. This fee was approved by the Laboratory's Board of Trustees for the years 1996-2000. The net appreciation on permanently and temporarily restricted net assets is reported together with temporarily restricted net assets until such time as all or a portion of the appreciation is distributed for spending in accordance with the total return utilization policy and applicable state law. Investment income on the pooled investment account is allocated to the participating funds using the market value unit method (Note D). Plant Assets Buildings and equipment are recorded at cost. Donated facility assets are recorded at fair market value at the date of the gift. Depreciation is computed using the straight-line method, beginning the month after the asset is placed in service, over the asset's estimated useful life. Estimated useful lives are generally three to ten years for equipment and 20 to 40 years for buildings and improvements. Depreciation expense for the year ended December 31, 1997 amounted to $1,483,203 and has been recorded in the statement of activities in the appropriate functionalized categories. When assets are sold or retired, the cost and accumulated depreciation are removed from the accounts and any resulting gain or loss is included in unrestricted income for the period. Annuities and Unitrusts Payable Amounts due to donors in connection with gift annuities and unitrusts are determined based on remainder value calculations, as of December 31, 1997, with varied assumptions of rates of return and payout terms. Deferred Income and Advances on Contracts Deferred income includes prepayments received on Laboratory publications and advances on contracts to be utilized within the next year. Advances on contracts includes funding received for grants and contracts before it is earned. In certain circumstances, long-term advances are invested in the master pooled account until they are expended. Revenue Recognition Revenue is recognized at the time it is earned. The sources of revenue include grant payments from governmental agencies, contracts from private organizations, and income from the rental of laboratories and classrooms for research and educational programs. The tuition income is net of student financial aid of $536,097 and $479,000 in 1997 and 1996, respectively. Fees for conferences and other services include the following activities: housing, dining, library, scientific journals, aquatic resources and research services. Contributions Contribution revenue includes gifts and pledges. Gifts are recognized as revenue upon receipt. Pledges are recognized as temporarily or permanently restricted revenue in the year received and are recorded at the present value of expected future cash flows, net of allowance for unfulfilled pledges. Gifts and pledges, other than cash, are recorded at fair market value at the date of contribution. Expenses Expenses are recognized when incurred and charged to the functions to which they are directly related. Expenses that relate to more than one function are allocated among functions using various methodologies. Other programs expense consists primarily of fundraising, year-round labs, and library room rentals, costs associated with aquatic resource sales and scientific journals. Total fundraising expense for 1997 is $1,226,360. Use of Estimates The preparation of financial statements in conformity with generally accepted accounting principles requires management to make estimates and assumptions that affect the reported amounts of assets and liabilities and disclosure of contingent assets and liabilities at the date of financial statements and the reported amounts of revenues and expenses during the reporting period. Actual results could differ from those estimates. Reclassification Certain prior year amounts have been reclassified to conform with current year presentation. The reclassifications had no effect on net assets. Tax-Exempt Status The Laboratory is exempt from federal income tax under Section 501(c)(3) of the Internal Revenue Code. R14 Annual Report C. Investments: The following is a summary of the cost and market value of investments at December 31, 1997 and 1996: Market Cost 1997 1996 1997 1996 Certificates of deposit $ 40.000 $ 55,350 S 40.000 $ 55,350 Money market securities 2,168,958 3.469,588 2.168.958 3.469.589 U.S. Government securities 1,292,600 1,233,690 1. 098,526 998,829 Corporate fixed income 2,587,861 3,431.333 2.472.653 3.278.396 Common stocks 5,279.266 3.793,156 4.271.853 2.742,841 Mutual funds 23,223.812 17.371.972 19,317,499 14.555,958 Limited partnerships 5.429.700 4.326,600 3,309.994 3,108,464 Total investments $40,022.197 $33.681.689 $32,679.483 $28.209.427 Investment portfolios for the years ended December 31, 1997 and 1996 are as follows: Marke r Cost 1997 1996 1997 7996 Short- Term Investments Certificates of deposit $ 40,000 $ 55 ISO S 40 000 $ 55 350 Money market 1784 Fund 1,759,589 2,755.593 1,759,589 2,755,593 Common stocks 55 1 .780 50.608 530.936 50,608 Mutual funds 2.056.677 1 .056.643 2.056.679 1.063.089 Total 4.408,046 3.918,194 4.387,204 3,924,640 Long-Term Investments Pooled investments: Master pooled investments $26,163.702 $21,858,658 $20,201.962 $17,889,881 Separately invested: General Chase trust 5.846.916 4,898,630 4,986.443 3,926,959 Library Chase trust 1,593,242 1,315.847 1,358.149 1.065.008 Annuity and unitrust investments -, QJQ TQJ 1.690,360 1,745.725 1 .402.939 Total 35,614,151 29,763,495 28,292.279 24,284.787 Total investments $40,022.197 $33.681,689 $32,679,483 $28,209,427 D. Accounting fur Pooled Investments: Certain net assets are pooled for investment purposes. Investment income from the pooled investment account is allocated on the market value unit basis, and each fund subscribes to or disposes of units on the basis of the market value per unit at the beginning of the calendar quarter within which the transaction takes place. The unit participation of the funds at December 31. 1997 and 1996 is as follows: 7997 7996 Unrestricted 4.192 4.415 Temporarily restricted 42.693 41.426 Permanently restricted 65.41 1 66.442 Advances on contracts 6.506 6.493 i is. so: 118.776 Financial Statements R15 Pooled investment activity on a per-unit basis was as follows: Unit value at beginning of year Unit value at end of year Total return on pooled investments 7997 $ 186.35 220.30 $ 33.95 1996 $ 159.37 186.35 $ 26.98 E. Long-Term Debt: Long-term debt consisted of the following al December 31: Variable rate (5.50% at December 31, 1997) Massachusetts Industrial Finance Authority Series 1992 A Bonds payable in annual installments through 2012 6.63% Massachusetts Industrial Finance Authority Series 1992B Bonds, payable in annual installments through 2012 5.8% The University Financing Foundation. Inc., payable in monthly installments through 2000 5.8% The University Financing Foundation, Inc., payable in monthly installments through 2002 Capital leases with various rates and due dates /997 $ 960,000 1,280,000 325,210 231,817 1996 t 990,000 1.330,000 418,821 57,260 $2,797,027 $2,796,081 The aggregate amount of principal due on long-term debt for each of the next five fiscal years and thereafter is as follows: 1998 1999 2000 2001 2002 Thereafter $ 229,657 243,274 267,404 173,664 148,028 1.735,000 Less current portion of long-term debt 2,797,027 (229,657) Long-term debt net of current return $2,567,370 In 1992, the Laboratory issued $1,100,000 Massachusetts Industrial Finance Authority (MIFA) Series 1992A Bonds and $1,500,000 MIFA Series I992B. These bonds pay varying annual interest rates ranging from 3.48% to 6.63%. Interest expense on this debt totaled $141,899 for the year ended December 31, 1997. The Series 1992 A and B Bonds mature on December 1. 2012 and are collateralized by a first mortgage on certain Laboratory property. The agreements related to these Bonds subject the Laboratory to certain covenants and restrictions. Under the most restrictive covenant of this debt, the Laboratory's operating surplus (before transfers), interest, expense and transfers from the quasi-endowment for debt service must equal or exceed all debt service payments, as defined by the agreement. The Laboratory was in compliance with these covenants and restrictions at December 31. 1997. In 1996, the Laboratory borrowed $500,000 with an interest rate of 5.8% from the University Financing Foundation, Inc. The interest expense for the year ended December 31, 1997 was $21.829. The loan matures in 2000 and is collateralized by 69,440 shares of a fixed income mutual fund with a fair value of $933,690 at December 31, 1997. In 1997, the MBL borrowed $250,000 with an interest rate of 5.8% per annum from the University Financing Foundation, Inc. The interest expense for the year ended December 31. 1997 was $5,866. This loan matures 2002 and is collateralized by 69,440 shares of a fixed income mutual fund with a fair value of $933,690 at December 31, 1997. The Laboratory has a line of credit agreement with BankBoston from which it may draw up to $1,000.000. No amounts were outstanding under this agreement as of December 31, 1997 and 1996. R16 Annual Report F. Plant Assets: Plant assets consist of the following at December 3 1 : Land Buildings Equipment 1997 $ 702.908 32.419.072 4.300.932 1996 $ 702.908 3 1 .730,830 4.270,278 Total Less: Accumulated depreciation 37.422.912 (17.396.332) 36,704,016 (16.008.392) Plant assets, net $20.026.580 $20,695.624 G. Retirement Plan: The Laboratory participates in the defined contribution pension plan of TIAA-CREF (the "Plan"). The Plan is available to permanent employees who have completed two years of service. Under the Plan, the Laboratory contributes 10% of total compensation for each participant. Contributions amounted to $715.858 and $661.089 for the years ended December 31, 1997 and 1996. respectively. Unconditional promises to give are included in the financial statements as pledges receivable and the related revenue is recorded in the appropriate net asset category. Unconditional promises to give are expected to be realized in the following periods: In one year or less Between one year and live years After live years 1997 $2,219.056 2,485,851 80,000 1996 $1,836,874 2,780,000 Total 4,784,907 4.616,874 Less: discount of $227,025 in 1997 and $373.650 in 1996 and allowance of $100.000 in 1997 and $10,380 in 1996 (327,025) (384,030) $4.457,882 $4,232.844 Pledges receivable at December 31 have the following restrictions: Research and education Permanently restricted net assets $3,787,882 670,000 $4,232,844 $4,457,882 $4.232.844 I. Pu.ttri'tirement Benefils: The Laboratory accounts for its postretirement benefits under Statement No. 106. "Employers' Accounting for Postretiremen! Benefits Other than Pensions." which requires employers to accrue, during the years that the employee renders the necessary service, the expected cost of benefits to be provided during retirement. As permitted, the Laboratory has elected to amortize the transition obligation over 20 years commencing on January 1. 1994. The Laboratory's policy is that all current retirees and certain eligible employees who retired prior to June I, 1994 will continue to receive postretirement health benetits. The remaining current employees will receive benefits: however, those benefits will be limited as defined by the Plan. Employees hired on or after January 1, 1995 will not be eligible to participate in the postretirement medical benefit plan. Financial Statements R17 Net postretiremen! benefits for 1997 and 1996 include: Service cost (benefits earned during period) Interest cost (on projected benefit obligation) Actual return on plan assets Net amortization and deferral Net postretiremen! benefits cost 1997 $ 29,095 135,892 (31.976) 59.07 1 $ 192,082 1996 $ 61,712 149.149 (20.700) 92.176 $ 282,337 Below is a reconciliation of the funded status of the Plan at December 31. 1997 and 1996: Accumulated postretirement benefit obligation: Retirees and dependents Fully eligible active participants Other active participants $1,439,899 163.188 316.778 $1.369.787 230,290 608.796 Total Market value of plan assets Unfunded obligations Unrecognized prior service cost (credit) Unrecognized net (gain) loss Unrecognized transition obligation Accrued postretirement benefit cost 1,919,865 701.140 1,218,725 (191,078) 1.389.159 $ 20,644 2,208.873 588.337 1,620,536 136.158 1.475.981 $ (8,397) The health care cost trend rate assumptions used in determining the projected benefit obligation begin at 9.5% in 1997 and gradually decrease to 5.0% in the year 2006 and thereafter. The effect of raising the assumed health care cost trend rate by one percentage point in each year would be to increase the accumulated postretirement benefit obligation as of December 31. 1997 by $173.001 and to increase the aggregate of the service and interest cost components of net periodic postretirement benefit cost for the year then ended by $14.528. The discount rate used in determining the accumulated postretirement benefit obligation is 7.5%, and the expected return on plan assets was 8.0%. During 1997, the Laboratory contributed $192,082 to fund the Trust for these postretirement benefits. >H Oi O < ct O CO < J J < u O S UJ Z 5 00 1 X w z (J U. O D Q UJ U CO tu U-l si CL D CO ON ON -o I 3 1) in r*-, O m oo so ri ^ r - ON n n m c J m ffi ^ rj m ON N *t CO r "i n"i r OS ON 1 OO OO 00 r-i r-l so "j O r- o co r- T 3 O n h rj o f . T " sO i (N t -T ' 3 1 O od r J _ ^- - -' ri ri r 1' * ri rj 11 8 oo rj >n ON n ON in Tt ON m SO P "-1 00 SO o S ""* Tt ON ON ^> r" ON -t \ ri ri sO ^o* la ri oo r-4 O 11 ON Tj- IT, ON r*"i (N "~1 ON O "* n si O K en S r ?3 rj pvj r- r^ 00 oo oo oo ^ r W", oo' r-l r-4 ? ^ -rf ON in ON ON O- QV ^ to Z fi -f ^ ^ ^ m r- c ^ f- O so -t ^~ ' ^ r- r- O m m ON r- o - O r- - - sO -t ON S In 1 i \O' rn r*^ 1 in M 5 * ON r- r-' "t ON 1 r o o -3 en t "d ON 1 O in ^ f S 3 ri >/~) O J- Tj- ON ^ od ri rn - -.-)[ r*"j Os ON ri ^_J oo in so c ri i/- ( ri o ^ ; CO C N in m m, ri oo O r*-* r-' rK ' ^ 6^ onal ^ 4J O .^- "D. ^ * S ^ , 9- 4J (rt rj ^ | c2 - i5'S"aS2FS5 tL.a.onWooHiioDQO I o R18 Report of the Library Director One of the goals of the MBL's recently announced Discovery Campaign is to raise $1 million in support of the MBLAVHOI Library. This is excellent news indeed for the library. Funding at this level will be a good start towards protecting for future generations of scientists one of the world's most comprehensive and valuable collections of books and journals in the biological, oceanographic, and ecological sciences. This support will also help in the development of information technology, which promises to play a vital role in the future of the library. The way the library will serve scientists in the next decade will be very different from the way it serves them today. More and more the library is becoming a conduit rather than a repository of information, providing on-line access to the burgeoning electronic literature. But even as electronic delivery alters the library and the way it is used, the MBL's irreplaceable book and journal collection continues to be a national treasure that must be conserved. The Library Committee has spent considerable time over the past year evaluating our collection as a whole, the library's physical plant, and the future of both the paper and electronic library. The committee found that even with the advent of more electronic resources in the library, still more room will soon be needed to house all the early monographic series now available in the stacks and the hardware required to serve the growing electronic resources. With fellow members of the Boston Library Consortium, we are looking into the feasibility of having an off-site storage facility. If this plan is realized, we can look forward to a cooperatively run depository that will offer electronic retrieval and courier service as well as warehousing for some of our lesser used volumes, making room for other, ever expanding information resources. Special collections A new Special Collections area has been created in the third floor book section of the library through a generous gift made in memory of Ivor Cornman. In this area, the expeditions, bound collected reprints, and books from the open shelves published before 1900 are being protected. Earlier works, currently on the open shelves in the stacks, are being cataloged and placed in the Rare Books Room. Digital library As we preserve the older collection, we are expanding the electronic collection. The library currently supports more than 300 full-text electronic journals on its web site (www.mbl.edu). We provide access to more than 35 commercial databases and provide public access to databases created here at the MBL on marine animals, E. coli, women-in-science, the Leuckart charts, and other educational material of interest to both scientists and students. From the Community of Science we have purchased funding and patent databases including the Expertise Database, which contains information on individual scientists and their projects. Woods Hole scientists who enter information about their work in the data bank are rewarded with timely information on available funding sources in their field. The library continues to update to the latest versions of Internet Grateful Med and PubMed and InfoBank Search. The latter service includes newspaper, magazine, academic, and general reference searching targeted to our student and visitor populations. Bibliographies of relevant resources for the many summer courses held at the MBL have been created and can be accessed via the web. The Don Flescher photo collection of fish is a new database available from the library home page. It is comprised of 1500 pictures searchable by species, common name, catch area, or taxonomy. Journal issues One of the newest issues facing libraries, along with the continued rise in subscription prices, is the new R19 R20 Annual Report phenomena of publishers collecting electronic information at the "article level" when library users log into their sites. Vendors and publishers can now gather specific user data right down to the exact article, phrase, and words that someone pulls up or orders. What happens to this information? If this user information should be sold to the right company, it could become important for determining who else in the world is interested in, or perhaps working on, the same research problem. Will investigators hesitate to look at publishers' electronic sites for fear of competitors getting a hint of what they are working on? What protections can the library provide? A cloak of anonymity, possibly; but right now information on article use is transmitted almost exclusively to the publishers and not provided to the library. Like other libraries, the MBLAVHOI Library is now managing licenses covering information content as opposed to purchasing the contents. At the same time, we are trying to negotiate protections for our users. Our newly appointed Technical Services Coordinator carries the responsibility of "watchdog" for these licenses. From the arena of fair use, we now are venturing into the arena of ethical use of information. Ensuring access to today's electronic resources tomorrow is another major concern. Since 1991 the library has canceled more than 500 journal subscriptions. In 1991 we borrowed 1858 articles for our patrons. In 1997 we borrowed 3259, representing a 42% increase over six years. This practice will continue to increase as it becomes necessary to cancel titles in order to meet publishers' price increases for "must have" journals. As a result, electronic document delivery, whether providing documents for other libraries or retrieving materials for our own scientists, is becoming an ever-increasing component of the library's services. Instruction The new Science Reference Librarian has quickly expanded and improved the library's instructional programs on all four campuses (MBL. WHOI, Fisheries, and USGS). Training and course instruction is now provided for researchers and students, and new on-line bibliographies and reserve lists have been created for MBL summer courses. There continues to be high demand for new courses in how to search scientific databases, JAVA, Web page design, Windows 95, Powerpoint, etc. Branch libraries Jacqueline Riley, the Science Reference Librarian at 'he Fisheries, has implemented many new services including launching a NEFSC library web page. The contract that provides library services for the National Marine Fisheries, while bringing a collegia! interchange at the local level, presents some technological problems at the national level that resulted in different automated library systems. The marriage of two different library systems seemed to be impossible until a method was devised by Laurel Duda (MBL) and Maggie Rioux (WHOI) that allows both libraries to share bibliographic records without redundant effort (Duda, L.E. and M.A. Rioux, One Library, One Bib Record- Two OPACs, Two Systems, Information Outlook Vol. 2, No 3: 31-36). A new library was incorporated into the design of the new WHOI Research Vessel Atlantis. Space was designed for a computer workstation with Internet access for the researchers and staff to fulfill their information needs while at sea. A reference collection, popular works, and a wonderful collection of World War II monographs donated by Ken Parda are also available aboard ship. The Data Library, using the Library's integrated computer systems, has installed a new image server, providing photographic images of the WHOI scientific instrument collection. The new ALVIN archive catalog provides retrieval of individual ALVIN dive records through our local on-line catalog. MARINER, including all related media the data archive holds and hot links to images from the dives. More than 700 of the photographs in the MBL archives, including the late 1800 and early 1900 collection of MBL's Baldwin Coolidge photos, have also been cataloged and are available through MARINER and the Web. A growing debate focusing on numerical data in the Earth and Environmental sciences and the issues relevant to the management of all non-text information was the focus of the Scientific Data Advisory Committee which met this past year at WHOI. The major problem identified was access to the data and how the library must make it easier for users of data sets to find, select, and retrieve that data. In order to do this the Data Library needs to encourage submission of data sets in a useable standard metadata format, with appropriate documentation. Because getting scientists to do the latter is referred to as "herding cats," the community needs to provide assistance and incentives. The future The National Library of Medicine awarded the library $1,033,278 in support of the project titled, "Professional Services in Support of NLM's Outreach Efforts to Encourage the Use of Computers and Information Science in Medicine." With this support, Library Director's Report R21 the library is able to plan for "change" over the next five years and to offer services that will fit evolving needs. However, all of the members of the scientific community must promote the value of academic access to information for research and teaching. They will also need to use innovative technology to improve scholarly communications without operating in an economic model tied to increasing costs from large commercial publishers. Libraries and administrators have realized that spiraling costs cannot be solved by strategies that merely include canceling titles and collective purchasing. Change must be effected through partnerships in new technologies, and the development of new models of publication and distribution. While the new electronic environments offer opportunities to provide new services, we also need to think about the sustainability of the information ecosystem: managing the life cycle of information for preservation and access. This year the Florence Gould Foundation awarded $30,000 to the library for the preservation and conservation of the French scientific rare books in our collection. The library will make these books available electronically and will use some of the high-end translation dictionaries now available on our Web page that allows French to English translation. As noted, one of the goals of the Discovery Campaign is to help preserve the library facility and the book and journal collections for future scholars. Managing this traditional collection will be much easier than managing the new electronic collection, which presents a number of new issues, not the least being the lack of planning for digital archival storage and migration to new technologies by the publishers. As we approach the 21st Century, "change" will be the MBL/ WHOI Library's primary challenge. Catherine Norton Educational Programs Summer Courses Biology of Parasitism: Modern Approaches (June I2-August 15) Directors Steven Hajduk. University of Alabama. Birmingham Edward Pearce, Cornell University Facult\ Robert Bullis. Marine Biological Laboratory Harry Dickerson. University of Georgia Steve Ealick, Cornell University Christopher Hunter, University of Pennsylvania School of Veterinary Medicine Susan Little. University of Georgia Phil LoVerde, State University of New York, Buffalo David Russell. Washington University School of Medicine Phillip Scott. University of Pennsylvania David Sibley, Washington University School of Medicine Christian Tschudi, Yale University School of Medicine Buddy Ullman, Oregon Health Sciences University Ehsabetta Ullu, Yale University School of Medicine Teaching Assistants Laura Rosa Brunei. Cornell University Nino Campobasso, Cornell LIniversity Dai-rick Carter, Oregon Health Sciences University Jaime Dant, Washington University Medical School Jerome Drain. University of Alabama. Birmingham Merle Elloso. University of Pennsylvania Wendy Freebern, State University of New York. Buffalo Maren Lingnau. Washington University School of Medicine Carol Lopez-Estrano. Yale University Joao Pedras-Vasconcelo, Cornell University Veterinary School Elizabeth Sabin. Cornell University Lecturers Norma Andrews, Yale University School of Medicine James Bangs, University of Wisconsin, Madison Stephen Beverley, Harvard University School of Medicine John Boothroyd, Stanford University Karen Day, Oxford University, United Kingdom John Donelson, University of Iowa College of Medicine Paul Englund. Johns Hopkins University School of Medicine Daniel Goldberg, Washington University Richard K. Grencis, University of Manchester. United Kingdom Kasturi Haldar, Stanford University School of Medicine John Heuser. Washington University School of Medicine Stephen Hoffman. Naval Medical Research Institute Peter Hotez, Yale University School of Medicine William Jeffery, Pennsylvania State University Patricia Johnson. University of California. Los Angeles Richard Komuniecki. University of Toledo Manfred Kept", Basel Institute for Immunology, Switzerland Jean Langhorne. Imperial College of Science and Technology. United Kingdom Timothy Nilsen. Case Western Reserve University Marilyn Parsons. Seattle Biomedical Research Institute William Petri. University of Virginia Health Sciences Center Meg Phillips. University of Texas Southwest Steven Reed. Infectious Disease Research Institute Steve Reiner. University of Chicago David Sacks, National Institutes of Health Alan Sher, National Institutes of Health Mitchell Sogin. Marine Biological Laboratory Kenneth Stuart, Seattle Biomedical Research Institute Sam Turco, University of Kentucky College of Medicine C. C. Wang. University of California, San Francisco Gary Ward, University of Vermont Tom Wellems, National Institutes of Health R22 Educational Programs R23 Dyann Wirth. Harvard University School of Public Health Maria Yazdanbakhsh, Leiden University, The Netherlands Course Coordinator Robert Sabatini, University of Alabama. Birmingham Course Assistant Brian Hondowicz, University of Pennsylvania Laboratory Assistant Joseph Paton, Tufts University Students Kimberly Brouwer, Johns Hopkins School of Public Health Maristela Camargo, Federal University of Minas Gerais, Brazil Soren Gantt, New York University School of Medicine Michelle Gleeson, University of Technology, Australia Meike Hensmann, Yale University David Jiang, Johns Hopkins University School of Medicine Andreas Lingnau, Washington University Medical School Gunnar Mair, Queen's University of Belfast. United Kingdom Antoinette Marsh, University of California. Davis Lillian Ouko, University of Pennsylvania Lisl Shoda, Washington State University Kevin Tan, National University, Singapore Carl Johan Treutiger, Karolinska Institute, Sweden Elisabeth Triplet!, Cornell University Eric Villegas, University of Pennsylvania Ulrike Wille, University of Tubingen, Germany Embryology: Concepts and Techniques in Modern Developmental Biology (June 15-July 26) Directors Marianne Bronner-Fraser, California Institute of Technology Scott Fraser, California Institute of Technology Faculty Andre Adoutte, Universite Paris-Sud, France Rosa Beddington, National Institute for Medical Research, United Kingdom R. Andrew Cameron, California Institute of Technology Andrea Collazo, California Institute of Technology Eric H. Davidson, California Institute of Technology Richard Harland, University of California, Berkeley Lee Niswander. Memorial Sloan-Kettering Cancer Center James Posakony, University of California, San Diego Nadia Rosenthal, Massachusetts General Hospital Joel Rothman, University of California, Santa Barbara Joshua Sanes, Washington University School of Medicine John Saunders, Jr.. Marine Biological Laboratory Martin Shankland, University of Texas Judith Venuti, Columbia University Teaching Assistants Maria Ina Arnone, California Institute of Technology Guillaume Balavoine. Wellcome/CRC Institute, United Kingdom Rebecca Beach, Hollins College Steve Gendreau, University of California. Santa Barbara Gabrielle Kardon, Duke University Carmen Kirchhamer, Harvard University Anne Knecht, University of California. Berkeley Catherine Krull. California Institute of Technology Julie Kuhlman, Memorial-Sloan Kettering Cancer Center Eric Lai. University of California. San Diego Nicholas Lartillot. Universite Paris-Sud, France David Nellesen. University of California. San Diego Craig Neville, Massachusetts General Hospital Robert Zeller, University of California, San Diego Lecturers Sean Carroll, University of Wisconsin Nancy Hopkins, Massachusetts Institute of Technology Michael Levine, University of California, Berkeley Tom Maniatis. Harvard University David McClay, Duke University Course Coordinator Linda Huffer. Marine Biological Laboratory Course Assistants Allison Freeman, Wheaton College Jason Rickles, Boston University School of Education Students Pia Aanstad, University of Newcastle, United Kingdom Dominique Bergmann, University of Colorado, Boulder Brian Cooper. National Institute for Medical Research. United Kingdom Wim Damen, Munchen University. Germany Gregory Davis. University of Chicago Eric Finkelstein. State University of New York Health Science Center Paola Giuliano, Zoological Station, Italy Tiffany Heanue, Harvard University Sharon Hesterlee, University of Arizona Christoph Kaufmann. Massachusetts General Hospital Peter Ladurner, University of Innsbruck, Austria Walter Lerchner, National Institute for Medical Research, United Kingdom Per Lindahl, University of Goteborg, Sweden Matthew Marlowe, University of California, Berkeley Sean Megason, Harvard University Raffaella Melfi, Universita Degli Studi di Palermo, Italy Axel Meyer, State University of New York Bennett Novitch, Harvard Medical School Catherine Olsen, University of California. Davis Eva Reissmann. Max-Planck-Institut, Germany Elaine Seaver. University of Texas. Austin James Spotts. California Institute of Technology Xin Sun, University of California. San Francisco Christina Takke, University of Koln, Germany Emily Walsh. University of California, San Francisco Microbial Diversity (June 15-July 31) Directors Edward Leadbetter, University of Connecticut Abigail Salyers, University of Illinois R24 Annual Report Faculty Kurt Hanselmann. University of Zurich. Switzerland Bruce Paster, Forsyth Dental Center Rolf Schauder. University of Frankfurt, Germany Teaching Assistants Elena Barbieri. University of Urbino, Italy Daniel Gisi, Swiss Federal Institute of Technology, Switzerland David Graham, University of Illinois Irena Levin. Forsyth Dental Center Caroline Plugge, Wageningen Agricultural University. The Netherlands Lecturers Bianca Brahamsha. University of California. San Diego Colleen Cavanaugh, Harvard University Yehuda Cohen. Hebrew University of Jerusalem, Israel Sharon Danielson. Rensselaer Polytechnic Institute Daniel Distel, University of Maine Stephen Farrand. University of Illinois. Urbana-Champaign James Fleming. University of Tennessee Tillman Gemgross, Metabolix. Inc.. Cambridge Robert Haselkorn, University of Chicago Holger Jannasch. Woods Hole Oceanographic Institution Jared Leadbetter, Michigan State University Mark McBride, University of Wisconsin, Milwaukee Terry Miller, New York State Department of Health Duane Moser, Center for Great Lakes Studies. Milwaukee Sandra Nierzwicki-Bauer, Rensselaer Polytechnic Institute Thomas Pitta. Rowland Institute for Science James Russell, Cornell University Daad Saffarini, University of Massachusetts Anglica Seitz, Harvard University Nadja Shoemaker, University of Illinois Alfred Spormann, Stanford University David Stahl, Northwestern University Anne Summers. University of Georgia Andreas Teske, Woods Hole Oceanographic Institution Pieter Visscher. University of Connecticut, Avery Point John Waterbury. Woods Hole Oceanographic Institution Lily Young, Rutgers University Course Coordinator Madeline Vargas, College of the Holy Cross Course Assistant Noah Horst, Tufts University Students Ehud Banin, Tel Aviv University, Israel Fikry Barghuthy. Hebrew University, Israel Brendan Bohannan, Michigan State University Alfred Boyle, Rutgers University Scott Dawson, University of California, Berkeley Jeffrey Dugas, University of Connecticut Kelly Evans, Queen's University. Canada Deborah Hughes, Scripps Institution of Oceanography Elke Jaspers. Universitat Oldenburg. Germany Hope Johnson, Stanford University Sabine Krause, Max-Planck-Instiltit, Germany Thomas Lie, University of Connecticut David Long, Montana State University Junko Munakata Marr. Colorado School of Mines William Sobczak, Cornell University Alexandra Stone. Ohio State University Acharawan Thongmee. University of North Texas Crisogono Vasconcelos, Universidade Federal Fluminense, Brazil S. Wenuganen, Bogor Agricultural University, Indonesia Ludek Zurek. University of Alberta. Canada Neural Systems & Behavior (June 15-Augmt 8) Directors Janis Weeks, University of Oregon Harold Zakon, University of Texas, Austin Faculty Ronald L. Calabrese. Emory University Catherine Carr, University of Maryland Kathleen French, Llniversity of California, San Diego David Glanzman, University of California. Los Angeles Scott Hooper, Ohio University Richard Hyson, Florida State University Masashi Kawasaki, University of Virginia William Kristan. University of California. San Diego Richard Levine, University of Arizona Pierre Meyrand, Universite de Bordeaux, France Michael Nusbaum, University of Pennsylvania School of Medicine Glen Prusky, University of Lethbridge, Canada William Roberts, University of Oregon Gary Rose, University of Utah Angela Wenning, Universitat Konstanz. Germany Teaching Assistants Cecilia Armstrong. University of Oregon Dawn Marie Blitz, University of Pennsylvania School of Medicine Shanping Chen, University of California, Los Angeles Richard Dyck. University of Lethbndge. Canada Michael Ferrari, University of California, San Diego Evelyn Field, University of Lethbridge. Canada Matthew Friedman. Cornell University Jorge Golowasch, Brandeis University Andrew Hill. Emory University Maria Kubke. University of Maryland David Lenzi, University of Oregon Lynne McAnelly. University of Texas Tasha McMahon, University of California. Los Angeles Geoffrey Murphy, University of California, Los Angeles David Sandstrom, LIniversity of Arizona Daphne Scares, University of Maryland Laura Wolszon, Columbia University Lecturers George Augustine, Duke University Medical Center Rita Balice-Gordon, University of Pennsylvania School of Medicine David Clayton, LIniversity of Illinois Roger Hanlon, Marine Biological Laboratory Ron Hoy, Cornell University Edward Kravitz, Harvard Medical School Educational Programs R25 Eduardo Macagno. Columbia University Gina Turrigiano. Brandeis University Scholars-in-Residence Sascha du Luc, Salk Institute Alan Gelperin. Bell Labs Rebecca Johnston, Colby College Simon Laughlin. University of Cambridge. United Kingdom Course Assistants Sarah Doernberg, Emory University Sarah Gaines, Princeton University Students Emre Aksay, Bell Labs Eric Bauer. University of Texas. Austin Ben Bonham. University of California. San Francisco Holly Campbell, University of Arizona Randy Chitwood, University of Texas. San Antonio Virginia de Sa. University of California. San Francisco Teresa Esch. University of Virginia Joe Pass. Michigan Tech University Paul Gasser, Arizona State University Timothy Holy, Princeton University William Kargo. Allegheny University Christopher Kilroy, University of North Carolina, Wilmington Susan Laessig. University of Maryland, Baltimore Lynne Merchant, University of California, San Diego Leslie Osbome, University of California, Berkeley Dorit Polnaii, Northeastern University Rex Robison. Stanford University Sen Song. Brandeis University Ayako Yamaguchi. Columbia University Shih-Rung Yeh. Georgia State University Neurobiology (June 15- August 16) Directors Gary Banker, University of Virginia Medical School Daniel Madison. Stanford University Medical Center Section Directors Michael Greenberg. Childrens' Hospital Stephen Smith. Stanford University Medical Center Facult\ Susan Birren. Brandeis University Mark Bovvlby, Wyeth-Ayerst Research Gabriel Corfas, Childrens' Hospital Kerry Delaney, Simon Eraser University. Canada Donald Faber. Allegheny University of the Health Sciences Steven Finkbeiner, Childrens' Hospital Philip Haydon, Iowa State University Stephen Jones. Case Western Reserve University Maurice Kernan, State University of New York. Stony Brook Stephen Lin, Wyeth-Ayerst Research Ed McCleskey, Oregon Health Sciences University Adam Rory McQuiston, Stanford University School of Medicine Thomas Reese. National Institutes of Health Morgan Sheng. Howard Hughes Medical Institute Carolyn Smith, National Institutes of Health Stuart Thompson, Stanford University Li-Huei Tsai, Harvard Medical School Susan Wray, National Institutes of Health Joshua Zimmerberg, National Institutes of Health Teaching Assistants F. Woodward Hopt, Stanford University Medical Center Alane Murdock, Stanford University Medical Center Paul Pavlidis, Stanford University Medical Center Alberto Pereda, Allegheny University Marline Usdin, Stanford University Medical Center Lecturers Yadin Dudai, Weizmann Institute of Science, Israel Justin Fallon, Brown University Julie Ann Kauer, Duke University Medical Center Maurine Linder. Washington University Medical School Diane Lipscombe, Brown University Luis Parada, University of Texas Robert Rosenberg, University of North Carolina, Chapel Hill Gary Ruvkun, Massachusetts General Hospital Thomas Sudhof, University of Texas Karel Svoboda, Cold Spring Harbor Laboratory Roger Tsien, University of California, San Diego Course Assistants Tara Bennett, Dartmouth College Eleanore Edson, Stanford University Students Alberto Bacci. University of Milano, Italy Elizabeth Brown, Medical College of Virginia Benjamin Cravatt. Scripps Research Institute Samuel Hess, Cornell University Warren Kim. Yale University School of Medicine Stuart Licht. Massachusetts Institute of Technology Johanna Montgomery, Otogo University. New Zealand Craig Nelson, Harvard University Eric Norman, University of Pittsburgh Rita Saltier, University of Toronto, Canada Marilee Shelton. University of North Carolina. Chapel Hill Stephan Sigrist. Max-Planck-Institiit. Germany Physiology: Cellular and Molecular Biology (June 16-July 27) Directors Kerry Bloom, University of North Carolina, Chapel Hill Mark Mooseker. Yale University Faculty William Bement, University of Wisconsin Richard Cheney, University of North Carolina. Chapel Hill Jonathan Chernoff, Fox Chase Cancer Center Ruth Empson. University of Oxford, United Kingdom Antony Galione, University of Oxford. United Kingdom Tom Hays, University of Minnesota, St. Paul Daniel Kiehart, Duke University Medical Center Daniel Lew. Duke University Medical Center R26 Annual Report Mary Ann Sells. Fox Chase Cancer Center Edwin Taylor, University of Chicago Joseph S. Wolenski, Yale University Elaine Yeh. University of North Carolina, Chapel Hill Teaching Assistants Elaine Bardes, Duke University Medical Center Lisa Evans, Yale University Tama Hasson, Yale University Amanda Hayward-Lester. Yale University Gary Michael Idelchik, University of Wisconsin Min-gang Li, University of Minnesota, St. Paul Feng Liu, Fox Chase Cancer Center Craig Mandate, University of Waterloo, Canada Ruth Montague. Duke University Medical College Samara Reek-Peterson. Yale University Melissa Reeder, Fox Chase Cancer Center Rey Antonio Sia. Duke University Medical Center Jenny Sider, University of Wisconsin Andre Silvanovich, University of Minnesota. St. Paul Lecturers Sid Altman, Yale University Angelika Amon. Massachusetts Institute of Technology Van Bennett, Duke University John Chant, Harvard University Paul Forscher, Yale University Martin Hemler, Dana Farber Cancer Institute Laurinda Jaffe, University of Connecticut Health Center Jennifer Lippincott-Schwartz. National Institutes of Health Bruce Mayer. Harvard University Ben Neel. Beth Israel Hospital Terry Orr- Weaver, Massachusetts Institute of Technology Michael Snyder, Yale University Karl Swann. University of Connecticut Health Center Mark Terasaki, University of Connecticut Health Center Lew Tilney, University of Pennsylvania Pat Wadsworth. University of Massachusetts. Amherst Sandra Wolen. Yale University Tian Xu, Yale University Medical School Bruce Zetter, Harvard University Course Coordinator Dawn Grant, Louisiana State University Medical Center Course Assistant Rchekah Harrison, Stanford University Students Richard Bayliss, University of Cambridge. United Kingdom Dale Beach, University of North Carolina. Chapel Hill George Bell, University of Arizona Jill Broome, University of North Carolina, Chapel Hill Julie Canman, University of North Carolina, Chapel Hill Anthony DePass, University of Massachusetts. Amherst John Diggins, Providence College Jean-Emmanuel Faure, Centre National de la Recherche Scientitique. France Rip Finst, Emory University Gundula Ones, University of Pennsylvania Edward Guo, Columbia University William Heinz, Johns Hopkins University Maria Holzmann, University of Geneva. Switzerland G. Karthikeyan, Tata Institute of Fundamental Research, India Kaoru Katoh. Marine Biological Laboratory Reed Kelso, Yale University Tijs Ketelaar. Wageningen Agricultural University, The Netherlands Rosy Lee, Stanford University Laura Linz, Louisiana State University Medical Center Jonathan Lyon, Scripps Institute of Oceanography Paul Maddox, University of North Carolina. Chapel Hill Gregory McGillem, University of Alabama, Birmingham Spontaneous McKnight. University of Arizona Brian Nihhelink, University of Vermont Helen Nilsson, University of Goteborg, Sweden Nesrin Ozoren, University of Pennsylvania Omar Quintero, Duke University Medical Center Rachael Ream. Hopkins Marine Station Eric Reese, University of California, Riverside Yasushi Satoh, University of Tokyo. Japan Matthew Simmons. University of Florida Justin Skoble, University of California, Berkeley Aline Valster, University of Massachusetts. Amherst Gang Wang, University of Iowa Yihong Wang, Ohio University Kathryn White, Scripps Institute of Oceanography Short Courses Analytical & Quantitative Light Microscopy (May 8 -May 16) Directors Greenfield Sluder, University of Massachusetts Medical Center. Worcester Foundation Campus David Wolf. University of Massachusetts Medical Center. Worcester Foundation Campus Faculty William B. Amos. Medical Research Council, United Kingdom Richard Cardullo, University of California, Riverside Walter Carrington, University of Massachusetts Medical School Jeff Gelles, Brandeis University Shinya Inoue. Marine Biological Laboratory Rudolf Oldenbourg. Marine Biological Laboratory Edward Salmon, University of North Carolina, Chapel Hill Randi Silver, Cornell University Medical College Kenneth Spring, National Institutes of Health Educational Programs R27 Yu-li Wang. University of Massachusetts Medical Center, Worcester Foundation Campus Warren Zipsel, Cornell University Teaching Assistants Christine Thompson. University of Massachusetts Medical Center. Worcester Foundation Campus Elizabeth Thompson. University of Massachusetts Medical Center. Worcester Foundation Campus Clare Waterman-Storer, University of North Carolina. Chapel Hill Course Coordinator Frederick Miller, University of Massachusetts Medical Center, Worcester Foundation Campus Students Julia Avery. Yale University Margaret Clarke. Oklahoma Medical Research Foundation Carol Cogswell. University of Sydney, Australia David Ceilings, North Carolina State University Rossella Conti. Brandeis University Carol Gregorio. University of Arizona. Tucson Alexey Khodjakov, Wadsworth Center for Labs and Research Helmut Knapp. Swiss Federal Institute of Technology. Switzerland Stephen Lambert. University of Massachusetts Medical Center, Worcester Foundation Campus Jeannie Lee. Whitehead Institute Edwin Levitan, University of Pittsburgh Ilona Linnoila. National Cancer Institute Frank Macaluso, Albert Einstein College of Medicine Richard MacDonald, Massachusetts General Hospital Ivana Novak. August Krogh Institute. Denmark Joan Packenham, National Institute of Environmental Health Sciences Brigitta Peteri-Brunback. Astra Hassle AB. Sweden Jonathan Pines, The Wellcome Trust/Cancer Research Campaign. United Kingdom Guillermo Romero. University of Pittsburgh Ichiro Sase, Laboratory of Molecular Biophotonics. Japan Shuji Toyonaga. Laboratory of Molecular Biophotonics, Japan Shlomo Trachtenberg. National Institutes of Health Steven Treistman, University of Massachusetts Medical Center Stefan Wilhelm. Carl Zeiss, Inc., Germany Ginger Withers, University of Virginia Steven Wooding. Medical Research Council, United Kingdom Ji-Hong Zang, Stanford University Julianne Zimmerman. Payload Systems, Inc. Medical Informatics (June 1-June 8) Director Daniel Masys. University of California. San Diego Faculty Paul Clayton. Columbia University James Grigsby, University of Colorado Health Sciences Center George Hripcsak. Columbia-Presbyterian Medical Center Stephen Johnson. Columbia-Presbyterian Medical Center Lawrence Kingsland. National Library of Medicine David Landsman. National Library of Medicine Donald D.A.B. Lindberg. National Library of Medicine Richard Rodgers, National Library of Medicine Jay Sanders, Global Telemedicine Group Students Michael Ashburn. University of Utah Lynda Baker, Wayne State University Janet Bangma. East Carolina University Health Sciences Library Stephen Bartold, Texas Tech University Health Science Center David Boilard. Medical College of Ohio Sekum Boni-Awotwi. Howard University Neil Busis, University of Pittsburgh Laura Cailloux, Northwest Portland Area Indian Health Board Thomas Clay, East Carolina University School of Medicine May Dobal. Wayne State University Janice Dreyer, Community Hospitals of Central California Kathel Dunn. New York Academy of Medicine David George. Reading Hospital & Medical Center Gale Hannigan, Texas A&M University Health Science Center Conor Heneghan, New York Eye & Ear Infirmary Ruth Hoist. Columbia Hospital Arthur Huntley. University of California. Davis Charles Jaffe, American College of Allergy Juan Larach. Mission Neighborhood Health Center Clare Leibfarth, Doctors Hospital of Stark County Ethel Madden, Ochsner Medical Institutions Faith Meakin, University of Florida Peter Pevonka, University of Florida Health Science Center Ted Rosenkrantz, University of Connecticut Health Center Floyd Russell, West Virginia University Gerald Segal, New York State Psychiatric Institute Marcus Simpson, George Washington University Elaine Trzebiatowski, Allina Health System Amy Warner, University of Michigan Ken Wolf, Charles R. Drew University of Medicine & Science Methods in Computational Nenroscience (August 3 -Au gust 30) Directors David Kleinfeld, University of California, San Diego David Tank. Bell Laboratories/Lucent Technologies Faculty Lawrence Abbott, Brandeis University Moshe Abeles, Hebrew University of Jerusalem. Israel Richard Andersen. California Institute of Technology Robert Barlow, State LIniversity of New York Health Science Center William Bialek, NEC Research Institute Thomas Collett, University of Sussex. United Kingdom Rob de Ruyter van Steveninck. NEC Research Institute Kerry Delaney, Simon Fraser University. Canada Bard Ermentrout, University of Pittsburgh David Hansel, Ecole Polytechnique, France John Hopfield, Princeton University Daniel Johnston. Baylor College of Medicine Nancy Kopell. Boston University Kevan Martin, Swiss Federal Institute of Technology, Switzerland Kevin Martin. University of North Carolina. Chapel Hill David McCormick, Yale University School of Medicine Wolfram Schultz. Universite de Fribourg. Switzerland Terrance Sejnowski, Salk Institute R28 Annual Report H. Sebastian Seung, Bell Laboratories Arthur Sherman, National Institutes of Health Karen Sigvardt, University of California, Davis Frederick Sigworth, Yale University School of Medicine Haim Sompolinsky. Hebrew University of Jerusalem. Israel Mircea Steriade, Universite Laval, Canada David Terman. Ohio State University Naftali Tishby, Hebrew University of Jerusalem, Israel Misha Tsodyks, The Weizmann Institute, Israel Steven Zucker, Yale University Lab Instructors Roderick Jensen, Wesleyan University Terrance Kovacs, Bell Laboratories John White, Boston University Course Assistant Joseph Paton, Tufts University Students Thomas Adelman. Cornell University Yoram Ben Shaul, Hebrew University Hadassah School of Medicine, Israel Karl Deisseroth, Stanford University Reiko Maki Fitzsimonds. University of California. San Diego Galit Fuhrmann, Hebrew University of Jerusalem, Israel Mark Goldman, Harvard University Audrey Guzik, Allegheny University Richard Hahnloser, Swiss Federal Institute of Technology, Switzerland Robert Harris, University of Cambridge, United Kingdom Katrina MacLeod. California Institute of Technology William Miller, Brandeis University Jonathan Murnick, Harvard University Medical School Louis Neltner, Ecole Polytechnique, France Duane Nykamp, New York University Anitha Pasupathy, Johns Hopkins University Sharon Peled, Massachusetts Institute of Technology Daniel Reich, Rockefeller University Mark Schnitzer, Princeton University Oren Shriki, Hebrew University of Jerusalem, Israel Jonathan Simon. University of Maryland William Vinje. University of California, Berkeley Martin Wainwright, Harvard University James Zackheim, Rutgers University Microinjection Techniques in Cell Biology (May 20-May 27) Director Robert B. Silver, Marine Biological Laboratory Faculty Suzanne Klaessig. Cornell University Douglas Kline. Kent State University Gernot Presting, Institute of Plant Breeding, Germany Eric Shelden, University of Michigan Course Assistant Lisa Mehlmann, University of Connecticut Health Center Students Annie Borgne, CNRS. France Tao Cheng, Massachusetts General Hospital Goffredo Cognetti, University of Palermo, Italy Maxim Dorovkov, University of Medicine and Dentistry of New Jersey/Robert Wood Johnson Medical School William Faught. Medical University of South Carolina Jennifer Hartt, University of Pennsylvania Lorayne Jenkins, Wyeth Ayerst Research Kathleen Jensen, U.S. Environmental Protection Agency Shuxian Jiang. Harvard Institutes of Medicine David Lagunoff. St. Louis University Yoshiaki Omura, Heart Disease Research Foundation Stelios Papaioannou. United Medical and Dental School of Guy's and St. Thomas' Hospitals. United Kingdom May Penarroyo, De La Salle University, Philippines Gloria Perez. Massachusetts General Hospital Margherita Randazzo. University of Palermo. Italy Vasantha Reddi, University of Virginia Carol Reinisch, Tufts University School of Veterinary Medicine Nancy Searby, NASA Ames Research Center/Stanford University Jennifer Stephens. University of Oklahoma Nirmala SundarRaj, University of Pittsburgh Molecular Mycology: Current Approaches to Fungal Pathogenesis (August 10-Augmt 30) Directors John Edwards, Jr., Harbor-UCLA Medical Center Paul T. Magee, University of Minnesota Aaron P. Mitchell, Columbia University Lecturers Arturo Casadevall, Albert Einstein College of Medicine Gary T. Cole, Medical College of Ohio Brendan Cormack, Stanford University School of Medicine Jim Cutler. Montana State University Judith Domer, Tulane Medical School Scott Filler, Harbor-UCLA Medical Center Gerald Fink, Whitehead Institute of Biomedical Research Carol A. Kumamoto, Tufts University School of Medicine Myra Kurtz, Merck Research Laboratories June Kwon-Chung, National Institutes of Health John McCusker, Duke University Medical Center John Perfect, Duke University School of Medicine Michael A. Pfaller, University of Iowa College of Medicine Judith Rhodes. University of Cincinnati Stewart Scherer, University of Minnesota School of Medicine Brian Wong. Veterans Administration Connecticut Healthcare System Teaching Assistants Janna Beckerman, University of Minnesota Maria Soushko, Columbia University Yang Xiao. Columbia University Educational Programs R29 Course Assistant Joseph Paton. Tufts University Students James Alspaugh, Duke University Kent Buchanan, University of Oklahoma Health Sciences Center Enda Clarke, Royal Postgraduate Medical School, United Kingdom Marianne De Backer, Janssen Research Foundation. Belgium Maurizio Del Poeta, Duke University Medical Center Tamara Doering, Cornell University Medical School Roy Hopfer, University of North Carolina. Chapel Hill Margaret Hosteller. University of Minnesota Linda Janusek, Loyola University of Chicago Herbert Mathews, Loyola University of Chicago Frank-Michael Miiller, National Cancer Institute Norbert Schnell, Zeneca Pharmaceuticals, United Kingdom Craig Thompson, Scriptgen Pharmaceuticals Jo-Anne van Burik. Fred Hutchinson Cancer Research Center Paul Verweij. University Hospital Nymegen. The Netherlands Mason Zhang, University of Nevada School of Medicine Neurobiology & Development of the Leech (August 9-August 30) Directors Pierre Drapeau. McGill University, Canada Martin Shankland. University of Texas, Austin Faculty Andreas Baader, University of Zurich. Switzerland Irmgard Dietzel-Meyer. Ruhr-Universitat Bochum, Germany Francisco Fernandez de Miguel. Universidad Nacional Autonoma, Mexico John Jellies. Western Michigan University Jorgen Johansen. Iowa State University Anna Kleinhaus. New York Medical College Eduardo Macagno, Columbia University Mark Martindale. University of Chicago Barbara Modney. Cleveland State University Kenneth Muller, University of Miami School of Medicine John Nicholls, University of Basel, Switzerland Christine Sahley, Purdue University Catherine Wedeen, New York Medical College David Weisblat, University of California Course Assistant Gabrielle Tomasky. Marine Biological Laboratory Students Istvan Albert, University of Notre Dame Dianne Allen, Louisiana State University Medical Center Maria Casanueva. Catholic University. Chile James Einum. Marquette University Chunta Jie. Iowa State University Lynette Nguyen. Smith-Kettlewell Eye Research Institute Elizabeth Perruccio, New York Medical College Aloysius Phillips. Columbia University Giulietta Pinato, International School for Advanced Studies, Italy Subhabrata Sanyal, Tata Institute of Fundamental Research, India Daniel Shain. University of California, Berkeley Patrick Wigge, Medical Research Council. United Kingdom Optical Microscopy and Imaging in the Biomedical Sciences (October 8-October 16) Director Colin Izzard. State University of New York. Albany Facitlt\ Joseph DePasquale, New York State Department of Health Robert Hard, State University of New York, Buffalo Brian Herman, University of North Carolina, Chapel Hill Shahid Khan. Albert Einstein College of Medicine Frederick Maxrield, Cornell University Medical College John Murray, University of Pennsylvania David M. Piston. Vanderbilt University Kenneth Spring. National Institutes of Health Jason Swedlow. Harvard University Medical School Teaching Assistants Ken Dunn. Indiana University Medical Center Lynda Pierini. Cornell University Medical College Wade Sigurdson. State University of New York, Buffalo Elizabeth Welnhofer, State University of New York, Buffalo Lecturers Jan Hinsch, Leica, Inc. Shinya Inoue, Marine Biological Laboratory H. Ernst Keller, Carl Zeiss. Inc. Rudolf Oldenbourg. Marine Biological Laboratory Martin Scott, Consultant in Scientific Imaging Course Assistant Kari Lavalli, Marine Biological Laboratory Students Diana Bartelt, St. John's University Michelle Burack, University of Virginia Dorthe Christensen, Biolmage, Denmark Conan Cooper, University of Calgary, Canada John Crocker. University of Pennsylvania Leanne Delbridge. University of Melbourne, Australia Mark Drew, Johns Hopskins University Tracey du Laney, University of North Carolina, Chapel Hill Lars Hansen, Hagedorn Research Institute, Denmark Brian Helmke, University of Pennsylvania Robert Hughes, University of Washington Eleanor Kable. Sydney University, Australia Peter Kaplan, University of Pennsylvania Timothy King. University of Texas Thomas Kinraide. United States Department of Agriculture David Marcey. Kenyon College Susanne Pedersen, University of Copenhagen, Denmark Marli Robertson, University of Calgary, Canada Pascal Stein, Harvard Medical School David Swift, University of Pennsylvania James Thomson. University of Wisconsin Sebastian Tille, Carl Zeiss Jena. Germany Mariko Tokito, University of Pennsylvania Frances Wang, National Institute of Standards & Technology Karen Zito. University of California. Berkeley R30 Annual Report Pathogenesis of Neuroimmunologic Diseases (August 17- August 29) Directors Celia F. Brosnan. Albert Einstein College of Medicine Jack Rosenbluth, New York University School of Medicine Faculty Barbara Barres, Stanford University School of Medicine Etty Benveniste, University of Alabama, Birmingham Joan Berman, Albert Einstein College of Medicine Peter Charles. Albert Einstein College of Medicine Patricia Coyle, State University of New York, Stony Brook Robert Darnell. Rockefeller University Judah Denburg. McMaster University, Canada David Felten, University of Rochester Robert M. Gould. New York State Institute of Basic Research Diane Griffin. Johns Hopkins University John Griffin, Johns Hopkins School of Medicine William Hickey. Dartmouth-Hitchcock Medical Center Gilla Kaplan. Rockefeller University Paul Knopf, Brown University Vijay Kuchroo, Brigham and Women's Hospital Jon Lindstrom. University of Pennsylvania School of Medicine James Martiney, Picower Institute for Medical Research Steven Pteifter, University of Connecticut Health Center Richard Ransohoff, Cleveland Clinic Foundation Bruce Ransom, University of Washington School of Medicine Anthony Reder. University of Chicago J. M. Ritchie, Yale University School of Medicine James Salzer, New York University Medical Center Clifford Saper, Beth Israel Hospital Moon Shin, University of Maryland, Baltimore Michele Solimena, Yale University Esther Sternberg, National Institutes of Health J. Wayne Streilein, Scheppens Eye Research Institute Byron Waksman. Foundation for Microbiology Students Amit Bar-Or, Massachusetts General Hospital Rita Baron-Faust. WCBS NewsRadio 88 Alexei Boiko. Russian State Medical University, Russia Emanuela Bonfoco, La Jolla Institute of Immunology Laurent Coscoy, Pasteur Institute, France Benedicte Dubois. Rega Institute, Belgium Urszula Fiszer, Institute of Psychiatry & Neurology. Poland Glen Greenough. Dartmouth-Hitchcock Medical Center Carolyn Hoban, Cambridge Neuroscience Inc. Liwei Hua. Albert Einstein College of Medicine Kee-Ching Jeng, Taichung Veterans General Hospital, Taiwan Pitagoras Justino. Universidade Federal de Uberlandia. Brasil Bernd Kieseier. University of Wurzburg. Germany Pia Kivisakk, Karolinska Institute, Sweden Dmitriy Labunsky. Institute of Neurology, Russia Igor Leykin, Weizmann Institute of Science, Israel Carrie McManus, Albert Einstein College of Medicine Mary McMenamin. Oxford University. United Kingdom Neelufar Mozaffarian, Albert Einstein College of Medicine Marcin Mycko, Medical Academy of Lodz, Poland Jitendra Patel, Zeneca Pharmaceuticals Barry Singer. New York Hospital-Cornell Medical Center Nevil Singh. Tata Institute of Fundamental Research, India Sulpicio Soriano. Children's Hospital/Harvard Medical School Workshop on Molecular Evolution (August 3-August 15) Directors Daniel B. Davison. Bristol-Myers Squibb PR1 Mitchell Sogin, Marine Biological Laboratory Faculty W. Ford Doolittle. Dalhousie University, Canada Douglas Eernisse. California State University Joseph Felsenstein. University of Washington Michael Gray. Dalhousie University, Canada Robert Haselkorn, University of Chicago David Hillis. University of Texas Mike Holder, University of Houston Richard Hudson, University of California, Irvine Thomas Kaufman, Indiana University David Maddison. University of Arizona, Tucson Geoffrey McFadden, University of Melbourne, Australia Lynn Miller, Genetics Computer Group, Inc. Michael Miyamoto, University of Florida Educational Programs R31 Rasmus Nielsen. University of California. Berkeley Gary Olsen, University of Illinois William Pearson, University of Virginia Kevin Peterson, California Institute of Technology David Roos. University of Pennsylvania Pamela Soltis, Washington State University David Swofford. Smithsonian Institution John Wakeley, Nelson Biological Labs Bruce Walsh, University of Arizona, Tucson Sam Ward. University of Arizona. Tucson Carl Woese, University of Illinois, Urbana-Champaign Teaching Assistant Steven Thompson. Washington State University Course Assistant Udeni Amit. Marine Biological Laboratory Students Michael Alfaro. University of Chicago Field Museum of National History John Archibald, Dalhousie University, Canada Pamela Arnofsky, Woods Hole Oceanographic Institution Vijay Aswani, Smithsonian Tropical Research Institute Andrew Baker, University of Miami Anne Bansemir, Rutgers University Elizabeth Barratt, Zoological Society of London, United Kingdom Mary Bateson, Montana State University Christiane Biermann. State University of New York, Stony Brook Carrine Blank, University of California, Berkeley Lisa Borghesi. Oklahoma Medical Research Foundation Adriana Briscoe, Harvard University Daniel Brumbaugh. University of Texas, Austin Nina Brunner. University of Essen, Germany Carmen Cadilla, University of Puerto Rico Susan Chien, University of Florida Manuela Coelho. University of Lisbon, Portugal Chris Conroy, University of Alaska Museum Colomban de Vargas. University of Geneva. Switzerland Joel Dore. Institut National de la Recherche Agronomique, France Mark Dorssi, University of Edinburgh, Scotland Amy Driskell. University of Chicago Field Museum of National History Lmdsey Dubb. University of Washington Mary Eubanks. Duke University Jeffry Fasick, University of Maryland Baltimore County Oisin Feeley, Dalhousie University. Canada Victor Fet. Marshall University Eric Gaidos. California Institute of Technology Jet Propulsion Laboratory/Woods Hole Oceanographic Institution Regme GroBkopf, Max-Planck-Institut, Germany Gabriel Gutierrez, University of Seville. Spain Malin Heldtander. National Veterinary Institute, Sweden Hiromi Imamichi, SAIC Frederic Alex Jeffries. National Research Council, Canada Dana Jones, Centers for Disease Control and Prevention Marie-Josee LaForest, University of Montreal, Canada Kirsten Lindstrom, University of California, Berkeley Jaw-Ching Liu. University of Texas, Houston Lei Liu, University of Connecticut Frieder Mayer, University of Erlangen, Germany Damhnait McHugh. Harvard University Kirsten Nicholson. University of Miami Link Olson, University of Chicago Field Museum of National History Bertil Pettersson. Royal Institute of Technology. Sweden Mary Poss, University of Washington Linda Prince. University of North Carolina. Chapel Hill Anne Marie Quinn. Yale University Patrick Reynolds. Hamilton College Frank Rosenzweig. University of Idaho Marco Salemi. Rega Institute. Belgium Andrew Salywon, Arizona State University Nikolaos Schizas, University of South Carolina Alastair Simpson, University of Sydney, Australia James Robert Stevens, University of Bristol. United Kingdom Suzanne Sukhdeo. Rutgers University Jin Gou Tone. Chinese University, Hong Kong Catherine Walton. University of Leeds, United Kingdom Hui Wang, University of Houston Jacqueline Weicker. University of Alaska. Fairbanks Connie Westhoff, University of Nebraska Adrian Whatmore. University of Warwick. United Kingdom Kenneth Wurdack. University of North Carolina, Chapel Hill Other Programs Semester in Environmental Science (September 8-December 19) Directors Jerry M. Melillo, Director Kenneth H. Foreman. Associate Director Faculty Charles Hopkinson. Aquatic Course Director Knute Nadelhoffer, Terrestrial Course Director John Hobbie. Microbial Ecology Edward Rastetter, Mathematical Modeling Anne Giblin Linda Deegan Bruce Peterson Christopher Neill Joe Vallino Mathew Williams R32 Annual Report Paul Steadier Peter Siver, Faculty Fellow. Connecticut College Research Assistants Jeffrey Hughes James Laundre Jane Tucker Lori Soucy Kristin Tholke Kathleen Regan Beth Hooker John Helfnch Martha Downs Amy Nolin Kathleen Newkirk Neil Bettez Teaching Assistants Martha Peterson Michele Bahr Patricia Micks Bonnie Kwiatkowski Robert Garritt Nat Weston Students Toby Ahrens, Connecticut College Hyacinth Armstrong. Mt. Holyoke College Noah Bleich. Brandeis University Abbey DeRocker. Bates College Lynn Diener, Bard College Janice Glass, Lafayette College Sarah Jackson, Connecticut College Samuel Kelsey, Dickinson College Christy Meredith. Allegheny College Sophie Parker. Wellesley College Stephanie Parker, Middlebury College Rachel Poretsky. Brandeis University Shana Rapoport, Brandeis University Amy Townsend-Small, Skidmore College Marlene Tsie. Brandeis University Rebecca Weidman, Carleton College Teachers' Workshop: Living in the Microbial World (August 17 -August 23) Course Director Lorraine Olendzenski, University of Connecticut Course Assistant Andy Heaford, University of Connecticut Presenters Lynn Margulis, University of Massachusetts Ricardo Guerrero, University of Barcelona, Spain Steve Goodwin. University of Massachusetts, Amherst Robert Bullis, Marine Biological Laboratory Holger Jannasch. Woods Hole Oceanographic Institute Greg Hinkle, University of Massachusetts, Dartmouth Art Girard, Pfizer Central Research. Groton. CT Norman Wainwright, Marine Biological Laboratory Adrian Smith. Marine Biological Laboratory Participants Charles Anastasia, Mashpee High School Diane E. Arnold, Bennie Dover Jackson Middle School Catherine Baker, East Junior High School Marcia A. Benvenuti. Bennie Dover Jackson Middle School Florence Berdan, Parsippany Hills High School Wesley Blauss, Indian Head Middle School William Cerino, Lyme-Old Lyme High School Paul J. Chamberlin. Nauset Regional High School Elizabeth Check. Attleboro High School Howard C. Estes, East Junior High School Maryrose L. Flynn, Indian Head Middle School Marsha R. Folger. Lyme-Old Lyme High School Mary Johnson. Parsippany Hills High School. Brooklawn Middle School Lucy Lupinacci. Griswold Intermediate School Joreen Mattras. Griswold Intermediate School Sheila E. McTigue, Lyme-Old Lyme High School M. Susan O'Donnell, Indian Head Middle School Jane Shute, Indian Head Middle School Judith J. Trotta, Nauset Regional High School James Watson, East Junior High School Summer Research Programs Principal Investigators Armstrong, Clay, University of Pennsylvania Armstrong, Peter B., University of California, Davis Augustine, George J., Duke University Medical Center Barbieri, Elena, Marine Biological Laboratory Barlow, Jr.. Robert B., State University of New York Health Science Center Beauge, Luis, Instituto M. y M. Ferreyra, Argentina Bennett, Michael V. L., Albert Einstein College of Medicine Berlin, Joshua, Bockus Research Institute Bloom, George, University of Texas Southwestern Medical Center Bodznick, David, Wesleyan University Boron. Walter F., Yale University Medical School Borst, David, Illinois State University Boyer, Barbara, Union College Brady, Scott T., The University of Texas Southwestern Medical Center, Dallas Brown, Joel E.. Albert Einstein College of Medicine Browne. Carole, Wake Forest University Burger, Max M., Friedrich Miescher Institute, Switzerland Burgos. Mario. Universidad Nacional de Cuyo-Conicet, Argentina Cardell, Robert R., University of Cincinnati Casagrand, Janet, University of Colorado Chappell. Richard L., Hunter College, City University of New York Cohen. Lawrence B., Yale University School of Medicine Cohen, William D., Hunter College, City University of New York Corwin, Jeffrey, University of Virginia De Weer, Paul, University of Pennsylvania School of Medicine DeMont, Edwin, St. Francis Xavier University, Canada Devlin, Leah, Penn State University DiPolo. Reinaldo. IVIC, Venezuela Eckberg, William, Howard University Edwards, Donald, Georgia State University Ehrlich, Barbara, University of Connecticut Health Center Fay, Richard, Loyola University of Chicago Fishman. Harvey M., The University of Texas Medical Branch, Galveston Gadsby, David, The Rockefeller University Garcia-Blanco, Mariano, Duke University Medical Center Garrick, Rita Anne. Fordham University College. Lincoln Center Giuditta. Antonio, University of Naples, Italy Goldman, Robert D., Northwestern University Medical School Gould. Robert. New York State Institute for Basic Research in Developmental Disabilities Gray, John, Queen's University, Canada Groden, Joanna, University of Cincinnati Gundersen, Gregg. Columbia University Haimo, Leah, University of California, Riverside Halstead, Matthew, University of Auckland, New Zealand Henry, Jonathan J.. University of Illinois Hershko, Avram. Technion, Israel Highstein, Steven M., Washington University School of Medicine Hines, Michael, Yale University School of Medicine Holz. George, Massachusetts General Hospital Hoskin, Francis, US Army Natick RTJ&E Center Howze, Gwendolyn, Texas Southern University Ip, Wallace, University of Cincinnati Johnston, Daniel, Baylor College of Medicine Joye, Samantha, Texas A&M University Kaczmarek, Leonard, Yale University School of Medicine Kaplan, Ilene M., Union College Khan, Shahid, Albert Einstein College of Medicine Khodakhah, Kamran. University of Pennsylvania Klerkx, J.H.E.M., Utrecht University, The Netherlands Kravitz, Edward, Harvard Medical School Kuhns, William, The Hospital for Sick Children. Canada Later, Eileen M.. University of Texas Health Science Center Landowne. David, University of Miami School of Medicine Langford, George, Dartmouth College Laskin. Jeffrey. University of Medicine and Dentistry of New Jersey Layne. John. Duke University Marine Laboratory Levandoski, Mark, Brown University Lipicky, Raymond J., Food and Drug Administration Llinas, Rodolfo R., New York University Medical Center Lovett, Donald, The College of New Jersey Martmdale. Mark. University of Chicago McNeil, Paul. Medical College of Georgia Mensmger. Allen. Washington University School of Medicine Metuzals, Janis, University of Ottawa Faculty of Medicine, Canada Minkoff, Charles, Duke University Medical Center Miyakawa, Hiroyoshi, Tokyo University of Pharmacy and Life Science, Japan R33 R34 Annual Report Moore. John. Duke University Medical Center Nasi. Enrico. Boston University School of Medicine Nguyen, Quoc Thang, University of California. Irvine Palazzo. Robert. University of Kansas Palma, Eleanora. Regina Elina Center Research Institute. Italy Pant, Harish, National Institutes of Health Parysek, Linda. University of Cincinnati Pixley, Sarah, University of Cincinnati Puga. Alvaro. University of Cincinnati Quigley, James P., State University of New York, Stony Brook Rakowski, Robert F., Finch University of Health Sciences/The Chicago Medical School Rasmussen. Howard, Medical College of Georgia Ratner, Nancy, University of Cincinnati Reese. Thomas S., National Institutes of Health Rieder. Conly. Wadsworth Center Ripps, Harris, University of Illinois College of Medicine Rome, Larry. University of Pennsylvania Rosenbluth. Jack. New York University Medical Center Ross, William. New York Medical College Ruderman, Joan V.. Harvard Medical School Russell, John M.. Medical College of Pennsylvania Saito. Takehito. University of Tsukuba, Japan Schweizer, Felix, Duke University Medical Center Shashar, Nadav, Marine Biological Laboratory Sheetz, Michael, Duke University Medical Center Simpson, Alastair, University of Sydney. Australia Siver, Peter. Connecticut College Sloboda. Roger D., Dartmouth College Spann, Timothy, Northwestern University Medical School Spiegel, Evelyn, Dartmouth College Spiegel. Melvin. Dartmouth College Standart, Nancy, University of Cambridge. United Kingdom Steffen. Walter. Institute of Biochemistry and Molecular Cell Biology. Austria Stuart, Ann E.. University of North Carolina, Chapel Hill Sugimon. Mutuysuki. New York University Medical Center Suszkiw, Janusz, University of Cincinnati Takahashi, Megumi, Kanagawa Psychiatric Center, Japan Telzer, Bruce, Pomona College Tokumam, Hiroshi, Duke University Medical Center Tran, Phong, University of North Carolina, Chapel Hill Troll, Walter, New York University Medical Center Tytell, Michael, Bowman Gray School of Medicine, Wake Forest University Wachowiak, Matt, University of California, Berkeley Weil, E. Jennifer. Marine Biological Laboratory Whittaker, J. Richard. University of New Brunswick. Canada Wicklein. Martina. University of Ari/ona Zigman. Seymour, University of Rochester Medical School Zottoli. Steven. Williams College Zukin, R. Suzanne. Albert Einstein College of Medicine Other Research Personnel Abe, Terno. Niigata University, Japan Altamirano, Anibal, University of Buenos Aires. Argentina Anderson. Erik, St. Francis Xavier University, Canada Antic. Srdjan, Yale University School of Medicine Araneda, Ricardo, Albert Einstein College of Medicine Armstrong, Clara. University of Pennsylvania Baikie, lain. Robert Gordon University. United Kingdom Barrera. Jose, New York University Medical Center Bearer, Elaine, Brown University Becker, Julie, Temple University Bezanilla. Francisco. University of California. Los Angeles Bobb. David, Huston-Tillotson College Boyle. Kim-Laura, Colby-Sawyer College Brackenbury. Robert, University of Cincinnati Breitwieser, Gerda E., Johns Hopkins School of Medicine Brown, Euan. Marine Biological Association. United Kingdom Burris, Jennifer. Carleton College Carnall, Nicola. University of Cambridge. United Kingdom Carvan. Michael. University of Cincinnati Christman. Emily, Goucher College Cimini. Ashley. Yale University Claessens, Luc. Marine Biological Laboratory Clay. John, National Institutes of Health Clifford. Patrick. College of New Jersey Connaughton, Martin, University of Pennsylvania Cooper, Gordon, Yale University School of Medicine Couch. Ernest. Texas Christian University Cserjesi, Peter. Columbia University Dadacay. Alma-Villa. Hunter College Davis. Bruce A.. Yale University DePma. Ana. Dartmouth College Dodge, Frederick. State University of New York Health Science Center Eddleman. Chris, University of Texas Medical Branch. Austin Edds-Walton. Peggy. University of California, Riverside Eyman, Maria, University of Naples, Italy Fang. Jing, Yale University School of Medicine Femstein, Douglas, Cornell University Fernandez-Busquets, Xavier, Friedrich Miescher Institut, Switzerland Flamarique, Inigo, University of Victoria, Canada Fukuda. Mitsunori, Tsukuba Life Science Center, Japan Fukui, Yoshio. Northwestern Medical School Summer Research R35 Gainer. Harold, National Institutes of Health Galbraith, James A., Duke University Gallant, Paul E., National Institutes of Health Gebhardt. Kelley, University of North Carolina, Chapel Hill Gerosa, Daniela, Friedrich Miescher Institute, Switzerland Goldman, Anne E., Northwestern University Medical School Gomez. Maria del Pilar, Boston University School of Medicine Goss. Thomas, Bancroft School Grant. Philip. National Institutes of Health Grassi, Daniel, Food and Drug Administration Gray, Richard. Baylor College of Medicine Grob, Marianne. Friedrich Miescher Institute. Switzerland Gross. Jeffrey, Duke University Medical College Hagar. Robert, University of Connecticut Health Center Hambe. Bjorn. Lund University, Sweden Hirata, Kazunari. New York University Medical Center Hogan, Emilia M., Yale University Medical School Holmgren. Miguel, Massachusetts General Hospital Horner, Michael, University of Gottingen, Germany Jarchow. Janina. Friedrich Miescher Institute, Switzerland Johannes, Eva. North Carolina State University Jonas, Elizabeth. Yale University School of Medicine Kaftan, Edward, University of Connecticut Health Center Kamino, Kohtaro, Tokyo Medical and Dental University, Japan Katz. Benjamin, Williams College Keyoung. Jinsoo, Albert Einstein College of Medicine Khater, Kevin, Chicago Medical School Khuon, Satya. Northwestern University Medical School Klimov, Andrei. University of Pennsylvania Koroleva, Zoya, Hunter College Koulen. Peter, Max-Planck-Institute for Brain Research. Germany Krasne, Frank, University of California, Los Angeles Krasne. Sally, University of California. Los Angeles Kummer. Rebecca, Williams College Lam, Ying-Wan, Yale University Lasser-Ross, Nechama. New York Medical College Lema-Foley, Christine. Hunter College Lew, Roger, York University, Canada Li, Lihong, Hunter College Linquist, Randall, Williams College Loboda, Andrey, University of Pennsylvania Ludin. Beat. Friedrich Miescher Institute, Switzerland Lyser. Katherine. Hunter College MacGillivray. Patrick, St. Francis Xavier University. Canada Major, Guy, University Laboratory of Physiology. United Kingdom Malchow, Robert, University of Illinois. Chicago Marks, Andrew, Mount Sinai School of Medicine Martys, Jayme, Columbia University McCleskey. Ed. Vollum Institute McSweeney, Maireade, Bowdoin College Meigs, Bridget, Millbrook School Melishchuk, Alexey, University of Pennsylvania Meyers, Jason, University of Virginia Mikhailov, Alexei, Columbia University Molyneaux. Bradley, Dartmouth College Morgan, Jennifer, Duke University Mulligan. Megan. Columbia University Musani, Farrah, Williams College Nakagawa. Masaya. New York University Medical Center Noailles, Pierre, Hunter College O'Connor, Vincent. Max-Planck-Institute for Brain Research. Germany Ogan. Jeffrey. Illinois State University Onigman. Timna, Brown University Parris, Tchaiko. Hunter College Pasman, Zvi, Duke University Medical Center Passaglia, Chris, State University of New York Health Science Center Patel, Leena, University of Illinois, Chicago Petersen, Jennifer, National Institutes of Health Petz, Anne, Wellesley College Piscopo, Stefania, University of Napoli. Italy Pivovarova. Natalia, National Institutes of Health Powers, Maureen, Vanderbilt University Pozzo-Miller. Lucas, National Institutes of Health Prasad, Kondury. University of Texas Health Science Center Quinn, Kerry. University of Connecticut Health Center Ramsey. John, Princeton University Ravula, Sharath, Texas A & M University Rebhun. Lionel, University of Virginia Rosenthal, Joshua, University of California, Los Angeles Ruta, Vanessa, Hunter College Schlemermeyer, Etha, Hunter College Schlief. Michelle, Hampshire College Searby, Nancy, NASA Ames Research Center Seyfarth, Emst-August, J. W. Goethe-Universitat, Germany Shaub, Amy, Illinois Wesleyan University Sheikh, Sarah, University of Edinburgh. United Kingdom Silva, Celso, Women's & Infants Hospital Simpson. Tracy, University of Hartford Singh, Satish. Yale University School of Medicine Smith. Benjamin, Southwestern University Smith, Cynthia R. Tufts University Smyers, Mark, University of Texas Sporhase-Eich. Ulrike. University of Gottingen, Germany Stenflo, Johan, Lund University, Sweden Stewart, Karen. Syracuse University Stockbridge. Norman, Food and Drug Administration Stokes, Darrell, Emory University Sudhot. Tom. University of Texas Southwestern Medical Center Suwa, Hiroshi, New York University Medical Center Swan. Justin. Wake Forest University Sweeney, Christina, Illinois Wesleyan University Terasaki. Mark, University of Connecticut Health Center Thomas. Abraham, Jester Center Tokumaru. Hiroshi. Duke University Medical Center Tokumaru, Keiko, Duke University Medical Center Twersky, Laura, St. Peters College Vasandani. Veena. University of Virginia Wagg, Jonathan. The Rockefeller University White. Benjamin. Yale University Williams, Tracey, Howard University Wollert. Torsten. University of Rostock, Germany Womack, Mary, Duke University Medical School Wu, Peter. Columbia University R36 Annual Report Zakevicius, Jane M., University of Illinois, Chicago Zavilowitz, Joseph, Albert Einstein College of Medicine Zecevic, Dejan, Yale University School of Medicine Zeki, Semir, University College London, United Kingdom Zhao. Jinhua, Yale University Zigman. Bunnie R.. University of Rochester Medical Center Library Readers Abbott, Jayne, Marine Research Inc. Adelberg, Edward A.. Yale University Ahmadjiian. Vernon, Clark University Alkon, Daniel, National Institutes of Health Allen. Garland E.. Washington University Alliegro, Mark, Louisiana State University Medical Center Anderson, Everett, Harvard Medical School Baccetti, Baccio, University of Siena, Italy Barry, Susan R., Mount Holyoke College Benjamin, Thomas L., Harvard Medical School Bentham, Dolores A., SPC Associates, New York Berne, Rosalyn W., University of Virginia Bernhard, Jeffrey D., University of Massachusetts Medical Center Bernheimer, Alan W., New York University Borgese, Thomas A., Lehman College Boyer, John, Union College Burgess, David, Pittsburgh, PA Campbell, Robert K., Ares Advanced Technology Candelas. Graciela C., University of Puerto Rico Cariello, Lucio, Stazione-Zoologica, Italy Chaet, A. B., University of West Florida Child. Frank M., Trinity College Clark, Arnold M.. University of Delaware Clark, Denise, University of New Brunswick Clark. Eloise E., Bowling Green State University Clarkson. Kenneth L., Bell Labs, Lucent Technologies Cogswell, Carol, University of Sydney, Australia Cohen, Seymour S., American Cancer Society Cohen, Leonard A., American Health Foundation Cohen, Yehuda, Hebrew University of Jerusalem, Israel Collier, Marjorie M., St. Peters College Comoglio, Paolo. Institute for Cancer Research, Turino, Italy Cooperstein, Shenvin J.. University of Connecticut Health Center Copeland, Eugene C., Woods Hole, MA Cowling, Vincent F., SUNY, Albany D'Allessio, Giuseppe, University of Naples deToledo-Morrell. Leyla, Rush Medical College Duncan, Thomas K., Nichols College Epstein, Herman T.. Brandeis University Feldman, Susan C., New Jersey Medical School Finch, Caleb E., University of Southern California Frenkel, Krystyna, NYU Medical Center Gehrke, Lee, Massachusetts Institute of Technology German, James L., New York Blood Center Goldman, Robert, Northwestern University Medical School Goldstein, Moise H., Johns Hopkins University Groden, Joanna, University of Cincinnati Gross. Paul, University of Virginia Grossman, Albert. NYU Medical School Gruner, John, Cephalon, Inc. Haimo, Leah, University of California Halvorson, Harlyn, University of Massachusetts, Dartmouth Herskovits, Theodore T.. Fordham University Hunter. Robert. Gartnaval Royal Hospital Ilan, Judith, Case Western Reserve University Inoue, Sadayuki. McGill University. Canada Issodorides. Marrietta, University of Athens, Greece Jacobson, Allan S.. University of Massachusetts Medical Center Josephson, Beth, Ocean Arks International Kaltenbach, Jane C.. Mount Holyoke College Kaminer. Benjamin, Boston University School of Medicine Kamino, Kohtaro. Tokyo Medical & Dental University, Japan Kaplan, Arnold, University of Illinois at Chicago Karlin. Arthur. Columbia University Kelly, Robert E., University of Illinois Keynan, Alex, Israel Academy of Sciences and Humanities King, Kenneth, Falmouth. MA Klein. Donald, Colorado State University Kramer, Fred R., Public Health Research Institute 1997 Library Room Readers Daniel Alkon National Institutes of Health Lucio Cariello Stazione-Zoolociga A. Dohrn A. Chaet University of Connecticut Paolo Comoglio Institute of Cancer Research, Turino, Italy Giuseppe D'Alessio University of Naples Robert Goldman Northwestern Univ. Medical School Harlyn Halvorson Marine Biological Laboratory Alex Keynan Israel Academy of Science Hans Laufer University of Connecticut Joe L. Martinez University of Texas, San Antonio Michael Rabinowitz Marine Biological Laboratory George Reynolds Princeton University Gerald Weissman NYU School of Medicine Summer Research R37 Laderman. Aimlee D.. Yale School of Forestry & Environmental Studies Laster. Leonard, University of Massachusetts Medical Center Laut'er, Hans, University of Connecticut Lee. John J.. City College of CUNY Leighton. Joseph, Aeron Biotechnology, Inc. Levy, Arthur L., St. Vincents Hospital of New York Lisman, John, Brandeis University Long, Carol, Allegheny University Lorand, Laszlo, Northwestern University Medical School Luckenbill-Edds, Louise, Ohio University MacNichol. Edward F.. Boston University School of Medicine Major. Guy. Oxford University Masland, Richard. Massachusetts General Hospital Martinez. Joe L., University of Texas, San Antonio Mauzerall, David, Rockefeller University Michaelson. James, MGH Cancer Center Miller, Daniel, Rockefeller University Mills. Eric L., Dalhousie University, Canada Mmkoff, Charles G.. Duke University Medical Center Mitchell. Ralph. Harvard University Mizell, Merle, Tulane University Nagel, Ronald L.. Albert Einstein Institute Narahashi, Toshio, Northwestern University Medical School Nathans, Jeremy. Johns Hopkins University Naugle, John, NASA Nicaise, Mari-Luz, Universite de Nice, France Nicaise, Ghislain, Universite de Nice, France Nickerson. Peter A., SUNY, Buffalo Pappas. George D., University of Illinois Pollen, Daniel A., University Massachusetts Medical Center Porter, Mary E., University of Minnesota Prusch, Robert D.. Gonzaga University Przybyszewski. Andrew W., University of Massachusetts Medical Center Rabinowitz, Michael. Marine Biological Laboratory Rafferty, Nancy S.. Northwestern University Rafferty, Keen, Northwestern University Reynolds, George, Princeton University Rosenbluth. Raja, Simon Fraser University Rosenkranz, Herbert S., University of Pittsburgh Ryan, Terence E.. Regeneron Pharmaceuticals Sanger, Joseph W., University of Pennsylvania Medical School Sanger. Jean M.. University of Pennsylvania Medical School Schauder. Rolf. University of Frankfurt, Germany Segal. Rosalind, Harvard Institute of Medicine Shepro, David. Boston LIniversity Microvascular Research Siwicki. Kathleen K., Swarthmore College Spector, Abraham, Columbia University Spotte, Stephen. University of Connecticut Sundquist. Eric. US Geological Survey Sweet, Frederick, Washington University School of Medicine Tilney, Lewis, University of Pennsylvania Trager, William, The Rockefeller University Tweedell. Kenyon S., University of Notre Dame Tykocinski, Mark L., Case Western Reserve University Van Holde, Kensal E., Oregon State University Walton, Alan J.. Cavendish Lab, Cambridge Univer