I I I
I !•••
1 »
,
- ,
.
.
I
;
. .
.
'
.
.
.
MBL
CONTENTS
No. 1, FUBRI ARV 1992
BEHAVIOR
Hermans, Colin O., and Richard A. Satterlie
Fast-strike feeding hehav ior in a pteropocl mollusk.
<-/ini/i: IniHiiiiiii Phipps
Wayne. Nancy L., and Gene D. Block
Effects of photoperiod and temperature on egg-lav- ing behavior in a marine mollusk. .\/il\siei californica
DEVELOPMENT AND REPRODUCTION
Amemiya, S., and R. B. Emlet
I lii' development and larval form of .111 ec liinoihu- rioid c'c lunoid, Astlienosoma ijiinin, revisiti-d
Ausio, Juan
I'm ilic.ilion and biochemical characterization of ihe mu lear sperm-specific proteins ol the- bivalve mol- \\isk.sAgl~iodesma saxicola and .M\iiliiiii'ini inil/nt/i ....
Blades-Eckelbarger, Pamela I., and Nancy H. Marcus The origin of conical vesicles and then role- in egg envelope formation in the "spiny" eggs of a calanoid copepod. Centropagei velificatus
Chandler. Resa M., Mary Beth Thomas, and Julian
P. S. Smith, III
The role of shell granules and accessory cells in eggshell formation in (.'.<m\'»luta pulrhra (Turbellaria, Ac oela)
Chia, Fu-Shiang, Ron Koss, Shauna Stevens, and Jeff
I. Goldberg
Isolation ol neurons of a nudibranch veliger ....
Holland, Linda Z., and Nicholas D. Holland
Early development in ihe lancelet (=amphioxus) Branchiostoma /inri/lac from sperm eniiv iliiough pronuc lear fusion: presence of vegetal pole plasm and lac k ol conspicuous ooplasmic segregation . .
Lee, Youn-Ho, and Victor D. Vacquier
The divergence of species-specific abalone sperm Ivsins is promoted bv positive Darwinian seleiiion
ECOLOGY AND EVOLUTION
Gil-Turnes, M. Sofia, and William Fenical
1-iubivos of HniiKinis aiiu'i'icaiiu* are proteiti-<l bv epibiotic bacteria
31
41
(ili
Williams-Howze, Judy, and Bruce C. Coull
Are temperature and photoperiod necessary cues for encystmi'iil in llic marine' bentlm harpacticoid copepod Hctt'iiiji^'llii^ I/HUH/ Coull?
GENERAL BIOLOGY
Jennings, Joseph B., Lester R. G. Cannon, and Adrian J. Hick
I lie nature and origin ol ihe epidei mal scales of Notodactylus handschini—3.n unusual temnocephahd turbellarian ectosymbiotic on crayfish from north- ern Queensland
Mangum, Charlotte P., James M. Colacino, and
Judith P. Grassle
Red blood cell oxygen binding in lapitclhd poly- chaetes .
PHYSIOLOGY
Singarajah, K. V., and F. I. Harosi
Visual cells and pigments in a demersal fish, the
blaik sea bass (O;//n</)jn//\ \lrinlu)
Tankersley, Richard A., and Ronald V. Dimock, Jr.
Quantitative anahsis ol (he- structure and function of the marsupial gills of (he freshwater mussel An-
i tiliii'intii .
117
129
135
145
RESEARCH NOTES
Feldgarden, Michael, and Philip O. Yund
Allorecognition in colonial marine invertebrates: does selection favor fusion wiih km. or fusion with self?
Rands, M. L., A. E. Douglas, B. C. Loughman, and
R. G. Ratcliffe
Avoidance of hypoxia in a cnidarian symbiosis In algal photosvnthclic o\\gen
1115 The Biological Bulletin Board
POETRY
Skinner, Dorothy M.. and John S. Cook
Carroll M. Williams
Mellon, Deforest, Jr.
How tin- axon got its tale
CONTENTS No. 2, APRIL 1992
165 167
Van Alstyne, Kathryn L., Chad R. Wylie, Valerie J.
Paul, and Karen Meyer
Antipredator defenses in tropical Pacific soft corals (Coelenterata: Alcyonacea). I. Sclerites as defenses against generalist carnivorous fishes 231
DEVELOPMENT AND REPRODUCTION
Hand, Cadet, and Kevin R. Uhlinger
The culture, sexual and asexual reproduction, and growth of the sea anemone Nematostella i'i'iieii\i\ 169 McEdward, Larry R.
Morphology and development of a unique type of pelagic larva in the starfish P/ennti'i /r\v7<//f<\ (Echi- nodermata: Asteroidea) 177
ECOLOGY AND EVOLUTION
Jeffries, William B., Harold K. Voris, and Sombat Poovachiranon
Age of the mangrove crab Sc\lla <n'mitn at coloni- sation by stalked barnacles of the genus ()itul(i\mi\ 188
Kim, Kiho, Walter M. Goldberg, and George T.
Taylor
Architectural and mechanical properties of the black coral skeleton (Coelenterata: Antipatharia): a com- parison of two species 195
Raimondi. Peter T.
Adult plasticity and rapid larval evolution in a re- cently isolated barnacle population 210
Shapiro, Daniel F.
Intercolony coordination of /ooid behavior and a
new class of pore plates in a marine brvo/oaii ... 221
NEUROBIOLOGY AND BEHAVIOR
Diaz-Miranda, Lucy, David A. Price, Michael J.
Greenberg, Terry D. Lee, Karen E. Doble. and Jose
E. Garcia-Arraras
Characterisation of two novel neuropeptides from
the sea cucumber Holotlniritt gluht'i 'mini 241
Mackie. G. O., C. E. Mills, and C. L. Singla
Giant axons and escape swimming in Eujilnl;iiiiii\ dunlapae (Ctenophora: Cydippida) 248
Saigusa, Masayuki
Phase shift of a tidal rhythm by light-dark cycles in
the semi-terrestrial crab Si'f-nruui /tntin/i 257
PHYSIOLOGY
Baker, S. M., and R. Mann
Effects of hypoxia and anoxia on larval settlement, juvenile growth, and juvenile survival of the oyster Crossostrfd I'lt^nuitt
Brown, A. Christine, and Nora B. Terwilliger Developmental changes in ionic and osmotic regu- lation in the Dungeness crab. C.auicr magintcr .... 270
Cronin, Thomas W.
Visual rhythms in stomatopod crustaceans observed
in the pseudopupil 278
No. 3, JUNE 1992
DEVELOPMENT AND REPRODUCTION
Fong. Peter P., and John S. Pearse
Evidence for a programmed circannual life cycle modulated by inc reasing da\ lengths in Xi'iinll/i^ Inn- i!;ro/fl(Polychaeta:Nereidae) from central California
Mita, Masatoshi, and Masaru Nakamura
Ultrastru< tural study of an endogenous energy sub- strate in spermatozoa of the sea urchin Hi'i>iicfiitn>tit\ juili In'ii nini\ 298
Rivcst, Brian R.
Studies on the- struc tine and function of the lar\al kicliic-\ complex of prosobranch gastropods 305
MARINE CELL BIOLOGY
Gates, Ruth D., Garen Baghdasarian, and Leonard Muscatine
Temperature stress causes host cell detachment in symbiotic cnidarians: implications tor coral bleach- ing 324
NEUROBIOLOGY AND BEHAVIOR
Mercier, A. Joffre, and Rune T. Russenes
Modulation of crayfish hearts 1>\ FMRFamide- velated peptides 333
CONTENTS
Kulk.it m, Gunderao K., and Milton Fingerman
Quantitative analysis by reverse phase high perfor- mance liquid chromatography of 5-hydroxytrypt- , innne in the central nervous system of the red swamp uavfish, Procambanti dnrkii 341
Page, Louise R.
New interpretation of a nudibranch central nervous system based on ultrastructural analysis of neuro- developinent in Mi-lih<- Ifminiti. I. Cerebral and vis- ceral loop ganglia 348
Page, Louise R.
New interpretation of a nudibranch central nervous system based on ultrastructural analysis of neuro- developim-iit in Mtlibeleonma. II. Pedal, pleural, and labial ganglia 366
PHYSIOLOGY
Bergles, Dwight, and Sidney Tamm
Control of cilia in the branchial basket of dona in- tr\tiiniln (Ascidacea) 382
Latz, Michael I., and James F. Case
Slow photic and chemical induction of biokmiines- cence in the midwater shrimp, Sergeste* unnlis Han- sen " 391
Fitt, W. K., and S. L. Coon
Evidence for ammonia as a natural cue for recruit- ment of oyster larvae to oyster beds in a Georgia salt marsh 401
Burton, Ronald S.
Proline synthesis during osmotic stress in megalopa stage larvae of the blue crab, Callinectes wpulm . . 409
Combs, Christian A... Nicole Alford, Angela Boynton,
Mark Dvornak, and Raymond P. Henry
Behavioral regulation of hemolvmph osmolarity through selective drinking in land crabs, Birgus Intro and Gecarcoidea lalanrlii 416
Ellers, Olaf, and Malcolm Telford
Causes and consequences of fluctuating coelomic pressure in sea urchins 424
Kraus, David W., Jeannette E. Doeller, and Jonathan
B. Wittenberg
Hydrogen sulfide reduction of symbiont cytochrome
<'552 i" gills of .S'n/ciww ri'nli (Mollusca) 435
Wilmot, David B., and Russell D. Vetter
Oxygen- and nitrogen-dependent sulfur metabolism
in the thiotrophic clam Snlemw reidi 444
VIEWS AND DISCUSSION
Grosberg, Richard K.
To thine own self be true? An addendum to Feld- garden and Yund's report on fusion and the evo- lution of allorecognition in colonial marine inver- tebrates 454
Yund, Philip O., and Michael Feldgarden
To thine own self be true? Yes! Thou canst not then
be false to any other. A reply to Grosberg 458
Index to Volume 182 . 460
Volume 182
THE
Number 1
BIOLOGICAL BULLETIN
fvlarine Biological Lat/uratory LIBRARY
MAR 3 irtZ
Woods Hole, Mass.
FEBRUARY, 1992
Published by the Marine Biological Laboratory
Marine Biological Laboratory LIBRARY
MAR 3 1992 Woods Hole, Mass.
THE
BIOLOGICAL BULLETIN
PUBLISHED BY THE MARINE BIOLOGICAL LABORATORY
Associate Editors
PETER A. V. ANDERSON, The Whitney Laboratory, University of Florida
DAVID EPEL, Hopkins Marine Station, Stanford University
J. MALCOLM SHICK, University of Maine, Orono
Editorial Board
DAPHNE GAIL FAUTIN, University of Kansas
WILLIAM F. GILLY, Hopkins Marine Station, Stanford University
K. RANGA RAO, University of West Florida STEVEN VOGEL, Duke University
Editor: MICHAEL J. GREENBERG. The Whitney Laboratory, University of Florida Managing Editor: PAMELA L. CLAPP, Marine Biological Laboratory
FEBRUARY, 1992
Printed and Issued by LANCASTER PRESS, Inc.
PRINCE & LEMON STS. LANCASTER, PA
THE BIOLOGICAL BULLETIN
THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, MBL Street, Woods Hole. Massachusetts 02543.
Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BUL- LETIN. Marine Biological Laboratory, Woods Hole, Massachusetts 02543. Single numbers, $30.00. Sub- scription per volume (three issues). $77.50 ($155.00 per year for six issues).
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, Woods Hole, Massachusetts 02543. Telephone: (508) 548-3705. ext. 428. FAX: 508-540-6902. E-mail: pamcl@hoh.mbl.edu.
POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN. Marine Biological Laboratory,
Woods Hole, MA 02543.
Copyright © 1992, by the Marine Biological Laboratory
Second-class postage paid at Woods Hole, MA, and additional mailing offices.
ISSN 0006-3 185
INSTRUCTIONS TO AUTHORS
The Biological Bulletin accepts outstanding original research reports of general interest to biologists throughout the world. Papers are usually of intermediate length (10-40 manuscript pages). A limited number of solicited review papers may be ac- cepted after formal review. A paper will usually appear within four months after its acceptance.
Very short, especially topical papers (less than 9 manuscript pages including tables, figures, and bibliography) will be pub- lished in a separate section entitled "Research Notes." A Re- search 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 Bulletin. Materials and Methods should be incorporated into appropriate figure legends. See the article by Lohmann et al. (October 1990, Vol. 179: 214-2 1 8) 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 acceptable for review purposes.) The original manuscript must be typed in no smaller than 12 pitch, using double spacing (in- cluding figure i 'ends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. I oaper, 8'/2 by 11 inches. Please, no right justification. Manuv. i should be proofread carefully and er- rors corrected legib.s . , ,-k ink. Pages should be numbered consecutively. Margins on al1 sides should be at least 1 inch (2.5 cm). Manuscripts should conform to the Council of Biology Ed- itors Style Manual, 5th Edition (Council of Biology Editors. 1983) and to American spelling. Unusual abbreviations should
be kept to a minimum and should be spelled out on first referenc as well as defined in a footnote on the title page. Manuscript, should be divided into the following components: Title page Abstract (of no more than 200 words). Introduction, Materials and Methods, Results, Discussion. Acknowledgments. 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 inches, should be kept in mind in preparing figures for publ cation. We recommend that figures be about 1 '/: times the line;1 dimensions of the final printing desired, and that the ratio c the largest to the smallest letter or number and of the thickest to the thinnest line not exceed 1:1.5. Explanatory matter gen- erally 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 numbered in consecutive order, with no distinction between text and plate figures. The author's name and an arrow indicating orientation should appear on the reverse side of all figures.
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
he Literature Cited. Figure legends should contain enough in- irmation to make the figure intelligible separate from the text, egends should be typed double spaced, with consecutive Arabic jmbers. on a separate sheet at the end of the paper. Footnotes lould be limited to authors' current addresses, acknowledg- ments or contribution numbers, and explanation of unusual bbreviations. All such footnotes should appear on the title page. rootnotes 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 riled as Jones et ai, 1980. Personal communications and ma- erial in preparation or in press should be cited in the text only, /ith author's initials and institutions, unless the material has :en formally accepted and a volume number can be supplied, he list of references following the text should be headed Lit- ature Cited, and must be typed double spaced on separate ages, conforming in punctuation and arrangement to the style f recent issues of The Biological Bulletin. Citations should in- clude complete titles and inclusive pagination. Journal abbre- /iations 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 THE WORLD LIST OF SCIENTIFIC PE- RJODICALS, 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 journals with notes of the USASI abbreviations where these differ. CHEMICAL ABSTRACTS publishes quarterly supplements of ad- itional abbreviations. The following points of reference style or 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. Comp. Physiol. NOT / cell. comp. Physiol.)
C. All 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 I 'fsindajjelags Islendinga with- out 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 receive their first 100 reprints (without covers) free of charge. Additional re- prints may be ordered at time of publication and normally will be delivered about two to three 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 authors' alterations. The Biological Bulletin does not have page charges.
Hi
ERRATUM
The Biological Bulletin, Volume 180, Number 2, page 314
The following correction should be made in the article by Anthony Pires and Michael G. Hadneld titled, "Oxidative breakdown products of catecholamines and hydrogen peroxide induce partial metamorphosis in the nudibranch Phestilla sibogae Bergh (Gastropoda: Opisthobranchia)" (Biol. Bull. 180: 310-317).
On page 314, the second sentence of the second paragraph in the left hand column, which reads "Con- centration threshold for velar loss after 7-h exposure to fresh DA. . ." should read, "Concentration threshold for velar loss after 7-h exposure to aged DA. . ." The word "aged" replaces the word "fresh."
Reference: Biol. Bull. 182: 1-7. (February, 1992)
Fast-Strike Feeding Behavior in a Pteropod Mollusk,
Clione limacina Phipps
COLIN O. HERMANS' AND RICHARD A. SATTERLIE
Department of Zoology, Arizona State University. Tempe, Arizona 85287 and Friday Harbor Laboratories. Friday Harbor. Washington 98250
Abstract. High speed cinematography and video re- cordings were used to evaluate the fast-strike feeding re- sponse by which Clione limacina captures its prey, Li- macina fielicina. The acquisition phase of feeding involves rapid mouth opening and extrusion of three pairs of buccal cones. Mouth opening occurs in 10 to 20 ms, while hy- drostatic inflation of the buccal cones takes 50 to 70 ms. Buccal cones are immediately retracted if prey are not contacted. Buccal cones surround the prey and release a viscous material that may be used as an adhesive attach- ment to the prey shell. Surface ultrastructure of the buccal cones reveals that they are studded with clusters of capit- ulate papillae, which appear to be the source of the viscous secretory material.
Introduction
The pteropod mollusk Clione limacina feeds exclusively on shelled pteropods (Lalli and Gilmer, 1989). Due to the extremely limited dietary breadth of Clione, as well as the active swimming characteristics of both predator and prey (Limacina helicina in Friday Harbor, Washing- ton), it is not surprising to find rapidly activated and highly specialized feeding structures in Clione. For prey acqui- sition, Clione rapidly extrudes three pairs of oral append- ages, called buccal cones, which surround and adhere to the prey (see Lalli, 1970). Each buccal cone is cone shaped when retracted, but becomes more cylindrical when ex- truded. Extrusion of buccal cones is primarily due to hy- draulic inflation (Lalli and Gilmer, 1989). The acquisition phase of feeding is followed by a manipulative phase, dur- ing which the prey is turned so that the shell opening is
Received 5 June 1991; accepted 25 November 1991. ' Present address: Department of Biology, Sonoma State University, Rohnert Park. California 94928.
over the mouth of Clione. Manipulation is performed by the buccal cones and is followed by the consumptive phase, during which the prey is extricated from its shell. Extrication involves the use of two specialized hook sacs that form part of the buccal apparatus (Lalli, 1970). Each hook sac contains tufts of recurved chitinous hooks, which are protracted into the shell opening to grasp and pull the prey from its shell. Soft tissues of the prey are dislodged by alternate protractions and retractions of the hook sacs. Swallowing is aided by protraction and retraction move- ments of the radula, which is also part of the buccal ap- paratus. The soft tissues of the prey are swallowed whole (Wagner, 1885; Litvinova and Orlovsky, 1985; for other references see Lalli and Gilmer, 1989).
Two distinct forms of feeding behavior are observed. In the first, referred to here as the fast-strike response, Clione enters the acquisition phase of feeding from an unexcited, slow swimming activity state. During acqui- sition, swimming changes from slow to fast, and continues fast throughout the consummatory phase. During fast swimming, bending of the tail leads to frequent turning and looping movements of the entire body. If a fast-strike fails, and prey is not acquired, the buccal cones are im- mediately withdrawn, and fast swimming is terminated. The fast-strike response, which is initiated by prey contact, thus represents a sudden change to feeding behavior; if unsuccessful, the response is terminated by an equally sudden return to pre-strike swimming activity.
The second type of feeding behavior is initiated without direct physical contact with the prey. This activity involves fast swimming with loops and turns, as well as buccal cone extrusion and is referred to as "hunting behavior" (Litvinova and Orlovsky, 1985). Hunting behavior can be induced by placing an animal in seawater containing prey homogenates, by placing non-feeding Clione indi- viduals close to feeding individuals, or by injecting sero-
C. O. HERMANS AND R. A. SATTERLIE
tonin into the hemocoel (Litvinova and Orlovsky, 1985; Kabotyanski and Sakharov, 1988). Hunting behavior is similar to fast-strike feeding behavior in that the mouth is held open with the buccal cones protruding, and swim- ming is fast with frequent changes in direction. The be- haviors differ in two important ways. First, hunting be- havior does not require direct contact with an intact prey. Second, buccal cone extension and fast swimming are maintained in hunting behavior, whereas both are ter- minated immediately in the fast-strike if a prey item is not acquired. Note that the prey acquisition responses of Clione form a continuum, with fast-strike feeding at one extreme and indefinite hunting behavior at the other.
In this paper, we describe behavioral and morphological aspects of the acquisition phase of fast-strike feeding; a cine analysis of mouth and buccal cone movements and a description of the surface morphology of the buccal cones are included. This work provides the background for an ongoing electrophysiological investigation into the acquisition phase of feeding behavior and the role of pu- tative modulators on the motivational states underlying feeding behavior. It also extends the cinematic analysis of Clione feeding behavior by Litvinova and Orlovsky (1985).
Materials and Methods
Both Limacina and Clione were collected from the breakwater at Friday Harbor Laboratories, Friday Harbor, Washington, and held in one-gallon glass jars in a seawater table. Individual animals were filmed in a small glass chamber filled with seawater at room temperature (16- 18°C). Fast-strike sequences were filmed, within five days of animal collection, at 100 frames/s with a Redlake Lo- cam high speed 16 mm camera containing Kodak Plus- X negative film. Additional feeding sequences were "filmed" with a Sony CCD video camera HVM-200, equipped with a Nikon Micro-Nikkor lens, at the equiv- alent of 60 frames/s and were recorded on a Canon VR- 30 4-head portable video recorder. Feeding sequences were obtained by touching the prey, Limacina helicina, to swimming individuals of Clione. Limacina were attached with "Super Glue" to a human head hair or held in fine forceps.
A hair was attached to the Limacina shell as follows. A Limacina was placed in a shallow container on the stage of a dissecting microscope and the water level in the container was lowered until the shell, which is very hy- drophobic, broke through the surface film of the water. The Limacina was then turned to achieve the desired ori- entation. The root of a human hair was quickly dipped in a small droplet of "super glue" and applied to the sur- face of the shell.
Fast-strike responses were recorded from five different individuals. One complete response (from initiation
through acquisition) was recorded from each of these an- imals, but unsuccessful strikes were often recorded before the complete event. Unsuccessful strikes were also re- corded from three other individuals that never produced a complete response. All animals were between 1 .4 and 2.2 cm in body length.
Film sequences were analyzed frame-by-frame by making photographic prints of the sequences, and by pro- jecting individual frames onto tracing paper. Tracings were made of body, wing, head, and buccal cone positions. In one case, the images from sequential frames were dig- itized from tracings with a Jandel Scientific digitizing pad and processed with a computer-assisted three-dimensional reconstruction software program, (PC3D™, Jandel Sci- entific, Corte Madera, California). Photographic prints were made by projecting 16-mm frames directly onto photographic paper with a standard photographic enlarger. Video sequences, advanced frame-by-frame, were traced directly from a television screen during viewing.
For scanning electron microscopical investigation, specimens that were not adhering to prey were anesthe- tized by immersion in a 1 : 1 solution of 0.33 A/ magnesium chloride and seawater. A Clione adhering to its prey was prepared as follows. First, a Limacina, glued to a hair, was dangled in an aquarium so as to contact swimming individuals of Clione. When one of the pteropods struck at and gripped the prey, it was immediately pulled out of the aquarium and dropped directly into the primary fix- ative solution. The specimen continued to grip its prey as they were both being fixed, and remained attached until CO2 turbulence, during critical point drying, accidentally separated them, exposing the adherent surfaces. Fixation was completed by immersion in isotonic, cacodylate-buf- fered 2% glutaraldehyde, pH 7.3, at room temperature for 2 h, and postfixation was in cacodylate-buffered 1% osmium tetroxide for 1 h at room temperature. The spec- imens were dehydrated in ethanol, critical point dried from carbon dioxide, and sputter coated with gold and palladium before examination with an AMRay 1000 (Figs. 1, 4A) or a JEOL JSM-35 (Fig. 4B) scanning electron microscope.
Results
Acquisition behavior
Fast-strike feeding behavior was initiated by bringing a tethered Limacina into contact with the oral region of a freely swimming Clione. In our experience, the success rate of inducing fast-strikes was extremely low. With some animals, a day or more would pass without a strike being elicited; Clione apparently feeds irregularly. The degree of satiation in individual animals could not, therefore, be determined. The success rate was equally low, however, in animals that had been held in a jar for more than a
FAST-STRIKE FEEDING
week. With other animals, strikes could he obtained with some dependability. On one occasion, a response was ob- tained although the prey was not in contact with the oral region of Clione. In this case, the Limacina began rapid swimming movements when brought near the oral region of Clione, triggering an immediate fast-strike response.
In all observed fast-strike responses, the initial response of the acquisition phase was rapid mouth opening. When closed, the mouth forms a dorsoventral slit on the anterior margin of the head (Fig. 1 A). Lip retraction pulls the lips laterally, causing mouth gaping and protrusion of the buccal cones (Fig. IB). The degree of mouth opening can be judged from the position of the anterior tentacles, as recorded on film and video prior to and during fast-strike responses (Figs. 2, 3). The mouth of Clione is flanked by a pair of anterior tentacles that project from the antero- lateral margins of the head (Fig. 1 A). When Clione is hov- ering or slowly swimming forward (upward), the anterior tentacles are normally inflated and project forward (Figs. 1A, 2 A). During mouth opening, lip retraction, and pro- trusion of the buccal cones, the anterior tentacles rotate laterally 90°, so that their projection is perpendicular to the longitudinal axis of the animal (Figs. 2D, 3). Mouth opening occurs in the first 20 ms of the fast-strike and is accompanied by full exposure and partial protraction of the buccal cones (Fig. 3). This can be demonstrated by pulling open the mouth of an anesthetized animal, which exposes the buccal cones and causes them to bulge slightly out of the mouth (similar to that seen in Fig. IB). Three buccal cones lie on either side of the buccal mass (a mus- cular organ containing the radula and a pair or hook sacs), in a line parallel to the lips. The retracted cones are not inverted, but rather are collapsed and retracted into small
cavities, or cheek pouches, adjacent to the buccal mass. Buccal cones protract when they are inflated with he- molymph (Lalli and Gilmer, 1989). This is supported by our physiological experiments in which induced activity in buccal cone protraction motor neurons causes mouth opening, contraction of head musculature, but only partial extension of buccal cones (Norekian and Satterlie, in prep.). In these preparations, full expansion of the buccal cones is impossible because the head hemocoel is com- promised to allow electrophysiological recordings. In in- tact animals, expanded cones can extend approximately one-half body length from the mouth. Expansion is ac- companied by a decrease in the diameter of the head and the appearance of a distinct circular constriction in the neck region (Figs. 2D, 3). In two recorded sequences in which the head and neck outlines were clearly shown, the reduction in head diameter averaged 22.7% while the re- duction in neck diameter averaged 20.2%. Full expansion of the buccal cones, including the initial mouth opening, takes from 50 to 70 ms (Fig. 3). If the prey is not contacted during buccal cone expansion, the cones are immediately retracted, the mouth is closed, and the animal returns to slow swimming. Retraction of buccal cones is not a passive deflation, because the cones can be fully retracted in 70 to 90 ms (based on three unsuccessful strikes). On two occasions, strikes were aborted when the buccal cones were inflated to only 10 to 20% of the body length. In these cases, the cones were immediately retracted as in unsuc- cessful strikes.
Inflation of the buccal cones occurs from the base out- ward; the tips of the cones do not inflate until late in cone expansion. The uninflated tips are more opaque than the inflated parts of the buccal cones (Fig. 2D). As the cones
Figure 1. Scanning electron micrographs showing ventral views of heads of Clione in normal swimming posture ( 1 A) and with mouth (m) open and five of six buccal cones (be) partially protruded ( 1 B). Note the pair of anterior tentacles (t) that bear ciliary tufts (c). The head (h) is covered with a coat of motile cilia, w — wings, Ic — tufts of large neck cilia.
C. O. HERMANS AND R. A. SATTERLIE
Figure 2. Representative frames from cinematographic series taken at 100 frames/s showing a tethered Limacina being offered to a Clione. cw — wings of Clione. wl — wings of Limacina. be — buccal cones, t — anterior tentacles of Clione. (A) Predator and prey 200 ms (20 frames) before first sign of response to contact. (B) First sign of response to contact. Note the slight bulge on head of Clione (arrow). (C) Next frame (10 ms) after (B), showing buccal cones exploding from cheek pouches and forming grasping tentacles. (D) 4 frames (40 ms) after (C), showing buccal cones near full extension and beginning to grip the Limacina. Note the decreased diameter of the head, and the prominent neck constriction.
are extruded, they project outward at approximately 45° with a slight concave curvature with respect to the mouth. As the cones reach full expansion, they bend around the prey and adhere to its shell (Fig. 3).
Limacina shells pulled from the grasp of Clione buccal cones were coated with a viscous residue in the regions contacted by buccal cones. Clear viscous material pro- duced by the buccal cones could be gripped with fine for- ceps and lifted in fine strands from the surface of the sea- water containing the Clione. During hunting behavior, the protracted buccal cones frequently adhered to the wall of the container following contact with it. Removal of an adhering animal revealed residue on the glass, apparently adhesive.
Surface ultraslnicture of the buccal cones
The surface of each buccal cone is studded with clusters or rosettes of capitulate papillae (Fig. 4). The number of papillae in each cluster varies from two or three to about a dozen. The clusters near the bases of the buccal cones contain the fewest papillae per rosette, those toward the tips contain more. Each papilla is about 1 5 nm high and somewhat less than 10 ^m in diameter. The tip of each papilla is slightly inflated, forming a lumpy capitulum about 10 //m in diameter. Each rosette has a common stalk, about 20 /urn in diameter and 20 yum in height. Long cilia protrude from the sides of the papillae and project from the surface of the buccal cone between the papillae.
Tight clusters of cilia protrude from the centers of some of the papillary rosettes. Isolated clusters of cilia, were also observed, but they were not common (Fig. 4A).
When the buccal cones are retracted, the epidermis be- tween the clusters of papillae is deeply folded, and the capitula and cilia form a tightly packed feltwork or welter on the surface of each cone. When the buccal cones are ex- tended, the rosettes of papillae stand up above a smooth, simple squamous epithelium that stretches tightly over the tentacular surface between the rosettes of papillae (Fig. 4A).
The surface of the shell of Limacina, to which the buccal cones adhere, is very smooth, transparent, and very hy- drophobic; it appears smooth when viewed with a scan- ning electron microscope. The shells of dead Limacina lose their hydrophobic properties rapidly. Where the shell of a Limacina is contacted by the buccal cones of a fast- striking Clione are fine threads, observable by SEM, that correspond to those that appear on the surfaces of the buccal cones where they contact the Limacina shell (Fig. 4B). These threads appear to originate from the tips of the capitulate papillae, but this possibility is difficult to establish with certainty.
Discussion
The fast-strike response of Clione consists of a rapid opening of the mouth and a hydraulic inflation of the six buccal cones, the entire response occurring in 50 to 70 ms. In aborted or unsuccessful strikes, withdrawal of buc-
FAST-STRIKE FEEDING
Figure 3. Tracings of Clione and Limacina from cine series with time intervals of 10ms between frames and covering the 100 ms interval from one frame (10 ms) prior to the first sign of response to the prey through the initial grasping of the shell. The sequence has been plotted twice with a 7° shift in the y-axis. When viewed with a stereoscopic viewer or with crossed eyes, the sequence will appear in 3-dimensions with time represented in the z-plane. Buccal cone labels: (Id) — left dorsal, (1m) — left median, (Iv) — left ventral, (rd) — right dorsal, (rm) — right me- dial. Right (rat) and left (lat) anterior tentacles are also labelled.
cal cones is nearly as rapid. This would suggest that both expansion and withdrawal are active responses. Fast-strike prey acquisition is thus distinct from the hunting behavior described by Litvinova and Orlovsky (1985) in which Clione rapidly swim with the buccal cones held in an ex- panded state. The initial phase of hunting behavior pre- sumably involves similar mouth opening and buccal cone inflation.
The low success rate in triggering a fast-strike under laboratory conditions suggests that the fast-strike response has a high threshold for activation. Lowering of this threshold could result in behavior that is more disposed toward feeding, such as hunting behavior. In this case, the difference between responses to prey during normal swimming and those during hunting behavior might be one of motivational state. This difference can best be il- lustrated by comparing buccal cone responses during hunting and during an unsuccessful fast-strike response. In the former, the buccal cones are held in an expanded position despite the lack of mechanical contact with the prey. In the latter case, lack of prey contact results in a
rapid withdrawal of the buccal cones and a return to nor- mal swimming behavior. In hunting behavior, therefore, buccal cone withdrawal must be suppressed, even in the absence of direct mechanical contact with prey.
The nature of the trigger underlying the change in be- havioral state, from hunting to fast-strike, is not known. It may, however, involve serotonergic input to the feeding system, because bath application or hemocoel injection of serotonin can trigger behavioral responses similar to those of hunting behavior; i.e., the responses can be evoked although the animal has not been exposed to prey or prey extracts (Kabotyanski and Sakharov, 1988). The external stimulus for a switch to hunting behavior pre- sumably involves chemosensory input because Limacina extracts, or proximity to feeding Clione, can initiate hunt- ing behavior (Litvinova and Orlovsky, 1985).
Inflation of the buccal cones is remarkable for its great speed. Expansion is associated with a decrease in head and neck diameter, suggesting that increased hemocoelic pressure is associated with buccal cone in- flation. Arshavsky et al. (1990) have shown that heart rate in Clione increases during hunting behavior, further supporting the idea that feeding responses are associated with increases in hemocoelic pressure. Pressure changes can be localized in the head as a muscular diaphragm separates head and body hemocoels. The diaphragm surrounds the anterior aorta and may act as a physio- logical valve further regulating blood flow to the hem- ocoel in the head (Lalli, 1967).
With buccal cones protruded, the Clione appear much like small squid. This led Wagner (1885) and Pelseneer (1885) to consider the possibility of homology between Clione buccal cones and squid tentacles. However, em- bryonic origins and innervation patterns demonstrate that they are not homologous (see Lalli and Gilmer, 1989. for a discussion of pteropod systematics and affinities). The mechanisms by which the two types of tentacles move to grasp their prey are quite distinct. Kier demonstrated that cephalopod tentacles are muscular hydrostats (Kier, 1982, 1987, 1988; Smith and Kier, 1989). Muscular hydrostats are readily distinguishable from hydrostatic skeletons that use a hydraulic mechanism in that their volumes are made up almost entirely of muscular tissue. Therefore, although they can undergo extensive changes in shape, muscular hydrostats do not substantially change volume. Hydraulic hydrostatic skeletons, in contrast, are fluid-filled cavities surrounded by muscular or fibrous tissues that resist the hydrostatic pressure within (Smith and Kier, 1989).
No clear differences are found when the speed of ten- tacle protractions in muscular hydrostats is compared with that of the hydraulic system of Clione, because the range of protraction speeds found in muscular hydrostat systems is very wide. For example, each of the 19 pairs of digital tentacles of Nautilus consists of an extensible, muscular.
C. O. HERMANS AND R. A. SATTERLIE
Figure 4. (A) Enlarged view of a part of one of the partially protruded buccal cones shown in Figure IB. The head of each papilla (p) in the rosettes is studded with bumps. Motile cilia (c) project from the shaft of each papilla, whereas tufts of cilia (sc) project from the centers of rosettes, or less commonly are isolated from the rosettes. (B) Similar view of a region on a buccal cone of a different specimen, which was allowed to adhere to a Limacina shell and was fixed while grasping the prey. Dense mats of thread-like structures (t) can be seen on the adherent surfaces of buccal cones. Some appear to originate from the tips of capitulate papillae (arrows).
adhesive cirrus enclosed in a protective sheath. Protrusion of the cirrus from the tip of its sheath, which is necessary for it to grasp prey, requires 5-10 s or longer (Kier, 1987). At the opposite extreme, the tentacles of squid elongate fully in 15 to 30 ms (Keir, 1982, 1985).
The buccal cones of Clione can be protruded in less than 100 ms, and this performance is best appreciated when compared to other fast invertebrate prey capture behaviors that have been subjected to cine analysis. For example, prey seizure in the opisthobranch mollusk Na- vanax occurs in 380 ms; this is a muscular phenomenon involving a pharyngeal lunge followed by lip closure around the prey (Susswein el a/.. 1984; Susswein and Achituv, 1987). Prey acquisition behaviors involving the movement of body parts that are supported by hard skel- etal elements can be much faster; e.g., the strike of the second thoracic appendages of stomatopod crustaceans occurs in 4-8 ms (Burrows, 1969).
Other gastropods can strike rapidly. In particular, the proboscides of toxoglossans, which contain poisonous.
dart-like radular teeth, are potentially as rapid as the Clione buccal cone system. Predatory strikes have been described and photographed in a turrid, Ophiodermella inermis (Shimek and Kohn, 1981), and in Conns (Ny- bakken, 1967). The proboscis of Conns is protruded as a hydrostatic skeleton (Greene and Kohn, 1989), but the speed with which these strikes occur has not been analyzed by high speed cine or video.
Whereas some gymnosomatous pteropods do appre- hend their prey with suction cups, somewhat like coleo- idean cephalopods (Lalli and Gilmer, 1989; Kier and Smith, 1990), the adhesiveness of the buccal cones of Clione resembles that of the digital tentacles of Nautilus. The digital tentacles grip prey by means of ridges on the cirri that protrude from the tips of the sheaths that form the bases of the tentacles (Fukuda, 1987; Kier, 1987). In both Nautilus and Clione, the adhesive structures are en- sheathed when not in use. In both cases, a question re- mains: to what degree are the prey simply gripped, and to what extent do the tentacles adhere? Fukuda (1987)
FAST-STRIKE FEEDING
suggested that the ensheathing of the cirri in Nautilus might serve to save mucus.
Apprehension of the prey by Clione may be partly by chemical adhesion and partly by the physical gripping of the Limacina shell by the enclosing buccal tentacles. The capitula of the papillae on the buccal tentacles might be thrust through the boundary layer of water, covering the hydrophobic surface of the Limacina shell and providing the means of attachment to, or gripping of the shell, just like the beaded gloves used by soccer goalies and football wide receivers aid in gripping the wetted, hydrophobic surfaces of footballs. In both cases, the bumps aid in adhesion; they penetrate the boundary layer of water, eliminating this weak boundary layer by driving it into the spaces between the bumps (Waite, 1987).
Because the buccal tentacles appear to be chemically adhesive and yet can detach to manipulate the shell of the prey so that the opening is aligned with the Clione mouth, the possibility that both adhesive and releasing chemicals are secreted must be considered (Hermans, 1983). Examination of the ultrastructure of the buccal cones and their secretions, as well as analyses of the control of feeding behavior, will help answer this and other ques- tions about prey acquisition in Clione.
Acknowledgments
We thank Dr. Tigran Norekian for translating Litvinova and Orlovsky (1985), Ms. Michelle Lagro for preparing specimens of Clione for electron microscopy. Prof. A. O. D. Willows of Friday Harbor Laboratories for space and equipment. Prof. R. Strathmann for use of his cine camera and the macro lens and video equipment. Dr. Tom Schroeder for instruction in SEM, also Mr. W. Sharp for instruction in EM and for the use of the Biological Electron Microscope Facility at Arizona State University, Mr. Chaz. Kazelik for help with the PC3D stereoscopic imaging program, Drs. Claudia Mills and Norm McLean for collecting and shipping specimens, and several other friends, colleagues, staff, and family at the Friday Harbor Labs for help in collecting specimens and for many other kindnesses. Thanks also to Sarah Cohen for suggesting the use of "Super Glue." Our perspective on the potential similarities between the strikes of toxoglossans and Clione has benefitted from discussions with Drs. Ron Shimek, Matt James, and Ed Smith.
Literature Cited
Arshavsky, Y. I., T. G. Deliagina, I. M. Gelfand, G. N. Orlovsky, Y. V. Panchin, G. A. Pavlova, and L. B. Popova. 1990. Neural control
of heart beat in the pteropod mollusc Clione limacina: coordination
of circulatory and locomotor systems. J. Exp. Biol. 148: 461-475. Burrows, M. 1969. The mechanics and neural control of the prey cap- ture strike in the mantid shrimps Stiuilla and Hemisquilla. Z I'ergl.
Physiol.62: 361-381. Fukuda, Y. 1987. Histology of the long digital tentacles. Pp. 249-256
in Nautilus: The Biology and Paleobiology of a Living Fossil. W. B.
Saunders and N. H. Landman. eds. Plenum, New York. Greene, J. L., and A. J. Kohn. 1989. Functional morphology of the
Conns proboscis (Mollusca: Gastropoda). J Zoo/., Loud. 219: 487-
493. Hermans, C. O. 1983. The duo-gland adhesive system. Oceanogr. Mar.
Biol. Ann. Rev. 21:283-339. Kabotyanski, E. A., and D. A. Sakharov. 1988. Monoamine-dependent
behavioural states in the pteropod mollusc Clione limacina. Symposia
Biologica Hungarica 36: 463-477. Kier, W. M. 1982. Functional morphology of the musculature of squid
(Loliginidae) arms and tentacles. J. Morphol. 172: 179-192. Kier, W. M. 1985. The musculature of squid arms and tentacles: ul-
trastructural evidence for functional differences. / Morphol. 185:
223-239.
Kier, W. M. 1987. The functional morphology of the tentacle mus- culature of Nautilus pompilius. Pp. 257-269 in Nautilus: The Biology
and Paleobiology of a Living Fossil. W. B. Saunders and N . H . Land- man. eds. Plenum. New York. Kier, W. M. 1988. The arrangement and function of molluscan muscle.
Pp. 21 1-252 in The Mollusca. Form and Function. Vol. 1 1. E. R.
Trueman and M. R. Clarke, eds. Academic Press, New York. Kier, W. M., and A. M. Smith. 1990. The morphology and mechanics
of octopus suckers. Biol. Bull. 178: 126-136. Lalli, C. M. 1967. Studies on the structure and biology of two gym-
nosomatous pteropods, Clione kincaidi Agersborg and Crucibran-
chaea macrochira (Meisenheimer). Ph.D. Dissertation, University of
Washington. 175 pp. Lalli, C. M. 1970. Structure and function of the buccal apparatus of
Clione limacina (Phipps) with a review of feeding in gymnosomatous
pteropods. J. Exp. Mar. Biol. Ecol. 4: 101-1 18. Lalli, C. M., and R. W. Gilmer. 1989. Pelagic Snails: The Biology of
Holoplanktonic Gastropod Mollusks. Stanford University Press,
Stanford. 259 pp. Litvinova, N. M., and G. N. Orlovsky. 1985. Feeding behavior of the
pteropod mollusc Clione limacina. Byull. MOIP. Old. hiol. 90: 73-
77. Nybakken, J. 1967. Preliminary observations on the feeding behavior
of Conus purpurascens Broderip, 1833. I'e/iger 10: 55-57. Pelseneer, P. 1885. The cephalic appendages of the gymnosomatous
pteropoda, and especially ofC/ione. Q. J. Microsc. Sci. 25: 491-509. Shimek, R. L., and A. J. Kohn. 1981. Functional morphology and
evolution of the toxoglossan radula. Malacologia 20: 423-438. Smith, K. K., and W. M. Kier. 1989. Trunks, tongues, and tentacles:
moving with skeletons of muscle. Am. Sci. 77(1): 28-35. Susswein, A. J., and Y. Achituv. 1987. Pharyngeal movements during
feeding sequences ofNavanax inermis (Gastropoda: Opisthobranchia)
in successive stages of dissection. / Exp. Biol. 128: 323-333. Susswein, A. J., Y. Achituv, M. S. Cappell, D. C. Spray, and M. V. L.
Bennett. 1984. Pharyngeal movements during feeding sequences
in Navanax inermis: a cinematographic analysis. J. Comp. Physiol.
155:209-218. Wagner, N. 1885. Die H'irbcllosen des Weissen Meeres. Wilhelm En-
gelmann, Leipzig. Pp. 1-168. Waite, J. H. 1987. Nature's underwater adhesive specialist. Inl. J
Adhesion and Adhesives 1: 9-14.
Reference: Biol. Bull. 182: 8-14. (February, 1992)
Effects of Photoperiod and Temperature on
Egg-Laying Behavior in a Marine Mollusk,
Aplysia californica
NANCY L. WAYNE AND GENE D. BLOCK Department of Biology, University of Virginia, C/iarlottesville, Virginia 22901
Abstract. The primary purpose of these studies was to determine whether photoperiodic signals could influence seasonal egg-laying behavior in the marine mollusk, Aply- sia californica. Egg-laying behavior was monitored from groups of animals that were collected at four times of year and maintained in different temperature and photoperi- odic conditions in the laboratory. Animals that were ob- tained in autumn and kept in warm water laid eggs more frequently than those in cold water, regardless of photo- period. Furthermore, animals maintained on short days and warm water laid eggs more frequently than those on long days and warm water. Animals in cold water showed little to no egg laying, and a photoperiodic response was not evident. Animals that were collected in either winter or spring and maintained in warm water showed little or no spontaneous egg laying throughout the study, regardless of photoperiod. As with the autumn animals, Aplysia in- dividuals obtained in summer and kept on short days and warm water laid eggs more frequently than those kept on long days and warm water. These results provide the first evidence that the reproductive system of A californica is responsive to photoperiod. Overall, the data suggest that warm water is permissive for egg laying, and that short days can further stimulate this behavior. However, there is a strong inhibition of spontaneous egg laying during the winter and spring, which neither warm water nor short photoperiod can overcome. The role of the eyes in me- diating the photoperiodic response was also investigated. A control group of intact animals kept on short days laid eggs more frequently than those on long days, but this photoperiodic response was not evident in eyeless
Received 12 August 1 99 1 ; accepted 31 October 1991.
animals. These results suggest that the eyes play a role in mediating the effects of photoperiod on egg laying behavior.
Introduction
Like many animals living in the temperate zone, the marine mollusk Aplysia californica breeds seasonally. Both field and laboratory observations indicate that this species is reproductively competent during the summer and autumn, and reproductively quiescent during the winter and spring (Strumwasser el al. 1969; Audesirk, 1979; Berry, 1982). The onset of the breeding season is indicated by a significant increase in the incidence of cop- ulation and egg laying (Strumwasser et a/., 1969; Audesirk, 1979), as well as increased synthesis of the hormone that controls egg laying (egg laying hormone; Berry, 1982).
Earlier work has shown that egg laying hormone, a peptide synthesized and secreted by the neuroendocrine bag cells, is responsible for triggering egg-laying behavior (Strumwasser et al.. 1969;Kupfermann, 1970; Arch, 1972; Dudek et al.. 1980; Stuart et al.. 1980; Chiu and Strum- wasser, 1981; Blankenship et al., 1983). Although much is known about the molecular biology of bag-cell peptides (Chmetal.. 1979; Heller et al.. 1 980; Scheller ?/«/.. 1982; Mahon and Scheller. 1983) and about the electrophysi- ological properties of bag cells (Kupfermann and Kandel. 1970; Kaczmarek et al.. 1978, 1982; Kaczmarek and Strumwasser, 1981), the seasonal regulation of bag-cell activity and egg laying remains obscure. The general goal of this and future studies is to gain insight into the mech- anisms underlying seasonal fluctuations in egg laying be- havior and in reproductive neuroendocrine function.
LIGHT AND TEMPERATURE AFFECT EGG LAYING
The occurrence of reproductive activity at a particular time of year suggests the involvement of some environ- mental timing agent (e.g., ambient temperature, photo- period, food availability, specific nutritional cue). Previous studies in Aplysia have shown that warm water can stim- ulate egg laying, whereas cold temperatures inhibit this behavior (Berry, 1984; Pinsker and Parsons, 1985). Al- though the authors interpreted their results to suggest that changes in the rate of egg laying are solely dependent on seasonal cycles of temperature, the studies did not test for effects of other environmental variables, such as photo- period.
The annual cycle of photoperiod is the most regular and predictable environmental factor, and is therefore used by a wide variety of temperate-zone species to time reproduction to the appropriate season (mammals: Turek and Campbell, 1979; birds: Rowan, 1926; reptiles: Licht, 1967: insects: Lees, 1966; terrestrial slugs: Sokolove et ai, 1984). A. californica are intertidal organisms, spending much of their time near the water surface (Audesirk, 1979); thus they would be exposed to annual changes in day length. A. californica might use photoperiodic infor- mation, as well as temperature cues, to synchronize re- production to a particular time of year. The main goal of this study was to determine whether egg laying behavior can be influenced by photoperiodic signals.
Materials and Methods
General
Specimens of Aplysia californica (200-300 g) were captured off the coast of California (approximately 34°N latitude) by Alacrity Marine Supply, Redondo Beach, California. At the collection sites, the annual range in water temperature is from approximately 10 to 20°C (Dan Stark, Alacrity Marine Supply, pers. comm.), and the an- nual range in photoperiod is from 1 1 to 15.5 h light/day (includes 1 h civil twilight). Upon arrival in the laboratory, animals were maintained in temperature- and light-con- trolled seawater tanks (475 liters; light intensity at water surface was 700 lux as measured with a photographic light meter). Water was recirculated through undergravel filters within the tanks. Treatment groups (initially, 12 animals per group; 0-3 animals/group died during the course of the studies) were maintained in separate tanks, and all animals were kept in single, perforated plastic buckets (20 cm in diameter) so that each individual could be moni- tored throughout the studies.
To document the egg laying capability (i.e.. reproduc- tive maturity) of each animal, atrial gland extract was injected into the hemolymph of all animals upon arrival in the laboratory (Nagle et al, 1985). Animals with a ma- ture reproductive system will lay eggs in response to atrial
gland extract, while immature animals will not lay eggs. An Aplysia that did not lay eggs spontaneously during the course of the studies was again treated with atrial gland extract at the end of each study to assess maturity. Only those animals that were reproductively mature by the end of the studies were included in the analysis. Animals were fed a combination of Romaine lettuce and dried seaweed (Msubi Nori, Japan Food Corp.) daily. Egg masses were recorded daily from individual buckets. Because Aplysia lays eggs at a maximal rate of once per day and does not consume its own eggs (unpub. obs.), the presence or ab- sence of an egg mass is an excellent indication of whether an animal exhibited egg laying behavior on any given day.
The effects of photoperiod on egg laying behavior were determined as follows. Specimens of Aplysia were col- lected and shipped to our seawater facilities at four dif- ferent times of year. Animals that arrived in the early AUTUMN 1988 (Sept. 22) were all reproductively mature at the beginning of the study and were divided into four treatment groups. Aplysia individuals were kept either on short days (8 h light/day) or on long days ( 16 h light/day); animals on these two photoperiods were further divided and maintained either in warm (20°C) or in cold (15°C) water. Thus the combined effects of photoperiod and water temperature on egg laying could be investigated. Animals maintained in cold water rarely, if ever, layed eggs, so we dropped the cold-water group from the remaining studies. Aplysia individuals that arrived in the early WINTER 1989 (Jan. 3) and the early SPRING 1989 (Mar. 31) were reproductively immature at the beginning of the studies; but they had all reached maturity by the end of the ex- periments. In these two studies, all animals were main- tained in warm water and kept either on short or on long days. Aplysia individuals that arrived in the early SUM- MER 1989 (June 23) were reproductively mature at the beginning of the study and were maintained in warm water and kept either on short or on long days.
The role of the eyes in mediating the effects of photo- period on egg laying was investigated with specimens of Aplysia that were brought to the laboratory in the late SUMMER 1990 (Aug. 7) and maintained in warm water and 14.25 h light/day (photoperiod in mid-August at 34°N latitude) for three days. All of these animals were im- mobilized with MgCl: (injected into hemolymph); half of them were bilaterally enucleated, and the other half served as intact controls. Following surgery, animals were further divided and kept either on short (8 h light/day) or on long days (16 h light/day), making a total of four treatment groups.
Analysis of data
Differences in egg laying between treatment groups were assessed by Chi-square analysis. Values were significantly different if P< 0.05.
10
N. L. WAYNE AND G. D. BLOCK
100
AUTUMN
I short days, warm I long days, warm
A. .
short days, cold long days, cold
B.
C.
6-10
11-15
16-21 1-5 6-10
Days of experiment
11-15 16-21
warm cold
shortday brgday shortday bngday
Figure 1. Percent of.-iplysia individuals laying eggs during the early AUTUMN 1988. Panel A: Animals were kept in warm (20°C) water and either short (8 h light/day) or long (16 h light/day) days. Data were averaged (+sem) into 5-day bins. Panel B: Animals were kept in cold (15°C) water and either short (8 h light/day) or long ( 16 h light/day) days. Data are presented as in panel A. Panel C: Data from the 4 groups are presented as the percent of animals laying eggs each day, averaged (+sem) over the entire 2 1-day study. Different letters indicate values are significantly different (at least P < 0.05).
Results
Photoperiodic effects on egg laying
Overall, photoperiod and temperature can both affect the frequency of egg laying. In the AUTUMN, Aplysia individuals kept in warm water laid eggs more frequently than those kept in cold water (Fig. 1). Furthermore, an- imals maintained on short days and warm water laid eggs more frequently than those kept on long days and warm water. However, in the WINTER (Fig. 2) and in the SPRING (Fig. 3), egg laying frequency overall was sup- pressed in all groups (even though animals were repro- ductively mature by the end of the studies; see Materials and Methods), and there was no apparent effect of pho-
toperiod on egg laying. In the SUMMER (Fig. 4), we once again observed the emergence of a photoperiodic effect: Aplysia maintained on short days and warm water laid eggs more frequently than those kept on long days and warm water. This photoperiodic response in the summer was not as robust as that seen during the previous autumn (compare Figs. Ic and 4b).
Photoperiodic effects in intact vs. eyeless animals
The eyes appear to play a role in transducing photo- periodic information to the reproductive axis responsible for regulating egg laying (Fig. 5). Once again, control an- imals kept in short days and warm water laid eggs more frequently than those kept on long days and warm water.
WINTER
100
Q. <
80"
60-
ra 4fj-
20-
short days, warm long days, warm
B.
50
40 >
"S ~°-
Q) *<
30 ^ £
0) —
20 | |
10 ~ I 0
1-5 6-10 11-15 16-20 21-25 26-30 31-36 short days long days Days of experiment
Figure 2. Percent of Aplysia individuals laying eggs during the early WINTER 1 989. Animals were kept in warm (20°C) water and either short (8 h light/day) or long (16 h light/day) days. Panel A: Data were averaged (+sem) into 5-day bins. Panel B: Data are presented as the percent of animals laying eggs each day, averaged (+sem) over the entire study. There was no significant difference between the values of the two groups.
LIGHT AND TEMPERATURE AFFECT EGG LAYING SPRING
II
|
iuu - |
B short days, warm A- |
|
|
in 8 |
80- |
long days, warm |
|
TO c |
60 - |
|
|
_<8 |
40- |
|
|
| |
20- |
~F |
|
3« |
^ Wlfa |
|
|
<c ^><rt -*H-«C *c f^r\ OH oc oc *si |
|
50 |
||
|
B. |
^ |
|
|
• |
40 |
II |
|
*< *2. |
||
|
" |
30 |
|
|
< — |
||
|
If |
||
|
20 |
||
|
^ ^ |
10 |
<a CO |
|
0 |
short days long days
Days of experiment
Figure 3. Percent ofAplysia individuals laying eggs during the early SPRING 1989. Data are presented as in Figure 2.
On the other hand, there was no significant difference in the frequency of egg laying between the two eyeless groups. Although the photoperiodic response in the intact control group was significant, it was not nearly as robust as that seen in a previous study (see Fig. 1 ).
Discussion
This study provides the first evidence that the repro- ductive system of Aplysia is responsive to photoperiodic signals. The results suggest that both photoperiod and temperature can influence the seasonal rhythm of egg lay- ing. Specifically, warm temperature is permissive for the expression of the stimulatory effects of short days. Studies in another poikilotherm, the lizard Anolis carolinensis, have also documented that the reproductive response to stimulatory day lengths is evident in warm, but not cool, temperatures (Licht, 1967). In addition, recent work in the edible snail Helix pomalia has shown that egg-laying behavior is regulated by both photoperiod and tempera- ture cues (Gomot, 1990). In the wild, the reproductive
activity of Aplysia califomica peaks in late summer-au- tumn (Strumwasser el al, 1969; Audesirk, 1979; Berry, 1982). At this time of year, water temperature is reaching a maximum off the coast of California, and day length is decreasing. Our findings that warm water and short days stimulate egg laying are therefore consistent with the be- havior of the animal in its natural environment.
Animals brought to the laboratory in the winter and spring layed eggs infrequently, if at all, regardless of en- vironmental treatment. That is, an average of less than 10% of the winter and spring animals laid eggs on any given day during the course of the two studies — even un- der stimulatory conditions of short days and warm water. Although these animals were reproductively immature at the onset, towards the end of the studies they had reached maturity and were capable of laying eggs following hor- monal stimulation (see Materials and Methods). There- fore, ovotesticular function was most likely not the lim- iting factor in these studies (however, we do not know when during the studies animals attained reproductive maturity).
SUMMER
100
short days, warm long days, warm
6-10 11-15 16-20 21-25
Days of experiment
2&31 short days long days
Figure 4. Percent of Aplysia individuals laying eggs during the early SUMMER 1989. Data are presented as in Figure 2. In Panel B, different letters indicate values are significantly different (P < 0.05).
12
N. L. WAYNE AND G. D. BLOCK
100
short days, intact long days, intact
short days, eyeless D long days, eyeless
1-5 6-1011-1516-2021-2526-30 1-5 6-1011-1516-2021-2526-30 intact eyeless
Days of experiment short-day long-day short-day long-day
Figure 5. Percent ofAplysia individuals laying eggs during the late SUMMER 1 990. All animals were maintained in warm (20°C) water. Panel A: Intact, control animals were kept on short (8 h light/day) or long (16 h light/day) days. Data were averaged (+sem) into 5-day bins. Panel B: Bilaterally enucleated animals were kept on short (8 h light/day) or long (16 h light/day) days. Data are represented as in panel A. Panel C: Data from the 4 groups are presented as the percent of animals laying eggs each day, averaged (+sem) over the entire 30-day study. The letters a and b indicate values that are significantly different (P < 0.05). 'Indicates that values approached significant difference compared to that of the intact, long-day control group (P < 0. 10).
A common phenomenon among some seasonally breeding vertebrates is a spontaneous shutdown of the reproductive system during the non-breeding season (liz- ard: Cueller and Cueller, 1977; birds: Hamner, 1967; Robinson and Follett, 1982; mammal: Robinson and Karsch, 1984). During this period of reproductive qui- escence, previously inductive photoperiodic cues no longer stimulate reproductive activity. This period of insensitivity to stimulatory photoperiod (commonly labelled 'photo- refractoriness') is an endogenous process and can be 'bro- ken' by exposing the animal to a bout of inhibitory pho- toperiod, followed by a stimulatory day length (Jackson el ai, 1988). In Aplysia. we have shown that previously stimulatory environmental cues (warm water, short days) did not stimulate spontaneous egg laying during the non- breeding season in winter and spring. Aplysia may there- fore behave like many other seasonal breeders and become refractory to stimulatory signals. If this is so, then pre- treatment with long days and cold temperatures should be able to break refractoriness to stimulatory short days and warm temperatures.
But mechanisms other than an endogenous refracto- riness to an environmental signal might equally well un- derlie the cessation of spontaneous egg laying by Aplysia during winter and spring. For instance, one or more key components of the reproductive neural axis may be de- velopmentally immature during the winter and spring (even though the reproductive tract can mature in the laboratory). Alternatively, some environmental cue (e.g., food or other nutritional item necessary for high levels of spontaneous egg laying) may be missing during that time of year.
Further, our results suggest that the eyes play a role in mediating photoperiodic information to the reproductive axis responsible for regulating egg laying behavior. Spe- cifically, photoperiod had no effect on egg laying in those animals that were bilaterally enucleated. The eyes of Aplysia contain both photoreceptors and a circadian pacemaker (Jacklet, 1969; Eskin, 1971). The circadian system is involved in the neural pathway mediating pho- toperiodic responses in most animals investigated (Follett and Sharp, 1969; Elliott, 1976; Almeida and Lincoln, 1982). Furthermore, both ocular and extraocular photo- receptors mediate photoperiodic responses in a variety of species (Reiter, 1969; Follett et a/.. 1975; Legan and Karsch, 1983; Foster and Follett, 1985). In Aplysia, pho- toreceptors are found not only in the eye, but also in structures as diverse as the abdominal ganglion (Andresen and Brown, 1982), the cerebral ganglion (Block and Smith, 1973), the rhinophores (Jacklet, 1980), the oral veil (Cook et ul. . 1 99 1 ). and the siphon ( Lukowiak and Jacklet, 1972). Therefore, the relative roles of the ocular photoreceptors and ocular pacemakers in mediating the effects of pho- toperiod on egg laying are not clear. For instance, both the ocular photoreceptors and ocular pacemakers may be playing a role in photoperiodic time measurement. Al- ternatively, extraocular photoreceptors may be transmit- ting light signals to the ocular circadian pacemaker, which then sends its signals to the next step in the photoperiodic response system.
Nevertheless, we must stress that our results are difficult to interpret, because the photoperiodic response in the intact controls in the last experiment was weak compared to that seen in the first experiment (compare Fig. 1 with
LIGHT AND TEMPERATURE AFFECT EGG LAYING
13
Fig. 5). An intriguing mystery arising from these studies is the source of the variability in the photoperiodic re- sponse from one year to the next. Because we work with animals captured in the wild, we have no control over the environmental history of the animal. That is, we cannot control for variations in microhabitat (i.e., local and year- to-year variability in water temperature, food availability, sexual experience). Large, year-to-year fluctuations in the availability of the algae that Aplysia feed upon were ob- served during the course of these experiments: algae were abundant in the summer to early autumn of 1988, but scarce in the summer to early autumn of 1989 and 1990 (Dan Stark, Alacrity Marine Supply, pers. comm.); and these fluctuations were associated with similar changes in the robustness of the photoperiodic response. In addition to photoperiodic and temperature signals, there are other environmental variables (e.g., food or nutritional cues) that might affect spontaneous egg laying. All of these en- vironmental cues may act in combination such that one variable alters the effectiveness of the other variables on the frequency of egg laying. For instance, the photoperi- odic response during the breeding season may be more robust in those animals with a high level of nutrition; low nutrition may weaken the photoperiodic response. Food availability has pronounced effects on the photoperiodic diapause response in some insect species (Saunders, 1979). Future work investigating the role of food cues may pro- vide some information on its importance in the expression of a robust photoperiodic response in Aplysia.
Acknowledgments
This work was supported by NIH grants NS-08725 to NLW and NS- 15264 to GDB.
Literature Cited
Andresen, M. C., and A. M. Brown. 1982. Cellular basis of the pho-
toresponse of an extraretinal photoreceptor. Experienlia 38: 1001-
1006. Almeida, O. F. X., and G. A. Lincoln. 1982. Photoperiodic regulation
of reproductive activity in the ram: evidence for the involvement of
circadian rhythms in melatonin and prolactin secretion. Biol. Reprod.
27: 1062-1075. Arch, S. 1972. Biosynthesis of the egg-laying hormone (ELH) in the
bag cell neurons of Aplysia californica. J. Gen. Physiol. 60: 102-1 19. Audesirk, T. E. 1979. A field study of growth and reproduction in
Aplysia californica. Biol. Bull. 157: 407-421. Berry, R. \V. 1982. Seasonal modulation of synthesis of the neuro-
secretory egg-laying hormone of Aplysia. J Neurobiol. 13: 327-335. Berry, R. W. 1984. Environmental temperature modulates the rate of
synthesis of egg-laying hormone in Aplysia. J. Comp. Physiol. B 154:
545-548. Blankenship, J. E., M. K. Rock, L. C. Robbins, C. A. Livingston, and
H. K. Lehman. 1983. Aspects of copulatory behavior and peptide
control of egg laying in Aplysia. Fed. Proc. 42: 96-100.
Block, G. D., and J. T. Smith. 1973. Cerebral photoreceptors in Aplysia. Comp. Biochem. Physiol. 46a: 115-121.
Chiu, A. Y., and F. Strumwasser. 1981. An immunohistochemical study of the neuropeptidergic bag cells of Aplysia. J. Neurosci. 1:81 2-826.
Chiu, A. Y., M. W. Hunkapiller, E. Heller, D. K. Stuart, L. E. Hood, and F. Strumwasser. 1979. Purification and primary structure of the neuropeptide egg-laying hormone of Aplysia californica. Proc. Nail. Acad. Sci. 76: 6656-6660.
Cook, D. G., M. Stopfer, and T. J. Carew. 1991. Identification of a reinforcement pathway necessary for operant conditioning of head waving in Aplysia californica. Behav. Neural Biol. 55: 313-337.
Cueller, H. S., and O. Cueller. 1977. Refractoriness in female lizard reproduction: a probable circannuai clock. Science 197: 495-497.
Dudek, F. E., G. Weir, J. Acosta-Urquidi, and S. S. Tobe. 1980. A secretion from neuroendocrine bag cells evokes egg release in vitro from ovotestis of Aplysia californica. Gen. Comp. Endocrinol. 40: 241-244.
Elliott, J. A. 1976. Circadian rhythms and photopenodic time mea- surement in mammals. Fed. Proc. 35: 2339-2346.
Eskin, A. 1971. Properties of the Aplysia visual system: in vitro en- trainment of the circadian rhythm and centrifugal regulation of the eye. Z r?/. Physiol. 74: 353-371.
Follett, B. K., and P. J. Sharp. 1969. Circadian rhythmicity in pho- toperiodically induced gonadotropin release and gonadal growth in the quail. Nature 223: 968-97 1 .
Follett, B. K., D. T. Davies, and V. Magee. 1975. The rate of testicular development in Japanese quail (Coturnix cotumix japonica) following stimulation of the extra-retinal photoreceptor. Experienlia 31: 48- 49.
Foster, R. G., and B. K. Follett. 1985. The involvement of rhodopsin- like photopigment in the photoperiodic response of the Japanese quail. / Comp. Physiol. A 157: 519-528.
Gomot, A. 1990. Photopenod and temperature interaction in the de- termination of reproduction of the edible snail. Helix pomatia. J. Reprod. Pert. 90: 581-585.
Hamner, W. M. 1967. The photorefractory period of the house finch. Ecology 49: 211-227.
Heller, E., L. K. Kaczmarek, M. W. Hunkapiller, L. E. Hood, and F. Strumwasser. 1980. Purification and primary structure of two neu- roactive peptides that cause bag cell afterdischarge and egg-laying in Aplysia. Neurobiology 77: 2328-2332.
Jacklet, J. W. 1969. Circadian rhythm of optic nerve impulses recorded in darkness from isolated eye of Aplysia. Science 164: 562-563.
Jacklet, J. W. 1980. Light sensitivity of the rhinophores and eyes of Aplysia. J. Comp. Physiol. 136: 257-262.
Jackson, G. L., M. Gibson, and D. Kuehl. 1988. Photoperiodic dis- ruption of photorefractoriness in the ewe. Biol. Reprod 38: 127-134.
Kaczmarek, L. K., and F. Strumwasser. 1981. The expression of long lasting afterdischarge by isolated Aplysia bag cell neurons. / Neurosci. 1: 626-634.
Kaczmarek, L. K., K. Jennings, and F. Strumwasser. 1978. Neuro- transmitter modulation, phosphodiesterase inhibitor effects, and cyclic AMP correlates of afterdischarge in peptidergic neuntes. Proc. Natl. Acad. Sci. 75: 5200-5204.
Kaczmarek, L. K., K. R. Jennings, and F. Strumwasser. 1982. An early sodium and late calcium phase in the afterdischarge of peptide-se- creting neurons in Aplysia. Brain Res. 238: 105-1 15.
Kupfermann, I. 1970. Stimulation of egg laying by extracts of neu- roendocrine cells (bag cells) of abdominal ganglion of Aplysia. J Neurophysiol. 33: 877-881.
Kupfermann, L, and E. R. Kandel. 1970. Electrophysiological properties and functional interconnections of two symmetrical clusters (bag cells) in abdominal ganglion of Aplysia. J Neurophysiol. 33: 865-876.
14
N. L. WAYNE AND G. D. BLOCK
Legan, S. J., and F. J. Karsch. 1983. Importance of retinal photore-
ceptors to the photoperiodic control of seasonal breeding in the ewe.
Biol. Reprod. 29: 316-325. Lees, A. D. 1966. Photoperiodic timing mechanisms in insects. Nature
210: 986-989. Licht, P. 1967. Environmental control of annual testicular cycles in the
lizard Anolis carolinensis. I. Interaction of light and temperature in the
initiation of testicular recrudescence. / Exp. Zoo/. 165: 505-516. Lukowiak, K., and J. W. Jacklet. 1972. Habituation and dishabituation:
interactions between peripheral and central nervous systems in Aply-
sia. Science 178: 1306-1308. Mahon, A. C., and R. H. Scheller. 1983. The molecular basis of neu-
roendocnne fixed action pattern: egg laying in Aplysia. Cold Spring
Harbor Symp. Quant. Biol. 48: 405-412. Nagle, G. T., S. D. Painter, K. L. Kelner, and J. E. Blankenship.
1985. Atrial gland cells synthesize a family of peptides that can
induce egg laying in Aplysia. J Comp Physiol B 156: 43-55. Pinsker, H. M., and D. W. Parsons. 1985. Temperature dependence
of egg laying in Aplysia brasiliana and A. califomica. J Comp. Physiol.
B 156: 2 1-27. Reiter, R. J. 1969. Pineal function in long term blinded male and
female golden hamsters. Gen. Comp. Endocrinol. 12: 460-468. Robinson, J. E., and B. K. Follett. 1982. Photoperiodism in Japanese
quail: the termination of seasonal breeding by photorefractoriness.
Proc. R. Soc. London B 215: 95-1 16.
Robinson, J. E., and F. J. Karsch. 1984. Refractoriness to inductive day lengths terminates the breeding season of the Suffolk ewe. Biol. Reprod. 31: 656-663.
Rowan, W. 1926. On photoperiodism, reproductive periodicity, and the annual migrations of birds and certain fishes. Proc Boston Soc. Natl. Hist. 38: 147-189.
Saunders, D. S. 1979. Insect Clocks. Pergamon Press, New York. Pp. 102-107.
Scheller, R. H., J. F. Jackson, L. B. McAllister, J. H. Schwartz, E. R. Kandel, and R. Axel. 1982. A family of genes that codes for ELH, a neuropeptide eliciting a stereotyped pattern of behavior in Aplvsia. Cell 28: 707-7 19.
Sokolove, P. G., E. J. McCrone, J. van Minnen, and W. C. Duncan. 1984. Reproductive endocrinology and photoperiodism in a ter- restrial slug. Pp. 189-203 in Photoperiodic Regulation of Insect and Molluscan Hormones. Pitman, London.
Strumwasser, F., J. W. Jacklet, and R. B. Alvarez. 1969. A seasonal rhythm in the neural extract induction of behavioral egg-laying in Aplysia. Comp. Biochem. Physiol. 29: 197-206.
Stuart, D. K., A. Y. Chiu, and F. Strumwasser. 1980. Neurosecre- tion of egg-laying hormone and other peptides from electrically active bag cell neurons of Aplysia. J Neurophysiol. 43: 488- 498.
Turek, F. W., and C. S. Campbell. 1979. Photoperiodic regulation of neuroendocrine-gonadal activity. Biol. Reprod. 20: 32-50.
Reference: Biol. Bull. 182: 15-30. (February, 1992)
The Development and Larval Form of an Echinothurioid Echinoid, Asthenosoma ijimai, Revisited
S. AMEMIYA1 AND R. B. EMLET2*
lMisaki Marine Biological Station, Miura-shi, Kanagawa 238-02, Japan and -Department of Biological Sciences, University of Southern California, Los Angeles, California 90089-0371
Abstract. The modified development from cleavage to late larval form of the echinothurioid echinoid, Asthe- nosoma ijimai, was re-examined using light microscopy and scanning electron microscopy of whole and sectioned stages. Although an original study (Amemiya and Tsu- chiya, 1979) reported direct development without evi- dence of a pluteus larva, we found that the unusual de- velopment can be interpreted as a topologically reflected, reduced pluteus, with vestigial larval arms and a greatly reduced larval skeleton. This developmental pattern pro- duces the third and most reduced pluteus form known among the six echinoid lineages with modified develop- ment that have been studied thus far. Features such as an equal fourth cleavage, extrusion of yolk into the blastocoel, and the presence of large numbers of cells within the blas- tocoel are convergent with traits reported for other species with modified development. Coelom formation is clearly modified from that of species with feeding larval devel- opment, but notably the hydrocoel begins to develop po- dial buds prior to separation from the archenteron. Echinothurioid sea urchins are considered to be the most primitive living euechinoids, and in A. ijimai the timing of mesenchyme cell ingression and the formation of epi- neural folds were similar to these features in other eue- chinoids. Indentation of the juvenile oral surface relatively late in larval development raises the possibility that the amniotic invagination (vestibule), common in all other euechinoids, may be a trait incorporated into the devel- opment of echinoids at the time of origin of the echino- thurioids. The structural comparisons reported here show
Received 31 July 1991; accepted 25 November 1991. * Order of authorship was determined alphabetically.
a need for further detailed morphological studies of de- velopmental modifications in other echinoid species.
Introduction
Most sea urchins species (ca. 66%) develop through a feeding larval stage, the echinopluteus, for several to many weeks before metamorphosing into juvenile echinoids (see review by Emlet et ai, 1987). At least 14 times among living taxa, however, the feeding larval stage has been lost, and these species undergo a modified and abbreviated development before juvenile sea urchins are formed (Em- let, 1990; see also Strathmann, 1978; Raff, 1987). At pres- ent, about ten species from six of the lineages with mod- ified development have been investigated, and some de- scription of their embryonic and larval development is available. Descriptive and analytical research has been conducted on the following: two cidaroids (Phyllacanthus imperialis, Olson et ai. 1988; P. pan'ispinus, Mortensen, 1921; Parks et ai, 1989); two echinothurioids (Astheno- soma ijimai, Amemiya and Tsuchiya, 1979; A. sp., Uehara and Amemiya, unpub. obs.); one temnopleuroid (Ho- lopneustes purpiirescens, V. Morris, Univ. Sydney, un- pub.); one echinometrid (Heliocidaris erythrogramma, Mortensen, 1921; Williams and Anderson, 1975; Parks et ai, 1988); two clypeasteroids (Peronella japonica, Mor- tensen, 1921; Okazaki and Dan, 1954; Okazaki, 1975; P. rubra. Amemiya and Emlet, unpub. obs.); and two brooding spatangoids (Abatus agassizi, Larrain. 1973; A. cordatus, Schatt, 1985, 1988). Many of the other echinoid lineages with modified development occur in deeper seas or antarctic seas and are difficult to collect for study (e.g.. lineages oftemnopleuroids, holasteroids and other lineages of cidaroids and spatangoids, <.•./.", Mortensen, 1936; Fell, 1976).
15
16
S. AMEMIYA AND R. B. EMLET
The most recent studies have focused on species with very highly modified development, often referred to as direct development. These studies have examined heter- ochrony (changes in relative timing) of developmental events, modifications of cleavage patterns, the resultant cell lineages, and cell movements (e.g.. Parks et al.. 1988, 1 989; Wray and Raff, 1989, 1990; Henry and Raff, 1990). Additional immunofluorescence studies have drawn in- ferences about gene expression from specific markers for gene products [e.g., the monoclonal antibody B2C2 to mesenchyme-derived antigens or antibodies to seroton- ergic neurons (above citations, Bisgrove and Raff, 1989)]. Due to the extreme degree of developmental modifica- tions, these studies usually emphasize how different the morphogenetic patterns are from those of species that de- velop through a pluteus larva (e.g., Heliocidaris erythro- gramnia, Wray and Raff, 1990). With the exception of the above mentioned studies on Ahatus cordatits, Helio- cidaris erythrogramma, and Peronellajaponica, either no information or only limited information is available on the internal morphological aspects of development of species with modified development. Both the extreme modification, and a lack of detailed morphological infor- mation, limit our understanding of how these develop- mental modifications may have evolved.
This report — an extension of an earlier one (Amemiya and Tsuchiya, 1979) — describes selected features of mor- phogenesis in the echinothurioid echinoid Asthenosoma ijimai, from cleavage through late larval development. Comparisons are also made with unmodified pluteal de- velopment, as well as with the modified developmental patterns occurring in other species. The echinothurioids are a particularly important group to study: first, because all of them seem to have modified development (reviewed in Emlet et al., 1987); and second, because they are con- sidered to be the earliest living branch of the euechinoid lineage, and thus the second oldest lineage of echinoids after the cidaroids (Smith, 1984). Because the cidaroids and euechinoids differ in many developmental features (Emlet, 1988), we should inquire whether the develop- ment of the echinothurioids shows greater affinity with other euechinoids, or with the more primitive cidaroids. The observations presented here, together with the com- parisons with other species, point up the need for addi- tional morphological studies of developmental modifi- cations in other echinoid species.
Materials and Methods
Adults and larvae
Adults of Asthenosoma ijimai Yoshiwara were collected at a depth of 20 m off Misaki Marine Biological Station in Sagami Bay, Japan. Adult specimens were dissected, and fully matured gametes were obtained. Eggs were
washed twice in filtered seawater (0.22 /urn) and fertilized by mixing with a small amount of undiluted sperm. Fer- tilized eggs were washed three times in filtered seawater, and cultured in unstirred, one-liter glass beakers at 20°C or at room temperature (25-28°C). The stages and times for sectioned material presented here are from cultures at room temperature. No food was added to the larval cul- tures. At various times after fertilization, living larvae were photographed under a dissecting microscope, and aliquots were fixed for examination by light or scanning electron microscopy (SEM).
Preparation of sectioned and stained material
Larvae were fixed for 1 h in seawater containing 4% or 10% formalin at room temperature and preserved in 70% EtOH. Preserved specimens were dehydrated through graded ethanol series and embedded in Spurr embedding media (Polysciences, Inc). Sections, 5-8 f*m thick, were stained with Richardson's stain (1% Azure II in distilled water combined with 1%. methylene blue in 1% sodium borate, Richardson et al., 1960). The serial sections of larvae embedded in epoxy resin were traced with camera lucida and digitized so that 3-D images of the sections could be constructed (PC3D program, Jandel Scien- tific, Inc.).
Immunofluorescence and H33258 staining
Immunofluorescence staining with skeletogenic mes- enchyme specific monoclonal antibody B2C2 was con- ducted according to the methods of Parks et al. (1988). Embryos, larvae, and juveniles were fixed for 50 min in seawater containing 4% formalin, washed in artificial sea- water, dehydrated in a graded ethanol series, embedded in polyester wax (BDH, Ltd), and sectioned at a thickness of 5 ^m. The rehydrated sections were washed with phos- phate-buffered saline containing 0.05% Tween 20 (PBS- TW20) and incubated with culture fluid containing the monoclonal antibody B2C2, diluted 1 :20 in PBS-TW20, for 40 min at room temperature in a humidified chamber. The slides were washed in PBS-TW20, incubated with FITC-conjugated, goat anti-mouse, IgG antibodies (di- luted 1:200 in PBS-TW20) for 40 min, and rinsed again. For detection of cell nuclei, some sections were incubated with H33258 (Hechst, Inc.) at a concentration of 0.5 /ug/ ml PBS for 10 min instead of, or after, treatment in pri- mary and secondary antisera. Fluorescence was observed and photographed with a Nikon fluorescence microscope.
Scanning Electron Microscopy (SEM)
Specimens were fixed and preserved as indicated above, or they were fixed for 1 h in a mixture of 2% gluteraldehyde (Taab Lab.) and 1% osmium tetroxide (OsO4, Taab Lab.)
ECHINOTHURIOID DEVELOPMENT, REVISITED
17
Figure 1. Early embryonic stages ofAsthenosoma ijimai. a. Sixteen-cell stage embryos randomly oriented, show that all cells are approximately the same diameter after the fourth cleavage. Scale bar. 1 mm. b. SEM of a 21.5-h, lobate blastula. Arrowheads mark pits in ectoderm. Scale bar, 1 mm. c. Close-up SEM of ectodermal pit marked by right arrow in b. Scale bar, 25 nm. d. Section of a 2 1 ,5-h blastula shows yolky cytoplasm in the blastocoel and ectodermal pits (arrowheads). Same scale as b.
in 0.45 M sodium acetate buffer (pH 6.4) at room tem- perature (Harris and Shaw, 1984) and preserved in 70% EtOH at 4°C. The preserved specimens were dehydrated through a graded ethanol series, dried at the critical point (Hitachi HCP-1 drier) with liquid CO: as a transitional fluid, and sputter-coated with gold (Eiko IB-3 ion coater). Observations were made with a Hitachi HHS-2R SEM. To examine the inside of larvae with SEM, specimens were embedded in polyester wax and sectioned by micro- tome to expose a particular cross section. These specimens were incubated in absolute ethanol at 40°C for 12 h to remove wax (Armstrong and Parenti, 1973) and then subjected to critical point drying as described above.
Clearing lan>ae
Live larvae ofAsthenosoma are opaque orange-yellow (Amemiya and Tsuchiya, 1979), and it was impossible to see internal structures in these or in fixed, preserved spec-
imens. However, larvae could be rendered translucent by clearing with solutions of benzyl benzoate and benzyl al- cohol mixed in ratios of 2:1, 1:1, or 1:2, depending on the desired refractive index. Fixed larvae were first de- hydrated to 100% EtOH, then transferred into the clearing solution where the remaining EtOH was allowed to evap- orate. Upon clearing, larval dimensions remained the same, and no osmotic effects were discerned. To search for calcareous deposits, cleared larvae were observed under crossed-polarized light.
Results
Obsen'ations on soft-tissue development
Eggs, cleavage, external aspects of larvae, and meta- morphosis have been described by Amemiya and Tsu- chiya (1979). The fourth cleavage of embryos ofAsthen- osoma ijimai was almost equal, giving rise to 16 blasto- meres of similar size. Figure la shows that there was some
18
S. AMEMIYA AND R. B. EMLET
•^•'^W-vi '- .
1
^£%l^*t J
•>,. " •« * *£>«**,1GsU5l ^ "<V-3,
- r <-?"•» -~,J*G!.-<&*?'fj,'** **' t
rf^^^^^l* IU^>**Jfei*^il3
&nt^>«wl
» JSJ^lf'1*! ^^?«
Figure 2. Later embryonic stages of Asthenosoma ijiinai. All embryos and sections are oriented with the animal pole up. a. Light micrograph of a live, 25.5-h-old, early gastrula. Same scale as b. b. Section through animal-vegetal axis of flattened gastrula (25.5 h), showing blastocoel filled with yolky cytoplasm. Scale bar, 1 mm. c. Lateral wall of gastrula (25.5 h), shows yolky cytoplasm exocylosing from basal ends of ectodermal cells. Same scale as d. d. Vegetal wall of gastrula (25.5 h), showing exocytosis of cytoplasm and ingression of cells. Scale bar, 200 urn. e. Section of a 25.5-h gastrula shows fluorescently staining nuclei (H33258 fluorescent dye) of ectodermal and probable mesenchyme cells. Same scale as f. f. Section of 35- h gastrula, with fluorescently stained nuclei. Scale bar, 200 Mm. g. Section of 51.5-h embryo, stained with B2C2 antibody, arrow shows first occurrence of expression of MSP- 130 glycoprotein associated with blas- tocoelic cells. Scale bar, 50 ^m. h, i. Section of an 88.5-h larva, doubly stained with B2C2 antibody (h) and H33258 fluorescent dye (i) shows not all blastocoelic cells express MSP- 130. The concentrated clusters of nuclei in (i) are epithelia of the archenteron (A) and a coelomic compartment (C). Scale bar, 200 ^m.
variation in size of the blastomeres in 16-cell embryos, but there was no evidence of micromeres at the vegetal pole. The absence of an unequal, fourth cleavage parallels the cleavage patterns of other echinoid species large yolky eggs and modified development (Williams and Anderson, 1975; Raff, 1987; Parks et til., 1989).
Like other echinoderm species with yolky eggs, embryos of Asthenosoma ijimai formed wrinkled blastulae (Ame- miya and Tsuchiya, 1979; Parks et at., 1989). This stage was followed by egression and loss of wrinkles and led to a lobate blastula (Fig. Ib). At this stage, small, cylindrical pits were present on the external surface of the embryo
and passed into the ectodermal layer (Fig. Ib, c, d). Serial sections showed that some of these pits terminated blindly within the ectoderm, while others passed through the ec- toderm to another external opening. Several of these pits or passages coincided with large indentations in the em- bryo's surface and, therefore, may have been the remnants of the wrinkled indentations. Parks et at. (1989) reported pits in the yolky embryo of the cidaroid Phyllacanthus pan'ispimts, though they did not see any association of pits with egression tracks, where wrinkles diminished. Further work is necessary to determine whether the pits in the two species arise by similar mechanisms.
ECHINOTHURIOID DEVELOPMENT. REVISITED
19
Figure 3. A 5 1 .5-h larvae of Asthenosoma ijimai, all with anterior end up. a. Light micrograph of a live larva with dorsal swelling on the right. Scale bar, 1 mm. b. Medial sagittal section through late gastrula. The tip of archenteron (A) has curved toward the ventral surface. Same scale as a. c. Frontal section shows a small outpocketing on the left side of the archenteron (A) that extends dorsally in other sections. Scale bar. 200 /im.
Sections of this 21.5 h stage showed anucleate, yolky cytoplasm being released into the blastocoel from the basal ends of most ectodermal cells (Fig. Id, see also Fig. 2c, d). Sections also revealed that one indented surface was extruding considerably more material than other surfaces into the blastocoel (Fig. Id). The fluorescent stain, H33258, revealed considerable numbers of nuclei in the blastocoel, indicating mesenchyme-like cell ingression at the onset of gastrulation.
Gastrulation had begun by 25.5 h after fertilization, and embryos were compressed along the animal-vegetal axis, with a large indentation at the vegetal pole (Fig. 2a). At this stage, the blastocoel was filled with yolky cytoplasm (Fig. 2b, c, d). Counts of fluorescently stained nuclei showed a mean of 98 cells per 5-^m section (n = 3 sections, S.D. = 11) and were scattered among the yolky cytoplasm in the blastocoel (Fig. 2e). The number of staining nuclei, and thus the number of cells, increased to a mean of 375 per section (n = 3 sections, S.D. = 9.6) in the mid-gastrula stage at 35 h (Fig. 2f). The source of these additional blastocoelic cells is either ingression from the vegetal pole (Fig. 2d) or cell division. Fluorescent staining also showed that cell division in the ectoderm was continuing because the number of nuclei in the ectoderm increased between 25.5 and 35 h (Fig. 2e, f).
A positive reaction of B2C2 antibody with blastocoelic cells, indicating expression of MSP- 1 30 glycoprotein, was
found first at 51.5 h post fertilization (Fig. 2g). Later ob- servations on mesenchyme cells associated with skeleton in larvae of Asthenosoma ijimai showed that these cells reacted with B2C2, suggesting that MSP- 130 is expressed by the skeletogenic cells in this species just as in other echinoids. By 88.5 h post fertilization, after skeletogenesis had begun, sections labeled with both nuclear stain (H33258) and B2C2 antibody revealed that only a fraction of blastocoelic cells were skeletogenic (Fig. 2h, i).
In the present paper, the identities of the dorsal and ventral surfaces are reversed from those described in the initial paper on development of Asthenosoma ijimai (Amemiya and Tsuchiya, 1979). By 35 h post fertilization, the embryos elongated along the animal-vegetal axis, with the blastopore located off center, toward the ventral sur- face. At 5 1 .5 h, the late gastrulae had swollen dorsal sides (Fig. 3a). Internally, the archenteron had grown over half the length of the embryo, and the apical (anterior) end curved toward the ventral side of the embryo (Fig. 3b). In addition to the large curved tip of the archenteron, another small outpocketing was forming on the left side of the archenteron and was growing dorsally (Fig. 3c).
Serial sections of stages at 51.5, 56.5, and 63 h post- fertilization showed progressive changes in the develop- ment of the archenteron and coelomic pouches (Fig. 4). When the archenteron reached its full length, one or two slender, epithelial projections on the left side and an-
20
S. AMEMIYA AND R. B. EMLET
DORSAL VIEW
Ls
LEFT LATERAL VIEW
/ ^/\h d*/
Figure 4. Three-dimensional reconstructions of the archenteron and coelomic pouches from serial sections of 51. 5-, 56.5-, and 63-h larvae ofAsthenosoma ijiinai. In each column the same larva is shown in approximate dorsal view and approximate left lateral view. The animal pole, corresponding to the anterior end, is toward the top of the figure: the vegetal pole, corresponding to the posterior end, is toward the bottom of the figure. Line segments are tracings of the inner surface of the archenteron and coelomic pouches (black). The outer surface of the larval ectoderm (gray) is included in a. c, d and f for reference. Small arrows, left lateral and dorsal projections from archenteron; large arrow, tip of archenteron; Rs, right somatocoel; Ls. left somatocoel; H, hydrocoel; Pb, podial bud of hydrocoel. See text for explanation, a., b. 51.5 h. c.. d. 56.5 h. e., f. 63 h. Scale bar. 1 mm.
terodorsal surface of the archenteron grew dorsally (small arrows, Fig. 4a, b). Of the three 51.5-h specimens that were serially sectioned, two showed both projections (Fig. 4a, b), and one showed a left lateral projection only (not figured). With further development, the much larger ven- tral tip of the archenteron bent and extended ventrally without contacting the blastocoel wall (large arrow. Fig. 4b, c). Between 51.5 and 56.5 h, the archenteron under- went torsion, twisting approximately 90° counterclock- wise, when viewed from the animal pole. This twist re- oriented the tip of the archenteron toward the left side of the larva, and the slender projections toward the right side of the larva (Fig. 4b, c). One of the two slender pro- jections extended laterally and posteriorly (toward the vegetal end) to form the right somatocoel (Fig. 4c-f). Comparisons among serially sectioned larvae suggest that either of the slender projections could form the right so- matocoel, and the other slender projection apparently did not continue to grow. Subsequent to 56.5 h, the tip of the archenteron grew a projection dorsally and posteriorly, which formed the left somatocoel (Fig. 4e, f). By 63 h.
while still attached to the main body of the archenteron, the tip of the archenteron began to develop into the left hydrocoel with buds that became the coelomic lining of the five, primary podia (Fig. 4e, f; Fig. 5c, d).
Externally, changes between the 56.5 and 63 h stages produced four large and rounded lobes that grew into projections called "para-arms" by Amemiya and Tsuchiya (1979). One pair of these bilaterally symmetrical projec- tions is located dorsally and laterally relative to the blas- topore and projects posteriorly, away from the animal pole. The second pair is located on the dorsal surface just anterior to the other pair and also projects dorsally (Fig. 5a). The surfaces of the larvae were uniformly ciliated (Fig. 5b). No developing stages, even in the region of the para-arms, showed cilia collected into discrete rows such as found in the ciliated bands of pluteus larvae. No dorsal hydropore was present despite internal development of somatocoels and the left hydrocoel.
By 75.5 h post-fertilization, the five bulges of the pri- mary podia were externally visible and arranged in a circle on the left lateral surface (Fig. 6a). Sections of this stage
ECHINOTHURIOID DEVELOPMENT, REVISITED
21
V?**
Figure 5. A 63-h larvae of Asthenosoma ijimai. all oriented with the anterior end toward the top of the figure, a. SEM of a whole larva, right ventral view (dorsal side is on the left) shows two right para-arms (arrows) and the blastopore (Bp). Scale bar. 0.5 mm b. Close-up of uniformly ciliated epidermis. Scale bar, 25 nm. c. Frontal section shows the leftward oriented archenteron with coelomic components of podia near its tip. The hydrocoel (H) is developing before being separated from the archenteron (A). (This larva was damaged during embedding, but a clear interpretation of sections was still possible.) Same scale as a. d. Higher magnification of a more dorsally located frontal section from same larva as c. Hydrocoelic components of podial buds (Pb) are present. Ls. left somatocoel. Scale bar. 200 ^m.
22
S. AMEMIYA AND R. B. EMLET
Figure 6. A 75.5-h larvae of Asthenosoma ijimai. a. Light micrograph of live specimen shows five primary podia just beginning to form on left side of larva. Scale bar, 0.5 mm. b. Frontal section with continued hydrocoelic (and podial) development. The hydrocoel (H) is almost completely separated from the archenteron (A). Same scale as a. c. Detail of hydrocoel (H) with parts of two podial extensions from a different section of the same larva as b. Scale bar. 100 ^m.
showed the hydrocoelic compartments with thickened epithelia beneath the podial swelling of the ectoderm (Fig. 6c). Serial sections revealed that the connection between the hydrocoel and archenteron was greatly reduced in one larva (Fig. 6b) and completely severed in a second larva examined. All coelomic and archenteric cavities contained stained materials that appeared to be yolky cytoplasm and some cells (Fig. 6b, c).
By 101 h post fertilization, primary podia elongated to 0.2 mm length (Fig. 7b, c, d). Sections of the juvenile oral surface showed folds of ectodermal tissue lying between the five primary podia (Fig. 7c, d). These folds were evi- dently epineural folds that were growing over the juvenile oral surface to form the epineural sinus (von Ubisch, 1913; Hyman, 1955; Emlet, 1988). SEM observations of the external surface of the developing juvenile oral region confirm that these epineural folds were spreading toward the oral center (Fig. 7f-h).
Coincident with the lengthening of the primary podia and development of the epineural folds, the oral surface sank to become indented in the surface of the developing larva. This indentation was notable in live specimens viewed from the side at 101 h (Fig. 7b), as well as in sec- tioned material (Fig. 7c) and in specimens fixed for SEM (Fig. 7f ). Though the developing juvenile oral surface was never deeply enclosed as occurs within the amniotic in- vagination (or vestibule) of the euechinoids, the oral sur- face was further sunken in living larvae nine days post- fertilization (Fig. 8a). Fourteen days after fertilization, the oral surface was no longer evidently sunken, and the larval para-arms and anterior yolky mass have moved away from
the oral surface toward the aboral surface of the juvenile (Fig. 8b, c).
At 101 h post-fertilization, a hydropore was evident on the dorsal surface of the larva (Fig. 7a). The location of this pore was near the median side of the base of right anterior para-arm. Sections of 101-h-old larvae showed that the hydropore was joined to the hydrocoel via a canal lined by a thick epithelium (Fig. 7e). In sections of younger larvae (88.5 h), this hydroporic canal invaginated from the larval surface but was not yet joined to the coelomic cavities. Sections of 14-day larvae showed the hydropore connected to the hydrocoel by a stone canal (Fig. 8c, e). Also by this stage, epineural folds had joined to form an epineural sinus (Fig. 8d, e).
Observations on the calcitic skeleton
Larval stages at 58, 63, 75.5, 88.5. and 101 h after fer- tilization were cleared to look for calcareous skeletal spic- ules within developing embryos. No evidence of calcifi- cation was seen in 58- and 63-h specimens, even though the latter had begun to form the para-arms (Figs. 4f and 9a). The first evidence of calcification was found in 75.5- h specimens (Fig. 9b). In these, para-arms were well formed, and podial bulges had just begun to form. One calcareous plate-like ossicle was embedded in the base of each para-arm. In the more advanced 75.5-h specimen of the two observed, a fifth calcareous ossicle was present and located centrally between the four para-arms (Fig. 9b). In 88.5-h specimens, the five ossicles had grown into plates, and those in the para-arms had formed fenestrated rods that projected toward the distal ends of the para-
ECHINOTHURIOID DEVELOPMENT, REVISITED
23
Figure 7. A 101-h larvae of Asthenosoma ijimai. a. Light micrograph of dorsal side of live specimen. Note the hydropore (Hp) and four para-arms (to right). The anterior end is to the left of figure. Scale bar, 0.5 mm. b. Light micrograph of ventral side of live specimen. The anterior end is to the right of figure. Same scale as a. c. Medial frontal section through larva shows developing internal structures and juvenile oral region. P, podia; Rs, right somatocoel; Ls. left somatocoel; G. remnant of archenteron and future gut. Same scale as a. d. Close-up of juvenile oral region, with podia (P). epineural folds (Ef), radial canals of water vascular system (R). and left somatocoel (Ls). Scale bar. 200 pm. e. Section at the level of the hydropore shows mvaginated canal (He). In an adjacent section, the canal joins the hydrocoel. Scale bar, 200 nm. f. SEM of oral region of larva, shows five podia, and bulges for spines (Sp) sunken into the left larval surface. Scale bar. 0.5 mm. g. Close-up SEM of oral region showing inward movement of epineural folds (Ef) between podia (P). The infolding epidermis is strongly ciliated whereas the original floor of the oral region is sparsely ciliated. Scale bar. 200 urn. h. High magnification view of a single epineural fold (Ef) moving between two adjacent podia (P). Scale bar. 50 ftm.
arms (Fig. 9c). These rods were particularly well developed in the two right para-arms and had just begun to form in the two left para-arms. The centrally located plate showed no evidence of an attached rod. Each of the calcined skel- etal plates, with or without rods attached, behaved opti- cally like a single crystal when rotated through polarized light (Fig. 9d-f). This observation confirmed the structural appearance that plates with attached rods were a single
skeletal unit. Also in the 88.5-h larvae, several other cal- cification centers had formed and ossicles were growing (Fig. 9c).
Calcification in 101-h larvae was even more developed (Fig. 9g). These larvae had well-developed podial buds (Fig. 7b) and, on one specimen, the buds for spines were developing on the circumference of the juvenile oral sur- face (see Fig. 7f). As with earlier stages, fenestrated rods
24
S. AMEMIYA AND R. B. EMLET
Figure 8. Later stages of larval development of Asthenoxoma ijimai. For all specimens, the anterior end is to the right of figure, a. Ventral side of live specimen nine days after fertilization. Scale bar, 0.5 mm. b. Ventral side of live specimen 14 days after fertilization. The larval para-arms and anterior yolky mass have been contorted toward the juvenile aboral surface. P, podia; Sp, spines. Same scale as a. c. Fourteen-day post fertilization, approximate frontal section at the level of the hydropore and hydroporic canal (He). Same scale as a. d. Close-up of juvenile oral region showing epineural sinuses (Es), gut (G), water vascular system (W), radial canal (R), and podia (P). Compare with e. Scale bar. 200 ^m. e. SEM of partially sectioned specimen showing similar structures as seen in d. He, hydroporic canal; Rs, right somatocoel (aboral part of body cavity); Ls, left somatocoel (oral part of body cavity). Scale bar, 200 pm.
were associated with plates in the para-arms and not with other ossicles. In one larva, each of the spine buds con- tained a growing spicule. In this same larva, the two os- sicles of the left para-arms and three additional ossicles formed a circle beneath the juvenile oral surface that rep- resented the five ocular plates of the adult skeleton.
Discussion
Larval structure o/'Asthenosoma ijimai
Our re-examination of the larval development of As- thenosoma ijimai has demonstrated several morphological features that were not reported in the initial study of this species. Amemiya and Tsuchiya (1979) reported that the early post-gastrula of .1. ijimai resembled an early bi- pinnaria and not a prism larva. That study also reported the appearance of para-arms later in development and distinguished these projections from pluteus larval arms, because the former apparently lacked larval spicules and apparently arose from different regions of the larva. On this basis Amemiya and Tsuchiya concluded that, during development, A. ijimai passes from the gastrula stage to
metamorphosis without showing any evidence of a pluteus larval form. They also concluded that the development of A. ijimai represents a second example of direct devel- opment (sensu Hyman, 1955) for an echinoid, the first being that of Heliocidaris erythrogramma (development originally described by Mortensen, 1921, but also by Wil- liams and Anderson, 1975). Amemiya and Tsuchiya (1979) identified the surface on which para-arms arose in embryos of Asthenosoma as the ventral surface because of its resemblance to the ventral (oral) surface of early bipinnaria larvae of asteroids. Amemiya and Tsuchiya ( 1979) also incorrectly stated that the five primary podia were on the ventral surface, although Amemiya (1980) reported that primary podia arise lateral to the ventral surface. In the present study, the surface on which the para-arms arose has been identified as the dorsal surface based on observations of internal structures and on com- parison with the primitive pluteus morphology. In this new orientation, the primary podia form on the left side of the larva.
A number of newly observed structures and their po- sitions lead us to reinterpret the larval development of
ECHINOTHUR1OID DEVELOPMENT. REVISITED
25
Figure 9. Skeletal development in various, cleared stages of larvae of Asthenosoma ijimai. All larvae are viewed from the dorsal side in partially polarized light, a. A 63-h larva shows no evidence of calcareous skeletal elements. Scale bar, 0.5 mm. b. Two 75.5-h specimens show the very first signs of skeletal development. One calcareous element is associated with each para-arm. The specimen on the right was an additional calcareous element centrally located between the para-arms. Same scale as a. c. A 88.5-h larva with continued skeletal development. Each calcareous element associated with a para-arm has formed a plate-like ossicle and shows substantial or initial formation of a rod attached to the plate. Other calcification centers have also begun. Same scale as a. d. Close-up of plate-like ossicle and rod from right posterior para-arm of a 88.5- h larva. Scale bar. 100 ^m. e. Another plate-like ossicle and rod from a 101-h larva. Same scale as d. f. Central plate-like ossicle without an associated rod from a 101-h larva. Scale bar, 100 ^m. g. A 101-h larva with manv calcification sites. Same scale as a.
Asthenosoma ijimai as that of a highly modified pluteus larva. The two pair of bilaterally symmetrical para-arms arising from posterior and dorsal parts of embryo, each one containing a calcareous, fenestrated skeletal element, appear to be vestigial larval arms. We reject an alternative
interpretation that the fenestrated rods are juvenile spines, because the spines form in association with plates that are separate elements (Gordon, 1926a, b). Because the skeletal elements are fenestrated, we interpret the para- arms as reduced post-oral and postero-dorsal arms (the
26
S. AMEMIYA AND R. B. EMLET
1st and 3rd pairs of arms) of a pluteus. Fenestrated skeletal rods are only known for these arm pairs in pluteus larvae ( Mortensen, 1 92 1 ; Emlet, 1 982). In typical plutei, the sec- ond pair of arms to form is the anterolateral pair that always contains simple calcareous rods (Mortensen, 1 92 1 ). Each anterolateral rod is an outgrowth from the pair of spicules that also form the postoral rods and body skel- eton. The postoral rods are so reduced in A. ijimai that anterolateral rods are absent.
There are also several differences in the early formation of fenestrated spicules in a pluteus and those in A. ijimai.
( 1 ) In the pluteus, a fenestrated rod grows from a triradiate spicule (Okazaki, 1975) and later elaborates a plate at its proximal base (Emlet, 1985. and unpub. obs.). In contrast, skeletal elements in A. ijimai form proximal, reticulate plate-like ossicles that later form reduced, fenestrated rods.
(2) In a pluteus, calcareous rods extend and consequently the arms elongate (e.g., Okazaki, 1975); in larvae of A. ijimai, para-arms are already present before spicules elongate. In actuality, formation of arm buds in the ab- sence of spicules can still occur in plutei (Yasumasu et a/., 1985; Emlet, pers. obs.) indicating that the epidermis of the arm regions is apparently distinct prior to its as- sociation with spicules. This last observation is consistent with the formation of arm buds in A. ijimai.
Additional support for the identification of the para- arms as homologues of the first and third arm pairs of a pluteus larva comes from the following evaluation of arm position. Rather than being directed anteriorly (in the di- rection of swimming) as they are for a pluteus larva, arms and their associated skeletal elements are reflected dorsally and posteriorly at the surface of the very large yolky larva (Fig. 10). In echinoids with plutei, the gastrula forms a prism larva when rods of the first pair of larval spicules lengthen into postoral, body, and anterolateral rods and deform the ectoderm (Horstadius, 1939; Okazaki, 1975). The prism's ventral surface (defined by the association of the archenteron tip with that surface) flattens to become the pluteus oral surface; the prism's dorsal surface (op- posite the ventral surface) distends to become the aboral surface, terminating at the posterior end of the pluteus. During the prism stage, the ciliated band forms and serves as a landmark dividing oral and aboral ectoderm (c.f., Davidson, 1 986). The postoral arms grow anteriorly from the positions lateral to the blastopore. Late in the four- armed stage, a second pair of triradiate spicules appears at dorsolateral edges of aboral surface near the ciliated band (see Fig. 10), and these form the (usually) fenestrated posterodorsal rods (e.g., Mortensen, 1921; Okazaki, 1975). The postoral and posterodorsal arms thus extend the cil- iated band anteriorly and are located at the edge of the concave oral and convex aboral ectoderm. If the posterior end of the pluteus were not convex, and if the aboral surface lay in one plane, the positions of the postoral and
DORSAL VIEW PO{
PD
Figure 10. Schematic of a larva ofAs/lienosonia ijimai and a pluteus (Strongylocentrotusfranciscanus) viewed from dorsal and left lateral ori- entations. In A ijimai the para-arms are reflected posteriorly and contain reduced skeletal elements. These bilaterally symmetric arms and spicules are in positions that can be considered homologous with the postoral (PO) and posterodorsal arms (PD) of the pluteus. The anterolateral arms and rods (AL) have been lost in A. ijimai- Hp, hydropore; S, stomach.
posterodorsal arms of a pluteus would conform with the para-arms of Asthenosoma ijimai (see Fig. 10). This de- scription is consistent with the hypothesis of homology between the identified arms and skeletal elements in plutei and larvae of A. ijimai.
The position of another newly observed structure, the hydropore, is also consistent with and supports this in- terpretation of vestiges of pluteus larval development. In both pluteus larvae and those of Asthenosoma ijimai, the hydropore opens medially to, and anterior of, the bases of the posterodorsal arms (Figs. 10, 7a). A clear difference is, however, that the hydropore opens just after coelom formation in pluteus development and it opens only after advanced coelomic development in A. ijimai.
If the boundary between oral and aboral ectoderm has remained associated with the epidermal regions of the arms, this reinterpretation of the larval form of Asthe- nosoma ijimai implies that the large, rounded, anterior end of the larva is covered by oral ectoderm and that aboral ectoderm may be restricted to that region associated with the para-arms. For A. ijimai. there may be a reversal in the relative area (and shape) of the oral ectoderm and aboral ectoderm compared to that in plutei (Fig. 10). It
ECHINOTHURIOID DEVELOPMENT, REVISITED
27
may be possible to test this hypothesis with cell lineage studies or with immunocytochemical probes to transcripts of the Cylll actin gene or the Spec gene, which are specific to aboral ectoderm in plutei of Strongylocentrotus pur- piiratus (Cox et ai. 1986; Davidson, 1986). Enlargement of the oral ectoderm and reduction of aboral ectoderm has been demonstrated in cell lineage studies of Helioci- daris erythrogramma (Wray and Raff, 1990). Further work will be required to determine whether this apparent similarity represents a new case of parallelism in echinoid developmental patterns.
Comparisons between pluteus development and modified development
Even though larvae of Asthenosoma ijimai retain sev- eral reduced pluteus structures, several other features are partially convergent with other echinoid species that have modified development. Developmental comparisons among species that form plutei, A. ijimai, and other spe- cies with modified development allows inferences about morphogenetic changes that may occur during evolution from pluteus development to highly modified (e.g.. direct) development.
An equal fourth cleavage, documented here for As- thenosoma ijimai, is a common feature of species with highly modified development and is correlated with the production of a large number of mesenchyme cells (Raff, 1987; Parks et al, 1989). Raff (1987) suggested that the large number of mesenchyme cells is a requirement for acceleration of development of the adult rudiment. For A. ijimai. only a fraction of the large number of blasto- coelic cells become skeletogenic and only after a delay relative to species with feeding larvae. A large number of mesenchyme cells is also produced from the relatively large micromeres of an unequal fourth cleavage by em- bryos of Peronella japonica, and some of these cells also produce larval skeletal rods (Okazaki and Dan, 1954; Okazaki, 1975). These comparisons suggest that the loss of the expression of larval skeleton is independent of, and follows amplification of, the cell lineage that putatively produces adult skeleton.
The growth and behavior of the archenteron and coe- loms of Asthenosoma ijimai appears to be intermediate between that of species with pluteus development and that of the other species with modified development. In species with feeding larvae, the archenteron grows into the blastocoel, reaching approximately % of the distance toward the animal pole prior to bending toward, and at- taching to, the blastocoel wall where the larval mouth forms. In most species for which modified development has been described, the archenteron invaginates less than halfway into the blastocoel: Peronella japonica (Morten- sen, 1921; Okazaki, 1975); Heliocidaris erythrogramma
(Williams and Anderson, 1975; Wray and Raff, 1989); Phyllacanthus parvispimis (Parks et al., 1989). In Asthen- osoma ijimai, the archenteron invaginated as much as % of the way into the blastocoel. Unlike what has been re- ported for other species with modified development, in A. ijimai the tip of the archenteron curved toward the ventral surface of the blastocoel and subsequently under- went torsion to the left side of the larva.
Enterocoely, where left and right coelomic pouches are budded off completely from the anterior end of the arch- enteron, is the only means of coelom formation known in pluteus development (e.g., Okazaki, 1975). Both of these pouches divide again to form anterior axocoelic and posterior somatocoelic sacs. The left axocoel subsequently grows a canal to form the dorsal hydropore, and another extension of this sac forms the left hydrocoel (Hyman, 1955). Development of bilateral coelomic pouches fol- lowed by posterior growth and formation of somatocoels also occurs in Asthenosoma ijimai, yet in a distinctive way: two outpocketings from the archenteron form the left and right somatocoels, and precocious hydrocoelic lobes grow from the tip of the archenteron (Figs. 4; 5c, d). No obvious axocoelic sacs and no hydroporic canal are formed during this sequence. Later, a hydropore does form and joins with the water vascular system. In other species with modified development, the coelomic pouches are usually produced in pairs at the tip of the archenteron, with one sac substantially larger than the other (e.g., Wil- liams and Anderson, 1975; Wray and Raff, 1989). Several species with highly modified development are reported to form additional coelomic sacs by shizocoely from aggre- gated mesenchyme cells (e.g., Williams and Anderson, 1975; Schatt, 1985). These patterns suggest that a tran- sition from enterocoely to a combination of enterocoely and shizocoely may take place only after considerable modification of development has already occurred. For species with modified development in general, detailed descriptions of how coelomic pouches give rise to different coelomic sacs are currently lacking, and additional studies are needed.
The orientation of the adult oral-aboral axis relative to the plane of bilateral symmetry of the pluteus is conserved in several species with modified development, but is ap- parently lost in others. Loss of symmetry is not related to the degree of loss of pluteus features. Evidence for reten- tion of pluteus larval symmetry and its relation to juvenile rudiment formation has already been presented for As- thenosoma ijimai. In Phyllacanthus imperialis, with two pairs of larval arms, a reduced preoral region, and no larval mouth, the juvenile oral surface forms on the left side of the larval body (Olson et al., 1988). A reduced bilateral symmetry is also present in Heliocidaris erythro- gramma, which has one coelomic pouch (the left one) larger than the other (Williams and Anderson, 1975) and
28
S. AMEMIYA AND R. B. EMLET
a bilaterally symmetric (larval) serotonergic ganglion (Bisgrove and Raft", 1989). In H. erythrogramma, the coincident arrangement of these two sources of bilaterality provides evidence for conservation of the adult oral-aboral axis (Bisgrove and Raft", 1989). Departures from conser- vation of relative positions of larval and adult axes occur for three other species. Peronella japonica and P. rubra (with very similar development) have bilaterally sym- metric larvae, but the juvenile rudiment forms centrally, with the juvenile oral surface directed anteriorly and dor- sally (Okazaki, 1975; Amemiya and Emlet, unpub. obs.). Loss of the primitive larval-adult arrangement of axes may be related to the retention of one pair of larval arms (pos- torals) and the loss of the more dorsal pair. Retention of only one well-developed pair of larval arms may force the rudiment to develop on the dorsal side, whereas in the absence or reduction of both pairs (e.g., H. erythro- gramma, A. ijimai), or retention of both pairs (P. impe- rialis), the juvenile oral surface would not be shifted. This mechanistic hypothesis does not apply to Phyllacanthus parvispimts which lacks a larval skeleton and has bilobed, asymmetric coeloms which do not coincide with the ori- entation of the serotonergic neurons (Park et a/., 1989). Further morphological studies of the formation and growth of the archenteron and coeloms of both P. japonica and P. parvispimts are needed to determine how primitive larval and adult oral aboral axes have been rearranged.
Eitechinoid characters in echinothurioid development
Early ingression of cells from the blastular wall in As- thenosoma ijimai (Fig. Id) is comparable to primary mes- enchyme ingression, which occurs prior to gastrulation in other euechinoids and is different from the later ingression known among the cidaroids (Schroeder, 198 1 ).
Beginning with the first appearance of podia and con- tinuing until the adult skeleton is well-developed (Fig. 8b, c), the juvenile oral region of Asthe nosoma ijimai sinks into the surface of the larva (Figs. 7b, c, f; 8a). Within this indentation, both podia and oral spines grow. Though there is no early invagination and enclosure of the juvenile oral surface that could be clearly identified as an amniotic invagination, the strong indentation may be a morpho- genetic process equivalent to vestibule formation. Other species with modified development either have or lack an amniotic invagination, consistent with their phylogenetic position as cidaroids or euechinoids (see Parks et ai. 1989), and this sunken condition is, therefore, not simply due to the yolkiness of the larva. A comparison of our figures with those of the cidaroid Phyllacanthus pani- spimis. which lacks an amniotic invagination, shows that the oral surface of A. ijimai is considerably more indented than that of the cidaroid (Parks ci al.. 1989, fig. 3f). Our observations raise the possibility that a partial, possibly
primitive, form of an amniotic invagination may be pres- ent in echinothurioids.
Parks et al. (1989) also examined sectioned material of A. ijimai and concluded that an amniotic invagination was absent. These authors hypothesized that an amniotic invagination arose in the euechinoid lineage after the echinothurioid branch. They hypothesized further that, because the two most primitive lineages of echinoids, the cidaroids and echinothurioids, lacked an amniotic invag- ination, the absence of this character was primitive for echinoids. Our observations suggest that their first phy- logenetic hypothesis may not be accurate, but our findings are consistent with the hypothesis that the amniotic in- vagination is a derived character in euechinoids. Based on comparisons of the fate of larval epidermis among echinoderm classes, Emlet (1988) suggested the same hy- pothesis that the primitive condition for echinoids is the absence of an amniotic invagination.
The formation of epineural sinuses in Asthenosoma iji- mai matches very closely the original descriptions of the same process occurring within the amniotic invagination of other euechinoids [compare Fig. 7f-h with original text- figs, e-h of von Ubisch, 1913 (text-figs, f, g reprinted in Hyman, 1955, p. 497)]. By contrast, epineural sinus for- mation in this echinothurioid differs from that described for the cidaroid, Eucidaris thouarsi (Emlet, 1988). In E. thouarsi, epineural folds were present, but not clearly ev- ident when observed by SEM. Sections of E. thouarsi showed epineural folds closely adhered to the developing juvenile oral surface on the left side of the larva, whereas in euechinoids, the epineural folds were not so closely adhered. Emlet (1988) hypothesized that the pattern in E. thouarsi might reflect a different mechanism of epi- neural fold movement (from that in euechinoids) but might also result from the open condition of the surface upon which this process occurs in E. thouarsi. The largely open nature of the oral surface in A. ijimai, and the dis- tinctly euechinoid appearance of its epineural folds, sug- gest that the open condition of the epineural folds in E. thouarsi was not a cause for their appearance. This ob- servation dismisses Emlet's (1988) hypothesized expla- nation for convergence of the epineural folds ofE. thouarsi and the ophiuroid, Ophiopholis aculeata (Olsen, 1942) but leaves standing the hypothesis that cidaroids and ophiuroids have similar means of epineural sinus for- mation (Emlet, 1988).
Conclusions
In this re-examination of the larval morphogenesis of Asthenosoma ijimai, evidence has been presented to show that A. ijimai has retained previously unrecognized, re- duced pluteus characters. As such, this larval form is the most reduced pluteus yet described, being considerably
ECHINOTHURIOID DEVELOPMENT, REVISITED
29
more modified than larvae of Phyllacanthus imperialis (Olson et ai, 1988) and Peronella japonica. This contri- bution brings to three the number of lineages with mod- ified development and with pluteus characters that are retained to varying degrees. In contrast, four lineages (Phyllacanthus parvispinus. Heliocidaris erythrogramma, a temnopleuroid, and Abalns species) have lost most, if not all, primitive larval characters. (The genus Phyllacan- thus stands alone as being represented in both groups, but it is not known whether non-feeding development has evolved independently for the two species: P. imperialis and P. pun'ispinus.) Comparative experimental and cell lineage information about species with reduced larval fea- tures must now be collected if we are to determine 1 ) whether there is a common, possibly convergent, theme of developmental changes, or 2) whether modifications to cleavage and cell lineage fates are additional changes occurring after the loss of feeding and the reduction of the pluteus form. The detailed description presented here adds to the growing collection of comparative data on modified development; but it also indicates that the basic morphological changes occurring in other species with modified development — including two that are already well studied, Heliocidaris erythrogramma and Peronella japonica — ought to be re-examined.
Acknowledgments
This research was supported by grants from the Japa- nese Society for Promotion of Science (to SA and RBE) and the United States National Science Foundation (BSR- 9058139 to RBE). We would like to thank E. Arakawa for technical assistance, M. McFall-Ngai for allowing us to use her digitizing software, I. Lagomarsino for digitizing so many serial sections, and R. A. Raff for providing the B2C2 antibody. We are also grateful to S. Smiley for advice on clearing opaque larvae. Comments of V. Morris and two anonymous reviewers helped improve the manuscript.
Literature Cited
Amemiya, S. 1980. Metamorphosis of an echinothurioid sea urchin. Aslhenosoma ijimai. Pp. 74-78 in An Outline of Modern Biology Vol. I IB. I. Yasumasu et al. eds. Kakayama-shoten, Tokyo, (in Jap- anese)
Amemiya, S., and T. Tsuchiya. 1979. Development of the echinothurid sea urchin Aslhenosoma iiunai Mar Bioi 52: 93-96.
Armstrong, P. B., and D. Parent!. 1973. Scanning electron microscopy of the chick embryo. De\: Bioi 33: 457-462.
Bisgrove, B. W., and R. A. Raff. 1989. Evolutionary conservation of the larval serotonergic nervous system in a direct developing sea ur- chin. Dev. Growth Differ. 31: 363-370.
Cox, K. H., L. M. Angerer, J. J. Lee, E. H. Davidson, and R. C. Angerer. 1986. Cell lineage-specific programs of expression of multiple actin genes during sea urchin embryogenesis. / Mol. Bioi 188: 159-172.
Davidson, E. H. 1986. (jene Activity in Early Development. 3rd edition. Academic Press. Orlando. FL.
Emlet, R. B. 1982. Echinoderm calcite: a mechanical analysis from
larval spicules. Bioi Bull. 163: 264-275.
Emlet, R. B. 1985. Crystal axes in recent and fossil adult echinoids indicate trophic mode in larval development. Science 230: 937-940. Emlet, R. B. 1988. Larval form and metamorphosis of a "primitive" sea urchin, Eucidaris thonarsi (Echinodermata: Echinoidea: Cida- roida). with implications for developmental and phylogenetic studies. Bioi. Bull 174:4-19.
Emlet, R. B. 1990. World patterns of developmental mode in echinoid echinoderms. Pp. 329-335 in Advances in Invertebrate Reproduction, I'ol. 5. M. Hoshi and O. Yamashita. eds. Elsevier Science Publ.. Amsterdam.
Emlet, R. B., L. R. McEdward, and R. R. Strathmann. 1987. Echino- derm larval ecology viewed from the egg. Pp. 55-136 in Echinoderm Studies. Vol. 2, M. Jangoux and J. M. Lawrence, eds. Balkema Press. Rotterdam.
Fell, F. J. 1976. The Cidaroida (Echinodermata: Echinoidea) of Ant- arctica and the southern oceans. Unpublished Ph.D. dissertation. University of Maine, Orono, Maine. 293 pp. Gordon, I. 1926a. The development of the calcareous test of Echinus
milians. Philos. Trans. R Six.: B 214: 259-312. Gordon, I. 1926b. The development of the calcareous test of Ecliino-
cardium cordatum Philos. Trans. R. Sac. B 215: 255-313. Harris, P., and G. Shavt. 1984. Intermediate filaments, microtubules and microfilaments in epidermis of sea urchin tubefoot. Cell Tissue Res. 236: 27-33.
Henry, J. J., and R. A. Raff. 1990. Evolutionary change in the process of dorsoventral axis determination in the direct developing sea urchin. Heliocidaris erythrogramma. Dev Bioi 141: 55-69. Horstadius, S. 1939. The mechanics of sea urchin development, studied
by operative methods. Bioi. Rev Camb. Philos. Soc. 14: 132-179. 1 1\ m:in. L. H. 1955. The Invertebrates: Echinodermata. I'ol. II'
McGraw-Hill, New York. Larrain, A. 1973. Los pedicelarios globiferos y el desarrollo de Abatus
agassi:i. Bol. Soc. Bioi Concepcion 46: 53-63. Mortensen, T. 1921. Studies of the Development and Larval Forms of
Echinoderms. G.E.C. Gad. Copenhagen. Mortensen, T. 1936. Echinoidea and Ophiuroidea. Discover}' Rep 12:
199-348.
Okazaki, K. 1975. Normal development to metamorphosis. Pp. 177- 232 in The Sea Urchin Embryo Biochemistry and Morphogenesis, G. Czihak, ed. Springer-Verlag. Berlin.
Okazaki, K.. and K. Dan. 1954. The metamorphosis of partial larvae of Peronella japonica Mortensen. a sand dollar. Bioi Bull. 106: 83- 99.
Olsen, H. 1942. Development of a brittlestar Ophiopholis aculeata. with a short report on the outer hyaline layer. Bergens Museum Aar- bok. .\atur. 6: 1-107.
Olson, R. R., J. L. Cameron, and C. M. Young. 1988. Larval devel- opment of the pencil urchin Phyllacanthus imperialis: a lecithotrophic pluteus. P. 807 in Echinoderm Biology. Proc 6th Int'l Echinoderm Conl. Victoria. R. D. Burke. P. V. Mladenov, P. Lambert. R. L. Parsley, eds. Balkema. Rotterdam.
Parks, A. L., B. A. Parr, J.-E. Chin, D. S. Leaf, and R. A. Raff. 1988. Molecular analysis of heterochronic changes in the evolution of direct developing sea urchins. / Evol. Bioi. 1: 27-44.
Parks, A. L., B. \V. Bisgrove, G. A. Wray, and R. A. Raff. 1989. Direct development in the sea urchin Phyllacanthus parvispinus (Cidaroidea): phylogenetic history and functional modification. Bioi. Bull 177: 96-109.
Raff, R. A. 1987. Constraint, flexibility, and phylogenetic history' in the evolution of direct development in sea urchins. Dev Bioi 119: 6-19.
30
S. AMEM1YA AND R. B. EMLET
Raff, R. A., B. Parr, A. Parks, and G. Wray. 1990. Radical evolutionary
change in early development. Pp. 71-98 in Evolutionary Innovations, M. H. Nitecki, ed. University of Chicago Press, Chicago, IL.
Richardson, K. C, L. Jarrett, and E. H. Finke. 1960. Embedding in epoxy resin for ultrathin sectioning in electron microscopy. Slain Technol. 35:313-323.
Schatt, Ph. 1985. Developpement et croissance embryonnaire de 1'oursin incubant Abatus cordatus (Echinoidea: Spatangoida). These de Doctoral de 1'Universite Pierre et Marie Curie. 151 pp.
Schatt, Ph. 1988. Embryonic growth of the brooding sea urchin Ahatux cordatus. Pp. 225-228 in Echinodcnn Biology. Proc. Mh Inl '/ Echi- noderm Con/.. Victoria, R. D. Burke, P. V. Mladenov, P. Lambert, and R. L. Parsley, eds. Balkema. Rotterdam.
Schroeder, T. E. 1981. Development of a "primitive" sea urchin (Eu- cidaris tribuloides): irregularities in the hyaline layer, micromeres, and primary mesenchyme. Biol liuli 161: 141-151.
Smith, A. B. 1984. Echinoid Paleobiology. George Allen and Unwin, London.
Strathmann, R. R. 1978. The evolution and loss feeding larval stages
of marine invertebrates. Evolution 32: 894-906. Ubisch, L. von. 1913. Die entwicklung von Slrongylocentrotus livuhis.
(Echinus microtuberculatus, Arbacia pustulosa). Z. Wiss. Zoo/. 106:
409-448. \\ illiams, D. H. C., and D. T. Anderson. 1975. The reproductive system,
embryonic development, larval development and metamorphosis of
the sea urchin Heliucidaris erythragramma (Val.) (Echinoidea: Echi-
nometndae). Aust. J Zoo/. 23: 371-403. Wray, G. A., and R. A. Raff. 1989. Evolutionary modification of cell
lineage in the direct-developing sea urchin He/iociiiaris erythro-
gramma. Dev Biol. 132: 458-470. Wray, G. A., and R. A. Raff. 1990. Novel origins of lineage founder
cells in the direct-developing sea urchin Heliocidaris erythrogramma.
Dev. Biol. 141:41-54. Yasumasu, I., K. Mitsunaga, and V. Fujino. 1985. Mechanism forelec-
trosilent Ca:+ transport to cause calcification of spicules in sea urchin
embryos. Exp. Cell Res 159: 80-90.
Reference: Biol. Bull 182: 31-40. (February, 1992)
Purification and Biochemical Characterization
of the Nuclear Sperm-Specific Proteins of
the Bivalve Mollusks Agriodesma saxicola1
and Mytilimeria nuttalli
JUAN AUSIO
Department of Biochemistry and Microbiology, University of Victoria, I 'ictoria. British Columbia I '8 W 3P6. Canada
Abstract. The proteins from the nuclei of the sperm from two different species of the subclass Anomalodes- mata of the class Bivalvia have been analyzed for the first time. In both instances — Agriodesma saxicola (Baird, 1863) and Mytilimeria nuttalli (Conrad, 1837)— the compositional pattern is very similar. The sperm chro- matin is organized by a major protamine-like PL-I protein. As in all PL-I, this protein has a trypsin-resistant core. In both species analyzed, PL-I contains cysteine residues that account for the presence of the monomer (M) and dimer (D) forms observed in the total nuclear HC1 extracts. The molecular mass of these proteins is 21,000 Da in A. sax- icola, and 25,000 Da in M. nuttalli. All of the specimens of A. saxicola analyzed were hermaphrodites. As a result, the nuclear sperm-specific proteins from several preparations were readily and extensively degraded by protease activity from the oocytes. Such degradation was always observed when cross contamination between the two gonadal tissues accidentally occurred during protein extraction.
Introduction
The presence of a highly specialized histone HI (PL-I protein) seems to be a common feature of the nuclear protein composition of the sperm of bivalve mollusks (Jutglar et al.. 1991). Despite the structural heterogeneity of the sperm proteins within this taxonomic group (Ausio, 1 986; Zalensky and Zalenskaya, 1980), a PL-I protein has been identified in each of the species analyzed in detail so far. Like histone
Received 18 June 1 99 1 ; accepted 25 November 1991. ' This species is more commonly referred to as Entodesma saxicola, but see Bernard. 1983.
HI, this protein is soluble in diluted perchloric acid, has a globular trypsin-resistant core, is lysine rich, and yet is com- positionally related to the protamines (PL = protamine-like) (Subirana et al., 1973). Of all the different nuclear sperm- specific proteins found within a given species, PL-I is the one with the lowest electrophoretic mobility in urea-acetic acid gels (Ausio, 1986). The presence of a protamine-like histone HI -like protein in bivalve mollusks may have im- portant evolutionary implications, not only within the phy- lum Mollusca (Subirana and Colom, 1987), but also within other taxonomic groups.
As pointed out by Kasinsky (1989), however, only some of the subclasses within the class Bivalvia have been an- alyzed so far. and only a few species have been thoroughly characterized. Nevertheless, at least one sperm-specific histone HI (PL-I) protein has been identified: in Mytilus californianus (by Jutglar et al.. 1991), Crassostrea gigas (by Sellos, 1985), and Glycymeris yesonensis (by Odin- tsova et al, 1989) (subclass Pteriomorphia); in Ensis minor (by Giancotti et al., 1983), Spisula solidissima (by Ausio etal., 1987) and Macoma nasuta (Ausio, 1988) (Subclass Heterodonta); and in Anodonta piscinallis (by Rozov et al., 1984) (Subclass Palaeoheterodonta). In some of these species, two histone HI -like proteins have been described.
Although the subclass Heterodonta has been widely studied (Ausio, 1986), three subclasses, according to Barnes's (1980) classification of the bivalve mollusks, have never been characterized: Cryptodonta, Palaeotaxodonta, and Anomalodesmata. In the present work, we have an- alyzed and characterized the sperm-specific proteins of two species within the subclass Anomalodesmata and have shown that each contains a highly specialized histone H 1 -
31
32
J. AUSIO
like (PL-I) protein that is the major protein component of the nuclei of the sperm of that organism.
Materials and Methods
Living specimens
Specimens of Agriodesrna saxicola and Mytilimeria nuttalli were collected along the west coast of Vancouver Island. British Columbia, Canada, by SCUBA divers from the Biology Department at the University of Victoria.
Nuclei preparation and protein extraction
Isolation of sperm nuclei and HC1 crude extraction of the nuclear basic proteins was performed as described elsewhere (Ausio, 1986). Briefly, after carefully opening the shell, a small incision was made in the gonadal tissue, and the spontaneously released sperm were resuspended in NaCl 0.15 M, Tris-HCl 20 mM pH 7.6, 0.2 mM PMSF (Phenylmethylsulphonyl Fluoride) (buffer A). Because,-!. saxicola is hermaphroditic, its sperm would sometimes be contaminated with oocytes released accidentally from the intimately associated ovaries (see below for discussion).
The sperm suspension was centrifuged at 3000 X g for 10 min in a SS-34 Sorvall rotor at 4°C. The pellet obtained was homogenized in buffer A containing 0.5% TritonX- 100. After standing for 10 min on ice, the suspension was spun down under the same conditions as before. This step is meant to solubilize most of the cytoplasmic membranes, including the sperm flagella and the acrosome. Notice that this step will also expose the sperm nuclei to egg lysates in those samples of A. saxicola contaminated with oocytes; such cytoplasmic contamination may be respon- sible for the protein degradation observed under these circumstances. The detergent-treated pellet was imme- diately homogenized in 0.4 N HC1. Solubilization was continued for 2 h under stirring at 4°C. Finally, the sus- pension was centrifuged at 12,000 X g for 10 min at 4°C, and the acid extract was precipitated with 6 volumes of acetone, overnight, at -20°C.
Gel electrophoresis
Polyacrylamide gel electrophoresis was carried out on urea-acetic acid gels, as described elsewhere (Ausio, 1986).
Protein purification and fractionation
Ionic exchange chromatography was carried out on a 10 X 100 mm Protein-Pak SP 8HR column from Waters- Millipore as described elsewhere (Mogensen el a/.. 199 1 ).
Gel nitration was carried out on a 10 X 300 mm FPLC Superose 12HR 10/30 column from Pharmacia. The elu-
tion buffer was 6 M guanidinium chloride (Gdn-HCl; Schwarz/Mann Biotech), 50 mM Tris-HCl pH 7.6.
Reverse phase high pressure liquid chromatographv (HPCL)
HPLC was carried out on a 5 n (25 X 0.46 cm) Vydac C4 column, with 0. 1% trifluoroacetic acid (TFA) as eluant with different acetonitrile gradients.
Determination of the molecular mass
The molecular mass of each protein was determined by gel nitration under denaturing conditions on a Superose 12HR column in the presence of 6 M Gdn-HCl (see above). Several protamines, protamine-like proteins, and histones of known molecular mass were used as standards: PL-I from Spisula solidissima (Mr: 33,500 Da) (Ausio and Subirana, 1982a); Histone HI from calf thymus (Mr: 22,000 Da) (DeLange, 1976); PL-Ill (01) from Mytilus edulis (Mr: 9600 Da) (Ausio and Subirana, 1982b); and unfractionated salmine from Oncorhynchus sp. (Mr; 4300 Da) (Ando et al, 1973). Histone HI from calf thymus was purchased from Worthington, and salmine (sulfate form) was obtained from Sigma. The rest of the prot- amines were prepared in my laboratory. Globular proteins were also used as a molecular mass markers: bovine serum albumin (Mr: 68,000); ovoalbumin (Mr: 46,000); chy- motrypsinogen A (Mr: 25,000); and ribonuclease A (Mr: 13,200). These proteins were purchased from Pharmacia; all of them were subjected to performic acid oxidation before being applied to the column (see below). Vitamin B12 (Mr: 1350) was purchased from Sigma. For the es- timation of the molecular mass, the column was calibrated with the above standard proteins, and a plot of Kav versus log Mr was constructed (Mr = molecular mass; Kav = distribution coefficient).
v, - v0
where vo and V, are the void and total volume of the column, and Ve = the elution volume of a given protein. Blue Dextran and dansyl-L-alanine (Sigma) were used to de- termine V0 and V, experimentally. The proteins of un- known Mr were mixed with the protein standards and run together through the column. Their molecular masses were estimated by interpolation of their Kav values on the best fitting line of the calibration plot.
Amino acid analysis
Amino acid analysis was carried out on an Applied Biosystems model 420A derivatizer-analyzer system. The hydrolysis was carried out in gas-phase 6 N HC1 and 1% phenol under an argon atmosphere at 165°C, for 1 h, 2
NUCLEAR PROTEINS FROM THE SPERM OF BIVALVE MOLLUSKS
33
h, and 4 h, the final amino acid composition was obtained by extrapolation of the data to zero time. So that cysteine could be quantified, all protein samples were pyridyl- ethylated before hydrolysis, as described below.
Chemical modification of proteins
Reduction of SH groups. The SH groups of cysteine were reduced as described by Kuehl (1979). Briefly, the proteins, at 1 mg/ml in 6 M urea 20 mAf Tris-HCl pH 7.6, were reduced in the presence of 8% |8-mercaptoeth- anol for 3 h at room temperature.
Oxidation ofSH groups. Oxidation was carried out un- der the same buffer conditions as above, but in the pres- ence of 0.72 mM O-phenanthroline and 0.36 mA/CuSO4.
Performic acid oxydation. Performic acid was prepared according to Hirs (1967). For the oxidation, 1-mg aliquots of protein were dissolved in 0.5 ml of performic acid, which had been previously cooled on ice. The reaction was allowed to proceed for 4 h in an ice bath in capped tubes. The sample was then resuspended in a 25-fold ex- cess of HPLC grade distilled water and lyophilized.
Cysteine pyridylethylation. Proteins were pyridyleth- ylated, providing for a quantitative estimate of cysteine in the amino acid analysis. The procedure used was as fol- lows: proteins (=1 nanomol) were dissolved in 44 ^1 of 6.8 Murea, 60 mM Tris-HCl, 1.25 mA/EDTA (pH 7.6), and 2.3% /3-mercaptoethanol. The solution was incubated for 3 h at room temperature in the dark. Subsequently, 8
CE AS
HIST
0 I II
X T J-
IV SA
J U
PL- 1
PHI
PL- 1
PL-IV
PR
Figure 1 . Urea acetic acid PAGE analysis of the nuclear sperm-specific proteins of Agriodesma saxicola (AS) and Mytilimeria nuttalli (MN) in comparison to a histone standard from chicken erythrocytes (CE) and to a protamine from salmon, salmine (SA). The nuclear sperm-specific proteins of one representative of each of the five groups (O, I, II, III. and IV) of the classification of the bivalve mollusks (Ausio. 1986) are also shown. The representative species chosen for each group were: O: Pecten maximus; I: Spisula solidissima; II: Ensis ensis; III: Mamma nasula. and IV: Mytilus edulis. The regions corresponding to the different protamine-like (PL-I, PL-II, PL-Ill, and PL-IV) proteins defined in Ausio (1986) are also shown. HIST: histone region. PR: protamine. D: dimer form. M: monomer form of the major sperm protein component in each species. X2, Y2: possible dimer forms of the minor sperm protein components X, Y.
34
J. AUSIO
A.
0.8
0.6
0.2
0.0
ss a b c d
8
16
32 40
TIME , min
64
72
80
2.0 1.5 -
1.0 -
O
CO
0.5 2
0.0
B.
1.5
1.0
o
fO CM
0.5
0.0
SS
L/L/UUUUUUU
10
20
30 40
TIME , min
50
60
60 50
UJ
405! 300
r- LU
20 <
10 0
Figure 2. Fractionation of a crude 0.4 A' HCl extract from the nuclei of the sperm of Agnodesma saxicola. (A) Ionic exchange chromatography on a (10 X 100 mm) Protein-Pak SP 8HR column. Proteins were eluted with a linear (0-2 A/) NaCl gradient in 50 m.1/ Na-phosphate buffer (pH 6.8) at a flow rate of 1 ml/min. The inset shows the urea-acetic acid PAGE analysis of the fractions indicated. (B) Reverse-phase HPLC on a Vydac C4 column. Elution was carried with an acetonitrile gradient in 0. 1% trifluoroacetic acid at a flow rate of 1 ml/min. The inset shows the electrophoretic analysis of the fractionation. The lanes shown in the inset, and the chromatogram has been aligned to match the fraction analyzed with its corresponding position in the chromatogram. SS: starting sample.
NUCLEAR PROTEINS FROM THE SPERM OF BIVALVE MOLLUSK.S
35
A.
1.2
0.8
0.4
0.2
0.0
IL
B.
mn a b c d e f 9
_
I
?v5SHHBI^HBM^H
as a b c d e f 9
_
16
24 32
TIME ,min.
48
Figure 3. (A) Gel filtration FPLC on a Superose 12 HR 10/30 column. The elution buffer was 6 M Gdn-HCI in 50 mA/ Tris-HCl pH 7.6. The flow rate was 0.4 ml/min. The elution profiles of HC1 nuclear
extracts from the sperm of Mytilimeria nuttalli ( ) and Agriodesma saxicola ( ) are shown together
with the elution profile (••••) of some of the standards used to calibrate this column: I: PL-I from Spisula solidissima; II: Histone HI from calf thymus; III: PL-HI from Mytilus edulis; IV: protamine salmine; V: vitamin B 12; VI: dansyl-L-alanine. (B) Electrophoretic analysis on urea-acetic acid gels of the fractions a, b. c. d. e, f, g of the elution profiles of At. nuttalli and A. saxicola. mn: starting sample of A/ nuttalli. as: starting sample of A. saxicola.
H\ of 4-vinylpyridine was added, and the reaction was al- lowed to proceed for 2 h at room temperature. The sample was then immediately desalted in an HPLC reverse phase C8Vydac column, which was eluted for 5 min with 0.1% TFA (trifluoroacetic acid), and for 20 min with a 0-70% acetonitrile gradient in 0.1%- TFA. /3-lactoglobulin from Applied Biosystems Inc. was used as a standard for this procedure.
Trypsin digestion. Trypsin digestion of proteins in high salt— 2 M NaCl, 50 mA/ Na-phosphate buffer (pH 6.8)— was carried out as described elsewhere (Ausio el ai, \ 987).
Results
Chromatographic analysis and purification of the sperm-specific nuclear proteins from A. saxicola and M. nuttalli
Figure 1 shows the 0.4 N HC1 protein extracts from the nuclei of the sperm of A. saxicola and Al. nuttalli. They are shown in comparison to the five groups previously
established for the classification of the nuclear sperm-spe- cific proteins of the bivalve mollusks (Ausio, 1986). In each of the two species analyzed, two major protein bands run in the region of the PL-I proteins (Ausio, 1986). In addition to these proteins, 10-20% of minor protein frac- tions X and Y, which run in the histone region, are also observed (see Fig. 1-AS and 1-MN). This protein com- position was sufficiently novel that the two organisms could not, at first, be assigned to any of the protein groups previously established in my classification of the bivalves (Ausio, 1986). That was not surprising because they belong to a subclass (Anomalodesmata) that had not been ana- lyzed before. I therefore decided to purify and characterize each of the major protein components of these organisms. The first attempt at fractionation by ionic exchange FPLC under non-denaturing conditions is shown in Figure 2 A. Most of the protein components coeluted in a single multiphasic peak at around 2 M NaCl, but some protein separation was clearly achieved as is shown in the inset of the same figure.
36
J. AUSIO
1.0
0.8
0.6
0.4
0.2
0.0
103
10" Mr
105
Figure 4. Calibration plot used to determine the molecular mass of the sperm proteins determined on a superose 12HR 10/30 column under the elution conditions described in Figure 3A. Globular O and nonglob- ular® proteins were used as standards. 1: Vitamin B 12 (Mr: 1350 Da); 2: ribonuclease A (Mr: 13.200 Da); 3: chymotrypsingen A (Mr: 25,000 Da): 4: ovoalbumin (Mr: 46,000 Da): 5: bovine serum albumin (Mr: 68,000 Da): 1*: protamine salmine (Mr: 4,300 Da); 2*: protein PL-I from Myl ilus editlis (Mr: 9,600 Da); 3*: histone HI from calf thymus (Mr: 22,000 Da); 4*: Protein PL-I from Spisula solidissima (Mr: 33,000 Da). M = monomer form and D = dimer form of the major sperm protein components of: Mytillineria mtnalli •, and Agriodesma saxicola O.
Characterization of the sperm-specific nuclear proteins from A. saxicola and M. nuttalli
The fractionation problems described in the preceding section began to be elucidated when the molecular mass of these proteins was analyzed. Figure 4 shows the cali- bration plot used to estimate the molecular mass of pro- teins from gel filtration analysis. The molecular mass of the two major protein components of the sperm nuclei in A. saxicola were: 21,000 Da for the fastest protein component and 43,000 Da for the slowest moving frac- tion. The values were 25,000 Da and 49,000 Da for M. nuttalli. These results suggested a monomer-dimer rela- tionship between the slow and the fast moving protein fractions present in each species. To analyze this rela- tionship, and to examine the nature of the association phenomenon, 1 incubated the crude HC1 extracts in the presence of either /3-mercaptoethanol or copper phen- anthroline. Figure 5 shows the results of these treatments in the case of A. saxicola, and identical results were ob- tained with M. nuttalli (results not shown). The slower moving band completely disappears under reducing con- ditions for cysteine. Under oxidizing conditions (in the presence of copper phenanthroline) the relative intensity of the faster moving band (see Fig. 5b) slightly decreases, and higher association complexes are formed (see arrow in Fig. 5b). We are therefore dealing with the association of the faster moving protein components. Although the association seems to involve primarily the formation of dimers, the number of cysteines present in the monomer
To increase the resolution in the separation, the 0.4 N HC1 protein extracts were fractionated by reverse-phase HPLC; the elution profile is shown in Figure 2B. Although two peaks corresponding to each of the two major com- ponents could be clearly separated, both of them exhibited different amounts of what appeared to be overlapping cross-contamination.
Size fractionation of the starting HC1 extracts by gel filtration under denaturing conditions in the presence of 6 A/guanidinium chloride (Gdn-HCl) is shown in Figure 3. Although the peaks could not be completely resolved, the sample was partially fractionated as shown in Figure 3B. Indeed, when some of the eluting fractions from the different regions corresponding to the two major protein components were pooled together and rerun under the same conditions, two distinct peaks could then be clearly resolved. This was used as a basis for estimating the molecular mass of each of these protein components. Nevertheless, when the fraction under each separate peak was analyzed by urea acetic acid PAGE, the same cross-contamination observed in Figure 2B was again observed, although to a lesser extent (results not shown).
PL-I
SS A B
. • . U
^^M
d •— •
x2
m — »»
Figure 5. Urea-acetic acid polyacrylamide gel electrophoresis of the nuclear sperm specific proteins from Agriodesma saxicola under A: re- ducing(6% /3-mercaptoethanol) or B: oxidizing(copper-phenanthroline) conditions. SS = starting sample, m, d = monomer and dimer of major sperm protein component (PL-I). X, X2 = monomer and dimer of the minor sperm protein component. The arrowhead indicates the presence of higher association oligomers obtained upon oxidative treatment.
NUCLEAR PROTEINS FROM THE SPERM OF BIVALVE MOLLUSKS
37
cannot be clearly ascertained from the above experiments. Thus, although the strong tendency toward dimer for- mation would suggest the presence only of one cysteine per molecule, the lack of complete dimerization observed in Figure 5b, and the presence of association complexes higher than dimers. would strongly suggest the presence of more than one cysteine. The presence of two cysteine residues per molecule, which could easily form an internal disulfide bond, would explain the incomplete polymer- ization of the monomer, otherwise expected under the oxidizing conditions used here (Fig. 5b).
To determine the number of cysteines, as well as to establish the amino acid composition of the major nuclear protein component of the sperm of A. saxicola and M. mittalli, protein fractions such as those shown in the insets of Figure 2A and B were pyridylethylated before amino acid analysis. A /5-lactoglobulin sample was simulta- neously treated and analyzed to check for the completion of the reaction. The amino acid analyses clearly show (see Table I) that the proteins of both A. saxicola and M. nut- talli contain two cysteine residues per molecule. Com- parison with the amino acid analyses of other PL proteins, reveals the PL-I nature of the major nuclear protein com- ponent of the sperm of the two species analyzed. Like other PL-I proteins (Ausio, 1986; Ausio, 1988; Jutglar et al., 1991), these have an internal trypsin resistant core (Fig. 6).
Besides the major protein components M and D, we have also characterized the minor component X of A. saxicola. This protein exhibits an amino acid composition that is almost identical to PL-I (see Table I). Although
Table I
Amino acid analysis (mol %) of the nuclear sperm-specific PL-I proteins o/'Agriodesma saxicola PL-I (AS) and Mytilimeria nuttalli PL-I (MN) in comparison to those of Spisula solidissima PL-I (SS) (Ausio and Subirana. I982a) and Macoma nasuta PL-I (MC) (Ausio. 1988)
Pl-I(AS) PL-I(MN) PL-I(SS) PL-I(MC) X (AS)
|
Lys |
18.7 |
16.3 |
24.8 |
21.8 |
15.1 |
|
His |
— |
0.4 |
— |
2.3 |
0.4 |
|
Arg |
34.8 |
33.8 |
23.1 |
26.9 |
34.7 |
|
Asx |
1.5 |
1.8 |
0.6 |
0.8 |
2.5 |
|
Thr |
2.0 |
1.0 |
4.3 |
4.0 |
1.6 |
|
Ser |
20.8 |
26.5 |
21.7 |
20.2 |
21.4 |
|
Glx |
1.1 |
0.8 |
0.6 |
0.8 |
4.6 |
|
Pro |
0.5 |
0.8 |
2.4 |
1.8 |
1.7 |
|
Gly |
6.0 |
5.5 |
3.0 |
2 2 |
5.0 |
|
Ala |
3.3 |
4.2 |
14.2 |
11.3 |
4.4 |
|
1/2 Cys |
0.9 |
0.6 |
— |
0.7 |
tr.* |
|
Val |
3.2 |
2.4 |
2.3 |
2.4 |
2.8 |
|
Mel |
— |
0.2 |
0.4 |
0.2 |
— |
|
lie |
1.2 |
0.9 |
0.5 |
1.4 |
1.0 |
|
Leu |
3.6 |
3.3 |
1.7 |
2.1 |
2.6 |
|
Tyr |
1.5 |
1.0 |
0.3 |
0.5 |
1.0 |
|
Phe |
0.7 |
0.7 |
0.3 |
0.7 |
1.0 |
|
Trp |
— |
— |
0.3 |
— |
AS'OM OD
2o
3o 4M 4o SM 5o
* Determination carried out in the absence of pyridilethylation treat- ment.
The amino acid analysis of the minor protein component X of A. saxicola (AS) is also shown.
the amino acid analysis was carried out without prior pyr- idilethylation, trace amounts of cysteine could still be de- tected. Because of its relative electrophoretic mobility, X2 (see Fig. 1-AS) most probably represents the dimer form of X. Indeed, X2 disappears upon /3-mercaptoethanol treatment of the starting protein sample (see Fig. 5).
i
Figure 6. Analysis of the time course of digestion by trypsin of the monomer (M) and dimer (D) of the PL-I protein ofAgriodesma saxicola. Digestions were carried out in 2 A/ NaCl, 50 mA/ Na phosphate (pH 6.8), at an enzyme:substrate ratio of 1:500. The digestion times were: 0: 0 min; 1: 5 min; 2: 15 min; 3: 30 min; 4: 60 min and 5: 120 min. AS*: nuclear sperm-specific proteins of A. saxicola slightly degraded by an egg protease (see legend to Fig. 7). r: peptide resistant to digestion by egg proteases.
Specific degradation oj PL-I in A. saxicola
Every specimen of .-1. saxicola analyzed was hermaph- roditic. Although the male and female gonads are com- pletely separated, some contamination of the sperm by oocytes sometimes occurred when accidental incisions were made in the ovary as the shells were being opened. The protein composition of the crude HC1 nuclear extracts thus obtained showed a complex and highly variable pat- tern in urea acetic acid gel electrophoresis. Figure 7 shows a light microscope and electrophoretic analysis of several sperm samples with different amounts of contamination by oocytes. In preparations containing pure sperm, the electrophoretic analysis showed two major bands, M and D (Fig. 7a). corresponding to the monomer and dimer of PL-I, as well as a 15-20% of X and X2. As the extent of contamination by female germinal cells increases (see Fig. 7b, c), the amounts of M and D present in the HC1 extracts
38
D
Figure 7. Microscopic analysis with phase contrast of pure sperm (A), and of sperm preparations containing an increasing amount of contamination by eggs (B) and (C). The samples were obtained from different specimens of Agrtiidesma saxicola. a, b, and c: electrophoretic analysis, in urea-acetic acid PAGE, of sperm preparations containing an increasing amount of contamination by eggs. As contamination increases, an extensive degradation of both the monomer (M) and dimer (D) forms of the major nuclear sperm-specific PL-I component is observed. A relatively resistant peptide — r — is generated during this degradation process. The white bar corresponds to 50 pm. X and X2 are as in Figure 5.
decrease, and the proteins finally disappear completely. This is accompanied by the appearance of a complex pat- tern of new bands with faster electrophoretic mobility (Fig. 7b, c). Such protein pattern transition is clearly indicative of a degradation process elicited by specific proteases from the contaminating eggs. A similar in vitro degradation of sperm histones by the cytoplasm of sea urchin eggs has also been reported (Betzalel and Moav, 1987). A quite resistant degradation peptide. designated r, is produced during this process. Although the composition and nature of peptide r are completely unknown, it is certainly much smaller than the trypsin-resistant peptide obtained under in vitro conditions (see Fig. 6). A nuclear HC1 protein extract from pure oocytes contained none of the proteins observed in Figure 7 (results not shown).
Discussion
In this work I have analyzed the protein composition of the nuclei of the sperm of two representatives of the subclass Anomalodesmata within the class Bivalvia. In both of the species analyzed — Agriodesma saxicola and Mytilimeria nuttalli — 80-90% of the nuclear sperm-spe- cific proteins consist of a mixture of dimer (D) and mono- mer (M) forms of a protamine-like (PL-I) protein. The remaining 10-20% includes the minor protein fractions X and Y.
The protamine-like nature of the major proteins is re- vealed by their amino acid composition (see Table I). They
clearly fulfill the compositional definition of protamines (Subirana, 1983): (Lys + Arg) = 45-80%, (Ser + Thr) = 10-25%. Indeed, of all the PL proteins characterized so far, the ones analyzed here exhibit the highest arginine content within the PL classification (Ausio, 1986). The presence of a trypsin-resistant core (see Fig. 6) indicates that these proteins are also related to the proteins of the histone H 1 family. Therefore, the PL major components of both A. saxicola and M. nuttalli should be undoubtedly classified within the PL-I class (Ausio, 1986). The presence of cysteine in these proteins does not seem to be an un- usual feature; indeed, two cysteines also occur in the PL- I component ofMacoma nasuta (Ausio, 1988).
Because the minor protein component X of A. saxicola has an amino acid composition indistinguishable from PL-I, these two proteins must be closely related, and X: (see Fig. 1-AS) may represent a dimer form of X. The same observations apply to the X and Y components of M. nuttalli (results not shown). The structural relation- ships among PL-I and the X and Y fractions remain ob- scure, but the similarity of their amino acid analyses to that of the major protein component suggests that X and Y may be proteolytic peptides from PL-I. They could arise from the activity of either a nuclear or an acrosomal sperm protease (Miiller-Esterl and Fritz, 1981) during protein extraction. However, they do not seem to be related to any of the protein fragments produced by the protease digestion resulting from oocytic contamination, because
NUCLEAR PROTEINS FROM THE SPERM OF BIVALVE MOLLUSKS
39
Table II
Classification of the class Bivalvia according to llieir protamine-like group (Ansid. 1986)
|
Subclass (a) |
Representative species |
Protein type (b) |
Reference |
|
Cryptodonta |
|
|
|
|
Palaeotaxodonta |
|||
|
Palaeoheterodon ta |
Anodonta pisciniallis |
[(?) |
(c) |
|
Heterodonta |
Spisu/a solidissima |
I |
(d) |
|
Ensis minor |
I! |
(e) |
|
|
Macoma masuta |
Ill |
(f) |
|
|
.\fytilus cilulis |
IV |
(g) |
|
|
Pteriomorphia |
Crassostrea gigas |
0 |
(h) |
|
Anomalodesmata |
Agriodesma saxicola |
II |
(i) |
|
Myli/imcria nuttalli |
I |
(i) |
(a) According to Barnes (1980).
(b) According to Ausio (1986).
(c) Rozovrt al. (1984).
(d) Ausio and Subirana ( 1982a).
(e) Giancotti et al. (1983). (I) Ausio (1988).
(g) Ausio and Subirana (1982c). (h) Sellos(1985). (i) This work.
the presence of X2 or X does not increase as the level of egg-induced degradation increases (see Fig. 7). Indeed, when whole sperm cells (without any prior preparation of the nuclei) were extracted with HC1 for '/2 h at 4°C immediately after sperm collection, the overall protein pattern observed was undistinguishable from the pattern of the HC1 extracts prepared from nuclei uncontaminated by oocytes. In particular, the X-band was still observed.
The presence of a major PL-I protein in the sperm of the two species analyzed here would allow us to classify them within the protamine-like group I of my earlier clas- sification of the bivalve mollusks (Ausio, 1986: see Table II). Species fulfilling this compositional pattern have also been described in other subclasses, including Palaeohet- erodonta and Heterodonta. The presence of a PL-I pro- tein, however, seems to be a common feature to all the species of the class Bivalvia (Jutglar et al., 1991).
All of the PL-I proteins that have been analyzed in detail have structures related to the histone H 1 superfam- ily (Ausio e/ a/.. 1987; Ausio, 1988; Jutglar et al., 1991). The structural similarities of PL-I to both histone H 1 and the arginine-rich protamines from the vertebrates suggests a close evolutionary relationship between these proteins. In this sense, the increase in arginine and the decrease in lysine and alanine observed in the case of the PL-I proteins analyzed in the present work, when compared to other PL-I proteins, would indicate a further departure from their H 1 nature and a closer relationship to the protamines from vertebrates.
Acknowledgments
I am very indebted to Debra Murie, Daryl Parkyn, and Joachim Schnorr von Carosfeld from the Biology De- partment at the University of Victoria for providing me with the biological specimens used in this work. I am also very grateful to Steve Carlos for his valuable assistance in running the HPLC and FPLC columns and for reading the manuscript. I also would like to thank Mrs. Denise Lunger and Ms. Cheryl Gonnason for typing the manu- script. This work was supported by NSERC Grant OGP 0046399 to Juan Ausio.
Literature Cited
Ando, T., M. Yamasaki, and K. Suzuki. 1973. P. 28 in Prolamines. Vol. 12 of Molecular Biology, Biochemistry and Biophysics. Springer- Verlag. Berlin. Heidelberg. NY.
Ausio, J., and J. A. Subirana. 1982a. A high molecular weight nuclear basic protein from the bivalve molluscs. J. Biol. Chem. 257: 2802- 2805.
Ausio, J., and J. A. Subirana. 1982b. Conformational study and de- termination of the molecular weight of highly charged basic proteins by sedimentation equilibrium and gel electrophoresis. Biochemistry 21: 5910-5918.
Ausio, J., and J. A. Subirana. 1982c. Nuclear proteins and the orga- nization of chromatin in spermatozoa of Mytilus edulis. Exp. Cell. Res. 141: 39-45.
Ausio, J. 1986. Structural variability and compositional homology of the protamine-like components of the sperm from the bivalve mol- luscs. Comp. Biochem. Physiol. 85B: 439-449.
Ausio, J., A. Toumadje, R. McParland, R. R. Becker, W. C. Johnson, Jr., and K. E. van Holde. 1987. Structural characterization of the trypsin resistant core in the nuclear sperm-specific protein from Spi- sula solidissima. Biochemistry 26: 975-982.
Ausio, J. 1988. An unusual cysteine-containing histone HI -like protein and two protamine-like proteins are the major nuclear proteins of the sperm of the bivalve mollusc: Macoma nasuta. J. Biol. Chem. 263: 10.141-10,150.
Barnes, R. D. 1980. Invertebrate Zoology. 4th ed. Saunders College, Philadelphia.
Bernard, F. R. 1983. Catalogue of the living Bivalvia of the eastern Pacific Ocean: Bering Strait to Cape Horn. Canadian Special Pub- lication of Fisheries and Aquatic Sciences 81: 83.
Betzalel, M., and B. Moav. 1987. Degradation of sperm histones in vitro by cytoplasm of the sea urchin egg. Cell Dtjier. 20: 125- 136.
Del.ange. 1976. Handbook of Biochemistry and Molecular Biology. 3rd ed. (Proteins), G. Fasman. ed. CRC press Boca Raton, FL, vol II, 294 pp.
Giancotti, V., E. Russo, M. Casparini, D. Serrano, D. Del Piero, A. W. Thorne, P. D. Car), and C. Crane-Robinson. 1983. Proteins from the sperm of the bivalve mollusc: Ensis minor. Eur. J. Biochem 136: 509-516.
Hirs, C. M. \V. 1967. Determination of cysteine as cysteic acid. Me/hods En:ymol. XI: 59-62.
Jutglar, L., J. I. Borell, and J. Ausio. 1991. Primary, secondary and tertiary structure of the core of a histone HI -like protein from the sperm of Mylilus. J Biol. Chem. 266: 8184-8191.
Kasinsky, H. E. 1989. Specifity and distribution of Sperm Basic Pro- teins. Pp. 73-163 in Histones and Other Basic Nuclear Proteins. Vol. I, L. S. Hnilica, G. S. Stein, and J. L. Stein, eds. CRC Press, Boca Raton. FL.
40
J. AUSIO
Km hi. J. 1979. Synthesis of high mobility group proteins in regenerating rat liver. / Biol. Chem. 254: 7276-7281.
Mogensen, C., S. Carlos, and J. Ausio. 1991 . Microheterogeneity and interspecific variability of the nuclear sperm proteins from Mytilus. FEES Lett. 282: 273-276.
Miiller-Esterl, W., and H. Fritz. 1981. Sperm acrosin. Methods En- :ymol. 80: 621-632.
Odintsova, N. A., S. M. Rozov, and I. A. Zalenskaya. 1989. The chro- mosomal proteins from the sperm of the bivalve molluscs Swiftopecten swift v and Glycymerys yesonensis. Comp. Biochem. Physiol. 93B: 163-167.
Rozov, S. M., V. A. Brednikov, F. K. Corel, M. V. Lavrenteva, and L. P. Solonenko. 1984. Structure of lysine-nch histone from sperm of Anodonta piscinalis. Mol. Biol. (transl) (USSR) 18: 1497-1508.
Sellos, D. 1985. The histones isolated from the sperm of the oyster Crassostrea gigas. Cell Differ. 17: 183-192.
Subirana, J. A., C. Cozcolluela, J. Palau, and M. Unzeta. 1973. Pro- tamines and other basic proteins from spermatozoa of molluscs. Biochim. Biophys. Ada 317: 364-379.
Subirana, J. A. 1983. Nuclear Proteins in Spermatozoa and Their In- teractions with DNA. Pp. 197-214 in The Sperm Cell, J. Andre, ed. Martinus Nijhoff B. V., The Hague.
Subirana, J. A., and J. Colom. 1987. Comparison of protamines from freshwater and marine bivalve molluscs: evolutionary implications. FEBS Lett. 220: 193-196.
Zalensky, A. O., and I. A. Zalenskaya. 1980. Basic chromosomal pro- teins of marine invertebrates II. The proteins from the Bivalvia mol- luscs. Comp. Biochem. Physiol. 66B: 415-419.
Reference: Biol Bull. 182: 41-53. (February, 1992)
The Origin of Cortical Vesicles and their Role in Egg
Envelope Formation in the "Spiny" Eggs of a
Calanoid Copepod, Centropages velificatus
PAMELA I. BLADES-ECKELBARGER' AND NANCY H. MARCUS2
1 Darling Marine Center, University of Maine, Walpole, Maine 04573 and Department of Oceanography, Florida State University, Tallahassee, Florida 32306
Abstract. The mature oocytes of the marine calanoid copepod, Centropages velificatus, contain two morpho- logically distinct populations of cortical vesicles that un- dergo sequential exocytoses at the time of spawning. The contents of the primary cortical vesicles are released first and form the primary egg envelope. This is followed by the exocytosis of the secondary cortical vesicles. These contain numerous intracisternal granules that, upon re- lease into the perivitelline space, transform into a mass of fine fibers. The continual accumulation of fibers con- stitutes an extracellular matrix between the primary en- velope and the egg's plasmalemma. Further amassment of the fibers beneath the primary egg envelope results in the formation of long, spiny projections. The evolution of the cortical vesicles was traced to the early vitellogenic oocytes and appears to be unique. The two populations of cortical vesicles are synthesized together within the same cisternal elements of rough endoplasmic reticulum (RER). The RER originates from membranous blebs off both the nuclear membrane and stacks of annulate lamellae in the early vitellogenic oocytes. Numerous intracisternal gran- ules are present within the RER. Some of these granules fuse, forming a dense, ring-like structure in the extremities of the cisternae. These bud off from the RER to become the primary cortical vesicles. The unfused intracisternal granules remain as discrete bodies within irregular profiles
Received 15 July 1 99 1 ; accepted 1 October 1991.
Contribution Nos.: (PBE) Harbor Branch Oceanographic Institution Contribution No. 892 and Darling Marine Center, University ofMaine Contribution No. 244; (NHM) Florida State University Marine Labo- ratory Contribution No. 1068.
of vesicular ER and comprise the secondary cortical vesicles.
Introduction
The subject of post-embryonic development in free- living copepods has been a favorite research topic for de- cades. Consequently, the literature abounds with descrip- tions of naupliar and copepodid developmental stages. However, studies relating to embyronic development, i.e., those stages from spawning to the emergence of the first nauplius, are limited to only a few early publications (Grobben, 1881; Fuchs, 1914; Witschi, 1934; Marshall and Orr, 1954, 1955). In particular, details of the mech- anisms of fertilization and egg envelope formation have yet to be elucidated in the Copepoda.
Some marine calanoid species spawn their eggs into ovisacs that remain attached to the female until the emer- gence of the first or second naupliar stage. The majority of marine calanoids, however, are broadcast spawners, releasing the eggs freely into the surrounding water where they undergo development. Within this latter group, eggs of a variety of shapes and sizes and with different types of surface ornamentation have been observed (Johnson, 1967; Koga, 1968; Kasaharac/a/.. 1974; Uye, 1983; Mar- cus, 1990). The egg surfaces of most species are smooth, but others may be adorned with flanges or spines of vary- ing shapes and lengths. The production of spiny eggs has been reported for numerous species, Acartia toma (Zil- lioux, 1969), Centropages ponticus (Sazhina, 1968), C. hamatus (Pertzova, 1974), Pontella mediterranea (Sa- zhina, 1 968; Grice and Gibson, 1 98 1 ; Santella and lonora, 1990), A. erythraea, C yamadai, C. abdominalis (Kasa-
41
42
P. I. BLADES-ECK.ELBARGER AND N. H. MARCUS
hara et al, 1974), A. steuri (Uye, 1983), C. velificatus (Marcus, pers. obs.), Calanus glacialis (J. Runge and Blades-Eckelbarger, pers. obs.), and Candacia pachydac- tyla (Blades-Eckelbarger, pers. obs.).
Some of the species listed above produce two morpho- logical types of eggs where the spiny form represents a diapause stage (hatching is delayed), and the smooth form typifies a subitaneous stage (no mandatory delay in hatching) e.g., Centropages hamatus (Pertzova, 1974) and C. ponticits (Sazhina, 1968). Ponlella mediterranea pro- duces three morphotypes; diapause eggs with long spines, and subitaneous eggs that are either smooth or adorned with short spines (Sazhina, 1 968; Grice and Gibson, 1981; Santella and lonora, 1990). Acartiatonsa(ZH\iou\, 1969) and A. steuri (Uye, 1983) have been reported to produce both smooth and spiny eggs, but their physiological clas- sification as diapause or subitaneous is still in question. For the remaining species, only spiny eggs have been ob- served, and there is no evidence to suggest that they are a diapause stage.
While conducting a morphological survey of copepod eggs found in sea bottom muds, we became intrigued with the spiny modifications of the egg envelopes of some spe- cies. Consequently, we initiated a study using light mi- croscopy along with scanning and transmission electron microscopy to investigate the stages of egg envelope de- velopment and spine formation in the eggs of Centropages velificatus.
Materials and Methods
Adult female Centropages velificatus (De Oliveira, 1947) were sorted from plankton tows collected approx- imately 10 miles due east off the coast of Fort Pierce, Florida. Female's carrying mature, pigmented oocytes were placed in small dishes with filtered seawater that were observed every few minutes for spawned eggs. The eggs were carefully picked up by drawn-out pipettes and placed onto pieces of 35 ^m mesh supported by Beem capsules (Flood, 1973). The Beem capsules sat in shallow glass dishes containing 2.5% glutaraldehyde in filtered seawater.
For transmission and scanning electron microscopy (TEM and SEM), eggs in varying stages of development, from polar body extrusion to advanced spine formation, were collected and fixed in this manner. After approxi- mately 100 eggs were placed in a Beem capsule, the capsule was transferred to a 5% Karnovsky's (1965) glutaralde- hyde-paraformaldehyde mixture in 0.1 A/ Sorensens phosphate buffer. The capsules were flushed several times with the latter fixative to prevent precipitate caused by seawater mixing with the phosphate buffer. As a matter of convenience, due to the long duration of the complete fixation process, the eggs were held in the Karnovsky's
fixative for varying times depending on the time of day collected. Those collected in the morning were held at room temperature for 3 to 6 h. Those collected in the evening were held overnight at 4°C. The lower temper- ature slows the fixation process. There were no apparent differences in cell or organelle structure among the varying times and temperatures.
Adult females carrying mature oocytes were prepared also for TEM. Initially each individual was placed in a small amount of the Karnovsky's glutaraldehyde mixture for approximately 1 5 min. The head and urosome were then removed with a sharp razor and the metasomes transferred to a vial containing fresh fixative and held for the same range of times as the eggs.
This primary fixation of both eggs and adult females was followed by 2 or 3 rinses in 0. 1 Al Sorensen's phos- phate buffer (pH 7.4) and then held in 2% OsO4 in 0. 1 M Sorensen's buffer at room temperature for 1-2.5 h. The samples were rinsed briefly with buffer and dehydrated through an ascending series of alcohols to 70%. At this point, some of the eggs were pipetted onto SEM stubs covered with double-sided sticky tape and allowed to air dry in a desiccator for 2 to 3 days. The air-dried stubs were coated with gold palladium and observed with a Zeiss Novascan 30 SEM.
For TEM examinations, the remaining eggs and female metasomes were dehydrated further to 100% ETOH fol- lowed by propylene oxide and infiltrated with three changes of Epon (Luft, 1961). For final embedding, the female metasomes were oriented in flat embedding molds. The eggs were carefully drawn into a wide bore pipette with fresh Epon and dropped into a Beem capsule, which was centrifuged at room temperature for 20 min at setting #6 in a clinical centrifuge. Because the Epon is of a slightly thickened consistency, centrifugation is needed to con- centrate the hardened eggs into the tip of the Beem cap- sule. Extensive sectioning of eggs prepared in this manner revealed no membrane or organelle damage. For light microscopy, l-/um thick sections were cut with glass knives on a Porter-Blum MT2B ultramicrotome, and stained with Richardson's stain (Richardson et al, 1960). Thin sections for TEM were stained with uranyl acetate followed by lead citrate and examined on a Zeiss EM9-S2 TEM.
It should be noted that the procedures for both SEM and TEM result in minor shrinkage of the eggs. Therefore, all measurements are approximations.
Results Live observations o/ spawning and spine formation
Females were observed spawning on several occasions, during which they remained active, swimming in a normal manner around the dish. The oocytes flowed out of one
COPEPOD EGG ENVELOPE FORMATION
43
Figures 1-5. SEMsof egg envelope and spine formation from emergence of first polar body (Fig. 1, unlabeled arrow) to 24-h-old embryo (Fig. 5). Figure 6. SEM. High magnification of spines.
^m
• '"• • -'
t * x&:;3*R3
Figure 7. Perinuclear region of vitellogenic oocyte showing nuclear bleb (large arrowheads) extending from nuclear envelope (Nm) in formation of rough endoplasmic reticulum (RER). Note intracisternal granules (g) within nuclear bleb and RER. NP, nuclear pores.
Figure 8. Perinuclear region of vitellogenic oocyte showing annulate lamellae (AL). Note swollen extremities (RER) containing granules (g). Nm. nuclear membrane; Nu, nucleus.
44
COPEPOD EGG ENVELOPE FORMATION
45
or both oviducts, emerging from the genital pore as a single mass. The female would periodically twitch the urosome, causing the amorphous mass of eggs to break free and fall to the bottom of the dish. Approximately 5-10 s after re- lease from the female, the eggs separated from each other and transformed from an oval to a spherical shape. Release of the first polar body occurred at this time (Figs. 1, 14). The second polar body was not observed. The actual pro- cess of sperm and egg fusion in copepods has never been reported, nor was it seen in the present study. Therefore, it could not be ascertained when egg envelope formation began relative to the moment of fertilization.
Figures 1 to 6 present comparative SEM views of the stages of spine formation from emergence of the first polar body (Fig. 1 ) to a 24-h-old embryo (Fig. 5). Approximately 5 min after spawning, large, rounded bumps appeared on the egg surface (Fig. 2). These bulges became more slender and pointed, forming short jagged spines (Fig. 3). It took approximately 15-20 min for long spines to form. A sur- vey of over 100 eggs that were at least 24 h old revealed individual variations in the morphology, number, and size of the spines.
Cortical vesicle formation in vitellogenic oocytes
Formation of the egg envelope in Centropages velifi- catus involves the exocytosis of two morphologically dis- tinct, membrane-bound inclusions present in the egg's cytoplasm. Prior to spawning, the mature oocytes that reside in the oviducts of the female contain a variety of morphologically distinct granules, vesicles, and inclusions. One type of inclusion, referred to here as the primary cortical vesicle, appears as a membrane-bound body con- taining an electron-dense, granular material that sur- rounds an electron-lucent core (Figs. 11. 12). Favorable sections through the center of these vesicles present the appearance of a darkly staining ring around a flocculent center (Fig. 12). The secondary cortical vesicles appear as irregularly shaped vesicles filled with several moderately dense granules (ca. 75-82 nm diameter) (Figs. 11, 12).
Primary and secondary cortical vesicles originate in the very early stages of vitellogenesis, where a blebbing process of the outer lamina of the nuclear membrane is observed (Fig. 7). These nuclear blebs contain numerous moderately dense granules (ca. 80 nm diameter) and pinch off to form lamellar and vesicular profiles of rough endoplasmic retic- ulum (RER). Stacks of annulate lamellae are also observed in the perinuclear region (Fig. 8), as well as in the central
cytoplasmic region of mid- and late- vitellogenic stages (Fig. 9). Vesicles containing several dense granules, morpholog- ically identical to the nuclear blebs, also pinch off from the extremities of the annulate lamellae. Fusion of some, but not all, of these intracisternal granules culminates in the formation of the ring-shaped densities that characterize the primary cortical vesicles (Figs. 8-10).
The cytoplasm of mid- vitellogenic oocytes is filled with elongate profiles of RER containing numerous unfused, intracisternal granules residing with one or more ring- shaped densities (Figs. 9, 10). Small Golgi complexes were observed infrequently, but did not appear to contribute to the contents of the RER. In the mature oocytes, the ring-shaped portions bud off from the RER to become the primary cortical vesicles (Figs. 10, 11). They are en- closed by a smooth membrane devoid of ribosomes. The unfused intracisternal granules remain as discrete bodies within irregular profiles of vesicular ER that also have lost the attached ribosomes. These represent the secondary cortical vesicles (Figs. 11, 12).
There is no elaboration of an egg envelope prior to spawning. The oocytes are enclosed by a simple oolemma that is coated with a lightly staining glycocalyx (Fig. 13). The glycocalyx, or vitelline envelope, is deposited over the oolemma by the associated follicle cells during the mid-stages of vitellogenesis ( Blades- Eckelbarger and Youngbluth, 1984).
Cortical reaction, egg envelope elaboration, and spine formation
Deposition of the egg envelope results from a cortical vesicle reaction involving two successive stages of exo- cytosis. Soon after spawning, the majority of yolk bodies and other inclusions accumulate toward the center of the egg, but the primary and secondary cortical vesicles re- main in the cortical cytoplasm (Fig. 14). The first cortical reaction is characterized by the exocytosis of the primary cortical vesicles. The bounding membrane of the primary cortical vesicles fuses with the egg's plasmalemma, and the enclosed material is released into the perivitelline space (Figs. 15, 16). This results in the formation of a narrow layer (ca. 20 nm thick) of darkly staining material situated slightly above the egg's plasmalemma (Figs. 15, 16, 18, 2 1 ). We refer to this first layer as the "primary egg en- velope." At this time, the egg surface has a "bumpy" ap- pearance (Figs. 2. 3, 17) where regions of the plasmalemma
Figure 9. Early stage of primary cortical vesicle formation (large arrowheads) in vesicular RER of mid-vitellogenic oocyte. Note annulate lamellae (AL) with swollen extremities (RER). M. mitochondrion; Y. yolk granules.
Figure 10. Mid-vitellogenic oocyte. Ring-shaped densities (large arrowheads) budding off of RER (*) in formation of primary' cortical vesicles. M, mitochondrion; Y. yolk granule.
Figure 11. Late vitellogenic oocyle with primary cortical vesicles (Pv) now separate from secondary cortical vesicles (SV). M, mitochondrion. Figure 12. High magnification showing structure of primary (Pv) and secondary (Sv) cortical vesicles.
Figure 13. Oolemma (Oi) of mature oocyte in oviduct of female covered by vitelline envelope (*). Fc. follicle cell; Oo, ooplasm. Figure 14. Light micrograph, 1-^m-thick section of newly spawned egg and formation of first polar body (arrowhead). Note centrally located yolk granules with primary and secondary cortical vesicles occupying cortical cytoplasm.
46
>Ffy;
•v
I ^ . * • •» - -V
'16 -
,- ..- "'
;
•
...
OPT--
E/T7*- ' "">- ii
" • I
19
- -7/ I m
•-* ~» '""' k """
**••
.. •
Figures 15 and 16. First cortical reaction. High magnification of egg surface showing fusion of primary cortical vesicles (Pv) with oolemma (Ol) and exocytosis of dense material in formation of primary envelope (Pe).
Figure 17. Light micrograph. 1-fjm-thick section of egg during first cortical reaction and exocytosis of primary cortical vesicles. Note that cytoplasm extends into projections of egg surface.
Figure 18. Cortical cytoplasm of egg in late stage of first cortical reaction showing fusion of granules within secondary cortical vesicles (Sv). M, mitochondrion; Ol. oolemma; Pe. primary envelope; Y, yolk granule.
Figures 19 and 20. High magnification of secondary cortical vesicles showing fusion of granules.
47
48
P. I. BLADES-ECKELBARGER AND N. H. MARCUS
Pe
3frr-
-
H4*©l ** -^Sg"
:*^f£&fir ^Jf^LJL^^M *Z
" r -• ^^ir\w-^'<;
i
:.:.iTit"E*ii«?f?f^'^.'. / A'^ -^^>"'-^^r,'^/SL>.-..V.*fe'f.*---''*'i'. . ___________
Figure 21. Second cortical reaction showing exocytosis of granules (large arrowheads) from secondary cortical vesicles (Sv). Ol, oolemma; Pe, primary envelope.
COPEPOD EGG ENVELOPE FORMATION
49
bulge out. The egg's cytoplasm projects into these ex- panded areas (Fig. 17).
A second wave of exocytosis follows soon after the first with the release of the intracisternal granules contained within the secondary cortical vesicles (Figs. 21, 23). Just prior to their release, however, some of the intracisternal granules fuse with each other, forming slightly larger and denser masses (Figs. 18-20). Once in the perivitelline space, the intracisternal granules transform into a mesh- work of fibers that adhere to the inner surface of the pri- mary egg envelope (Figs. 22, 23, 25, 27, 28). Concomitant with the second wave of exocytosis is the appearance of numerous endocytotic pits and vesicles along the egg's plasmalemma (Figs. 22, 23).
With the continual accumulation of fibers from the secondary cortical vesicles, the primary envelope lifts higher above the egg's plasmalemma forming an irregular surface sculpturing (Fig. 22), the plasmalemma withdraws from the core of the spines and the egg proper becomes spherical again (Fig. 26). Observations of eggs in multi- cellular stages (approx. 24 h), during or just after synthesis of the naupliar cuticle, revealed both long and short spines with a crenulated surface (Figs. 27, 28). The space between the cuticle and the egg envelope is composed of a thick mass of fibers (Figs. 27, 28).
Discussion
Based on our observations of the oocytes and eggs of Centropages veliftcatus, we present here the first identi- fication of cortical vesicles, and a description of the cortical reaction and subsequent egg envelope formation in the Copepoda. These processes follow the same general se- quence of post-spawning events as reported in other an- imal species (Schuel, 1985; Longo, 1988). However, where the eggs of some species contain a single, morphologically heterogeneous population of cortical vesicles (or granules), those of C. velificatits were found to have two. Conse- quently, the cortical reaction in the eggs of C. veliftcatus, involves not one, but two exocytotic events.
The presence of structurally different populations of cortical granules has been demonstrated in other crusta- ceans, the crab Carcinus maemis (Goudeau and Becker, 1982) and the decapod shrimp Sicyonia ingentis (Pillai and Clark, 1988, 1990). Talbot and Goudeau (1988) re- ported four distinct cortical vesicles in the oocytes of the
lobster Homarus. In all cases, the various populations of cortical vesicles exhibited distinctly different morpholo- gies, underwent temporally separated exocytoses, and in S1. ingentis (Pillai and Clark, 1990) were found to be chemically heterogeneous. Each type of cortical vesicle contributed to different layers of the egg envelope.
During the first cortical reaction in the eggs of Centro- pages veliftcatus. the contents of the primary cortical ves- icles form the outer, or primary, egg envelope. This layer may correspond to the fertilization envelope of other an- imals, which is formed from the mixing of the vitelline layer with the exocytosed contents of the cortical vesicles (Kay and Shapiro, 1985; Somers and Shapiro. 1989). Exocytosis of the secondary cortical vesicles in the eggs of C. veliftcatus follows soon after the primary egg enve- lope is complete. The secondary vesicles contain several discrete intracisternal granules that, upon their release into the perivitelline space, transform into a myriad of fibers. The accumulation of these fibers between the egg's plas- malemma and the primary egg envelope forms an extra- cellular matrix (ECM) that exhibits a similar morphology to ECMs surrounding the eggs and embryos of other ma- rine invertebrates (Spiegel el ai, 1989).
The present paper further illustrates the cellular mech- anisms by which the two populations of cortical vesicles are synthesized in the vitellogenic oocytes of Centropages veliftcatus. In the oocytes of many animals, the cortical vesicles are derived from the Golgi complex (see Schuel, 1985, and references therein). In the decapod shrimp, Si- cyonia ingentis (Pillai and Clark, 1988), one population of cortical vesicles is derived from Golgi complexes and the second population from within the cisternae of RER. Cortical vesicle formation in C. veliftcatus, in general, is similar to that of Carcinus (Goudeau, 1 984) and Homarus (Kessel, 1968: Talbot and Goudeau, 1988) in which the vesicles are produced by the ER, and Golgi complexes do not appear to contribute. Other aspects of cortical vesicle formation in C. velificatus are unique; ( 1 ) both nuclear blebs and annulate lamellae appear to be involved in for- mation of the vesicular RER that synthesizes the intra- cisternal granules and. (2) these intracisternal granules appear to be the precursors of both the primary and sec- ondary cortical vesicles. Fusion of some of these granules within the RER cisternae forms the dense, ring-shaped contents of the primary cortical vesicles. The other intra- cisternal granules do not fuse, but remain distinct and
Figure 22. Second cortical reaction showing lifting of primary envelope (Pe) and filling of perivitelline space with fine fibers (*). Note surface sculpturing of primary envelope as well as coated micropinocytotic pits (p) and vesicles (v) along the oolemma (Ol). Sg. granules from secondary cortical vesicles.
Figure 23. Early spine formation showing massive exocytosis of secondary cortical vesicle (Sv). Large arrowhead denotes transformation of granular material into fine fibers. P, coated micropinocytotic pits; Sg. granules from secondary cortical vesicles in perivitelline space.
Figure 24. Mid-stage of spine formation.
Figure 25. High magnification of penvitelline space in 24-h-old embryo showing transformation of granules (small arrowheads) from secondary cortical vesicles into line fibers (*). Cu, early cuticle of nauplius.
Figure 26. Light micrograph of live 24-h-old embryo with advanced spine formation. Note that cytoplasm has receded from spines (small arrowheads).
Figures 27 and 28. Advanced spine formation of 24-h-old embryo (Em) showing thick mass of fibers filling spines (*). Cu. cuticle of nauplius.
50
COPEPOD EGG ENVELOPE FORMATION
51
comprise the secondary cortical vesicles. The primary vesicles separate from the secondary vesicles in the later stages of vitellogenesis.
In the eggs ofCarcinus maenus (Goudeau and Lachaise. 1980a. b), the cortical vesicles of one type are filled with "ring-shaped" granules that are the precursor of the main layer of the embryonic capsule. The authors emphasized that these ring-shaped granules are homologous to the "disc-shaped granules" or "intracisternal granules" pre- viously considered as endogenous yolk in the vitellogenic oocytes of several decapod crustaceans ( Beams and Kessel, 1962, 1963: Kessel, 1968; Ganion and Kessel, 1972). Subsequent studies have confirmed that, instead of pos- sessing nutritive qualities, the ring-shaped granules in these crustacean eggs play a structural role in formation of the egg envelope (Goudeau and Becker, 1982; Goudeau, 1984; Talbot and Goudeau, 1988; Pillai and Clark, 1990).
Within the Calanoida, the secondary cortical vesicles of Centropages velificatus appear homologous to the "in- tracisternal granules representing the endogenous yolk" in the oocytes of Centropages typicus (Arnaud ct a/.. 1982) and to the "granular form of type 1 yolk" in the oocytes of Labidocera aestiva (Blades-Eckelbarger and Young- bluth, 1984) and Ponlella mediterranea (Santella and la- nora, 1990). Our present observations parallel those of Goudeau and Lachaise (1980a, b) and Talbot and Gou- deau (1988), illustrating that the intracisternal granules previously assumed to represent endogenous yolk in co- pepod eggs, are actually precursors of the egg envelope. The distinctive morphology of the primary cortical vesicles in C. velificatus, however, has no correlate in the eggs of other copepod species studied thus far, even in the oocytes of a congeneric species, C. typicus (Arnaud el a/.. 1982). The fact that the eggs produced by C. typicus do not elab- orate spines warrants a closer look at morphological dif- ferences between the eggs of these congeners.
One consequence of the two exocytotic episodes in the eggs of Centropages velificatus is the addition of large quantities of membrane to the egg's plasmalemma. This occurs when the limiting membrane of the cortical vesicles fuses with the plasmalemma of the egg. However, the di- ameter of the egg does not increase significantly. The presence of numerous endocytotic pits and vesicles ob- served along the egg's plasmalemma during the second exocytotic event provides a mechanism for the recycling of at least some of the extra surface membrane. This pro- cess has been illustrated in the eggs of other animals (see review by Longo, 1988) and conforms with similar ob- servations on mammalian secretory tissues (Mata and Christensen. 1990).
Earlier studies have described two membranes sur- rounding the copepod egg (see reviews by Davis, 1968, 1981 ). During hatching, the outer membrane cracks and the inner membrane pushes out. The outer membrane
slips off and the nauplius is enclosed within the more delicate inner membrane. The nauplius then breaks open this membrane with its appendages and swims free. The ultrastructural features of the primary egg envelope in Centropages velificatus do not exhibit a trilamellar com- position indicative of a true membrane. Therefore, we suggest that this layer should be referred to as the hatching envelope, such as described for other crustaceans (Gou- deau and Becker, 1982; Pillai and Clark. 1988). The pres- ence and structure of an inner egg membrane around the copepod nauplius has yet to be validated because we did not examine the later embryonic stages.
The morphology of the subitaneous egg envelope of Centropages velificatus is very different from that of the envelope encasing diapause eggs as reported for Hemi- diaptomus ingens privinciae (Champeau, 1970), Diapto- mus sangiiineus (Hairston and Olds, 1984), Pontella mediterranea (Santella and lanora, 1990), and Anomal- ocera patersoni (lanora and Santella, 1991). The thick, multi-layered envelope surrounding diapause eggs appears as a lamellar arrangement of microfibrils in a helicoidal array and is considered comparable to the typical integ- ument of arthropods (Hairston and Olds, 1984).
The function of the spines that characterize the eggs of Centropages velificatus and those of other copepod species remains elusive. During a discussion session of the sym- posium entitled "Cultivation of Marine Invertebrates" held in Princeton in 1967, it was suggested that spines on copepod eggs might retard sinking (Allen, 1969), enhance gas exchange, and afford protection from predation (Shel- bourne, 1969). However, while it seems reasonable that the spines would deter predators, Zillioux ( 1 969) reported that spiny eggs were consumed by adult female Acartia. More recently, Santella and lanora (1990) suggested that the four-layered egg envelope and accompanying spines present on the diapause eggs of Pontella mediterranea may supply nutritive material during diapause and provide ex- tra protection from harsh environmental conditions.
Numerous functions have been proposed for the spines that cover the surface of other marine invertebrate eggs. The eggs of some sea urchins, starfish, and sea anemones present a spiny appearance due to the elongation and bundling of cortical microvilli (Schroeder, 1982, 1986). It is suggested that these microvillous "spikes" and "spires" play a role in reinforcing the egg surface (Schroe- der, 1982), function to facilitate absorption, and aid in adhesion between dividing cells of the embryo (Schroeder, 1986). Copepod oocytes are not known to possess mi- crovillar modifications of the oolemma (Blades-Eckel- barger and Youngbluth, 1984), nor do the eggs form mi- crovilli after fertilization (present study). Furthermore, the spines of Centropages velificatus are not cytoplasmic, but are projections of the outer or primary egg envelope caused by the amassment of an extracellular matrix (ECM) within
52
P. I. BLADES-ECKELBARGER AND N. H. MARCUS
the perivitelline space. Recent studies on the ECMs sur- rounding the embryos of other marine invertebrates may hint to the role of the ECM that coats the eggs of C. ve- lificatus. ECMs are believed to provide support and pro- tection to the developing cells, aid in cell movement and cell adhesion, and form a semi-permeable filter for uptake and concentration of substances from the environment needed for growth and differentiation (Spiegel el a!., 1 989).
Literature Cited
Allen, J. A. 1969. Marine Biology Vol. V. Proceedings of the Fifth Interdisciplinary Conference on Marine Biology, J. D. Costlow Jr., ed. Gordon and Breach, Science Publishers, New York, London, and Paris. 606 pp.
Arnaud, J., M. Brunei, and J. Mazza. 1982. Etude de 1'ovogenese chez Centropages typicus (Copepoda, Calanoida). Reprod. Nutr. Dev. 22: 537-555.
Beams, J. \V., and R. Kessel. 1962. Intracisternal granules of the en- doplasmic reticulum in the crayfish oocyte. J. Cell Biol. 13: 158- 162.
Beams, J. W., and R. Kessel. 1963. Electron microscope studies on developing crayfish oocytes with special reference to the origin of yolk. J. Cell Biol. 18: 621-649.
Blades-Eckelbarger, P. I., and M. J. Youngbluth. 1984. The ultra- structure of oogenesis and yolk formation in Labidocera aesliva (Co- pepoda: Calanoida). / Morphol. 179: 33-46.
Champeau, A. 1970. Etude de la vie latent chez les Calanoides (Co- pepodes). Caracteristiques des eaux temporaires de Basse-Provence. Ann. Fac. Sci. Marseille 44: 155-189.
Davis, C. C. 1968. Mechanisms of hatching in aquatic invertebrate eggs. Oceanogr. Mar. Biol. Ann. Rev 6: 325-376.
Davis, C. C. 1981. Mechanisms of hatching in aquatic invertebrate eggs. II. Oceanogr. Mar Biol. Ann. Rev 19: 95-123.
De Oliveira, L. P. H. 1947. Estudos sobre o Microplancton capturado durante a viagem do navio hidrografico Lahmeyer nas baias de Ilha Grande e Sepetiba. Mem. Inst. Oswaldo Cm: 44: 441-488.
Flood, P. R. 1973. A simple technique for the prevention of loss or damage to planktonic specimens during preparation for transmission and scanning electron microscopy. Sarsia 54: 67-74.
Fuchs, K. 1914. Die KeimblatterentwickJung von Cyclops viridis. Zoo/. Jalirh. Anal. 38: 103-56.
Ganion, L. R., and R. G. Kessel. 1972. Intracellular synthesis, transport, and packaging of proteinaceous yolk in oocytes of Orconcctes im- munis. J Cell Biol. 52: 420-437.
Goudeau, M. 1984. Fertilization in a crab: III. Cytodifferentiation of vesicles enclosing ring-shaped elements involved in the cortical re- action. Gamete Res. 9: 409-424.
Goudeau, M., and J. Becker. 1982. Fertilization in a crab. II. Cytological aspects of the cortical reaction and fertilization envelope elaboration. Tissue and Cell 14: 273-282.
Goudeau, M., and F. Lachaise. 1980a. Fine structure and secretion of the capsule enclosing the embryo in a crab (Carcinus maenus (L.)). Tissue and Cell 12: 287-308.
Goudeau, M., and F. Lachaise. 1980b. "Endogenous yolk" as the pre- cursor of a possible fertilization envelope in a crab (Carcinus maenas). Tissue Cell 12:503-512.
Grice, G. D., and V. R. Gibson. 1981. Hatching of eggs of Pontella medilerranea Claus (Copepoda: Calanoida). I'ie et Milieu 31: 49- 51.
Grobben, C. 1881. Die Entwicklungsgeschichte von Cetochilus sep- temtrionalis Goodsir. Arb. Zoo/. Inst. Univ. H'ien 3: 1-40.
Hairston, N. G., and E. J. Olds. 1984. Population differences in the timing of diapause: adaptation in a spatially heterogeneous environ- ment. Oecologia 61: 42-48.
lanora. A., and L. Santella. 1991. Diapause embryos in the neustonic copepod Anomalocera patersoni. Mar. Biol. 108: 387-394.
Johnson, M. W. 1967. Some observations on the hatching of Tortanus discaudatus eggs subjected to low temperatures. Limnol. Oceanogr. 12:405-410.
Karnovsky, M. J. 1965. A formaldehyde-glutaraldehye fixative of high osmolarity for use in electron microscopy. /. Cell Biol 27: 137A.
Kasahara, S., S. Uye, and T. Onbe. 1974. Calanoid copepod eggs in sea-bottom muds. Mar. Biol 26: 167-171.
Kay, E. S., and B. M. Shapiro. 1985. The formation of the fertil- ization membrane of the sea urchin egg. Pp. 45-80 in Biology of Fertilization. A. Monroy and C. B. Metz. eds. Academic Press, Inc., New York.
Kessel, R. G. 1968. Mechanisms of protein yolk synthesis and depo- sition in crustacean oocytes. Z Zellforsch. 89: 1 7-38.
Koga, G. 1968. On the pelagic eggs of Copepoda. J. Oceanogr. Soc. Jpn. 24: 16-20.
Longo, F. J. 1988. Reorganization of the egg surface at fertilization. Int. Rev Cylol. 113: 233-269.
Luft, J. H. 1961. Improvements in epoxy resin embedding methods. / Biophys. Biochem. Cytol. 9: 409-414.
Marcus, N. H. 1990. Calanoid copepod, cladoceran, and rotifer eggs in sea-bottom sediments of northern California coastal waters: identification, occurrence, and hatching. Mar. Biol. 105: 413- 418.
Marshall, S. M., and A. P. Orr. 1954. Hatching in Calanus fmmar- chicus and some other copepods. J. Mar. Biol. Assoc. U.K. 33: 393- 401.
Marshall, S. M., and A. P. Orr. 1955. The biology of a marine copepod, Calanus finmarchicus (Gunnerus). Oliver & Boyd, Edinburgh and London.
Mata, R. L., and E. I. Christensen. 1990. Redistribution and recycling of internalized membrane in seminal vesicle secretory cells. Biol. Cell. 68: 183-193.
Pertzova, N. M. 1974. Life cycle and ecology of a thermophilous co- pepod, Centropages hamalus in the White Sea. Z Zhurnal 53: 101 3- 1022.
Pillai, M. C., and W. H. Clark Jr. 1988. Hatching envelope formation in shrimp (Sicyonia ingentis) ova: origin and sequential exocylosis of cortical vesicles. Tissue Cell 20: 941-952.
Pillai, M. C., and W. H. Clark Jr. 1990. Development of cortical ves- icles in Sicyonia ingentis: their heterogeneity and role in elaboration of the hatching envelope. Mol. Reprod. Dev 26: 78-89.
Richardson, K. C., L. Jarett, and E. H. Finke. 1960. Embedding in epoxy resins for ultrathin sectioning in electron microscopy. Stain Technol. 35: 313-323.
Santella, L., and A. lanora. 1990. Subitaneous and diapause eggs in Mediterranean populations of Pomella medilerranea (Copepoda: Calanoida): a morphological study. Mar. Biol 105: 83-90.
Sazhina, L. I. 1968. On hibernating eggs of marine Calanoida. Zoo/. Zhurnal 47: 1554-1556.
Schroeder, T. E. 1982. Novel surface specialization on a sea anemone egg: "spires" of actin-filled microvilli. J. Morphol. 174: 207-216.
Schroeder, T. E. 1986. The egg cortex in early development of sea urchins and starfish. Pp. 59-100 in Developmental Biology. L. W. Browder, ed. Plenum Publishing Corp.
Schuel, H. 1985. Functions of egg cortical granules. Pp. 1-43 in Biology of Fertilization. A. Monroy and C. B. Metz, eds. Academic Press, Inc.. New York.
COPEPOD EGG ENVELOPE FORMATION
53
Shelbourne, J. C. 1969. In Marine Biology Vol. V Proceedings of the
Fifth Interdisciplinary Conference on Marine Biology. J. D. Costlow
Jr., ed. Gordon and Breach, Science Publishers, New York, London,
and Paris. 606 pp. Somers, C. E., and B. M. Shapiro. 1989. Insights into the molecular
mechanisms involved in sea urchin fertilization envelope assembly.
Dev. Growth Differ. 31: 1-7. Spiegel, E., L. Howard, and M. Spiegel. 1989. Extracellular matrix
of sea urchin and other marine invertebrates. J. Morphol. 199:
71-92. Talbot, P., and M. Goudeau. 1988. A complex cortical reaction leads
to formation of the fertilization envelope in the lobster, Homarus.
Gamete Res. 19: 1-18. Uye, S. 1983. Seasonal cycle in abundance of resting eggs of Acartia
steuri Smirnov (Copepoda, Calanoida) in sea-bottom mud of Ona-
gawa Bay, Japan. Criistaceana 44: 103-105. Witschi, E. 1934. On determinative cleavage and yolk formation in
the harpactid copepod Tisbe jurcata (Baird). Bio/. Bull 68: 335-340. Zillioux, E. J. 1969. In Marine Biology Vol. V, Proceedings of the Fifth
Interdisciplinary Conference on Marine Biology, J. D. Costlow Jr.,
ed. Gordon and Breach, Science Publishers, New York, London, and
Paris. 606 pp.
Reference: Biol. Bull. 182: 54-65. (February. 1992)
The Role of Shell Granules and Accessory Cells
in Eggshell Formation in Convoluta pulchra
(Turbellaria, Acoela)
RESA M. CHANDLER,1 MARY BETH THOMAS,1 AND JULIAN P. S. SMITH, III2
Department of Biology, The University oj North Carolina at Charlotte,
Charlotte, North Carolina 28223
Abstract. Most turbellarian embryos are surrounded by a sclerotinized eggshell originating from polyphenol-con- taining eggshell-forming granules (EFGs). Although em- bryos of the acoel Convoluta pulchra are surrounded by a shell, it is not sclerotinized. Therefore, in the absence of polyphenols as a marker for EFGs, it was not clear which, if any, of the granules of the oocyte function in eggshell synthesis. In this study, electron-opaque, elliptical granules with a characteristic frothy component and a diameter of 480 nm were identified in the oocyte as EFGs by their participation in eggshell formation. In addition, it was shown that accessory cells to the oocyte initiate synthesis of the shell by producing a thin, granular, elec- tron-opaque primary shell, against which the contents of the EFGs are released by exocytosis. Morphological com- ponents of the shell and stages of its synthesis are de- scribed. A second type of membrane-bound granule and the lipid droplets that occur in the ooplasm were found not to be involved in eggshell formation and are probable sources of nutrients for the developing embryo. Possible implications of the findings for taxonomy and phylogeny are discussed.
Introduction
The zygotes of acoel turbellarians, like those of all pla- tyhelminthes studied to date, are enclosed in a shell fol- lowing fertilization (see Rieger et ai, 1991, for Turbellaria; Fried and Haseeb. 199 1 , and Coil, 199 1 . for parasitic pla- tyhelminths). In all cases so far examined, the eggshell
Received 24 May 1991; accepted 31 October 1991.
Present addresses: 'Department of Biology, The University of North Carolina at Charlotte. Charlotte, NC 28223, and :Department of Biology. Winthrop College, Rock Hill. SC 29733.
appears to arise from eggshell-forming granules (EFGs) exocytosed from one of the cells that ultimately comes to lie within the shell (oocytes in the entolecithal archo- ophorans, yolk cells in the ectolecithal neoophorans and parasitic platyhelminths). Among archoophoran turbel- larians, EFGs have been described by transmission elec- tron microscopy (TEM) from the oocytes of at least one member of most orders (Polycladida: Boyer, 1972; Do- menici et ai, 1975; Gammon, 1979; Ishida et ai, 1981; Espinosa, 1986; Ishida and Teshirogi, 1986; Macrosto- mida: Gremigni et al.. 1987; Kucera, 1987; Acoela: Gre- migni, 1988; Smith et al., 1 988; Nemertodermatida: Smith et ai, 1988). Relatively few TEM studies of archoophor- ans, however, have examined formation of the eggshell (see Ishida, 1989; Falleni and Gremigni, 1989).
Whereas EFGs in most platyhelminths examined to date contain polyphenols, they have not been found in the oocytes of the Acoela (Thomas et ai, 1985; Chandler and Thomas, 1986, 1987; Gremigni, 1988; Smith et ai, 1988; Falleni and Gremigni, 1989, 1990) or the Nemer- todermatida (Thomas et ai. 1985; Smith et ai. 1988), the presumably primitive turbellarian orders that consti- tute the Acoelomorpha. In the acoels, the shell is protein- aceous and non-sclerotinized (Falleni and Gremigni, 1989). Because the eggshells of all other non-acoelomor- phan platyhelminths studied so far appear to be sclero- tinized, the process of eggshell formation in the acoels merits further study. For example, in the absence of the polyphenolic marker, it is difficult to ascertain with cer- tainty which, if any, of the several types of granules in the oocyte give rise to the eggshell. Falleni and Gremigni (1989) have implicated a population of electron-opaque granules with a diameter of 0.4-0.5 /urn as EFGs in the oocytes of the acoel " Convoluta psammophy la" (? = Pae-
54
SHELL DEPOSITION IN AN ACOEL
55
domecynostomum psammophilum Beklemischev, 1957), but have not examined the mechanism by which these granules contribute to the formation of the shell. Although it seems likely that exocytosis of the granules, which occurs during eggshell formation in other turbellarians, is in- volved in production of the eggshell of the acoels (Falleni and Gremigni, 1990), that is not clear from the published studies.
Also unclear is the question of whether cells other than the oocyte are involved in synthesis of the eggshell in acoels, as they are in other turbellarians (e.g., Giesa, 1966; Bunke, 1982; Ishida, 1989). Mature oocytes in both acoels and nemertodermatids are nearly always surrounded by "follicle" or "accessory" cells, whose function has not been elucidated, although it is usually suggested that they are responsible for heterosynthetic yolk production or that they assist in the production of the eggshell (see Rieger et ai. 1991).
The present study examines eggshell formation in the acoel Convoluta pulchra with particular attention to the following questions: ( 1 ) do any of the granules of the oo- cyte participate in the formation of the eggshell; (2) if so, what is the mechanism by which they participate; (3) are other cell types involved in eggshell synthesis; and (4) what is the morphology of the shell itself? A preliminary report of these findings has been presented elsewhere (Chandler eta/.. 1988).
Materials and Methods
Experimental organism
Convoluta pulchra (Family Convolutidae; Smith and Bush, 1991) was extracted according to the methods of Hulings and Gray ( 1 974) from sediment samples collected at mid-tide level at a coastal inlet near Fort Fisher, North Carolina.
Procedures for preliminary observations
To determine the time course of egg-laying in Con- voluta pulchra. gravid acoels were isolated in pairs in wells of Falcon® 96-well plates. The individual cultures were maintained in Millipore®-filtered seawater (MFSW) at a constant temperature of 20°C and a light:dark cycle of 16:8 h. The worms were monitored closely to deter- mine if the acoels lay their eggs according to a diurnal pattern. Under these conditions, worms began egg-lay- ing within approximately 1 to 1.5 h after exposure to light.
Experimental design
Ten pairs of acoels with large eggs were placed in Fal- con® plates. The animals were placed into darkness at 9:30 p.m. and returned to light at 6:30 a.m. Worms were
fixed for electron microscopy every half hour, from 7:00 a.m. until 1 1:30 a.m. Other worms were allowed to lay eggs, which were fixed for electron microscopy.
Procedures for electron microscopy
Adult worms and laid eggs were fixed in 1% glutaral- dehyde, 4% paraformaldehyde, 0. 1 M HEPES buffer (pH 7.4), 1 mA/ CaCl2, and 10% sucrose [modified from McDowell and Trump (1976)], rinsed in buffer, post-fixed in HEPES-buffered 1% OsO4, and embedded in Spurr's low viscosity epoxy resin (Spurr, 1969). Ultrathin sections were stained with uranyl acetate (Watson, 1958) and lead citrate (Reynolds, 1963), and examined with a Philips 20 1C transmission electron microscope.
Procedures for morphometric analysis
The Feret diameters (see Weibel, 1979) of the three types of granules were measured with a Zeiss ZIDAS dig- itizer. The Feret diameter of each granule profile was measured between two lines parallel to the long axis of the photographic print. The most mature stage of devel- opment in which all types of granules were present (Pri- mary Shell Synthesis Stage, described below) was chosen for measurement. To avoid measuring the same granule more than once, 1 .8 /nm separated the thin sections mea- sured and non-overlapping micrographs were taken from each thin section examined. Approximately 200 granules from electron micrographs magnified 28.000X were mea- sured and the size-frequency distribution of Feret diam- eters for granule Types A and B were plotted. For the Type A granules, the distribution was corrected for profiles overlooked in the smallest categories as described in Wei- bel (1979). The actual diameter of Type A granules (D) was estimated from the average Feret diameter (d) using
4 the relationship D ^ --d (Weibel, 1979).
7T
Oocytes at different stages of shell maturation were an- alyzed to determine the volume densities ( Vv; % of oocyte volume occupied by granules) of Types A and B granules and lipid droplets. Non-overlapping micrographs along two right-angled transects were taken from the germinal vesicle to the oolemma. Volume densities were deter- mined by point-count analysis, using the oocyte as the reference space (Weibel. 1979). To determine whether these volume densities changed in the oocyte as devel- opment of the shell proceeded, the volume densities were arcsin-square root transformed and subjected to ANOVA Planned Comparison.
The embryos contained within laid eggs fixed for elec- tron microscopy were observed to be separated from the inner edge of their shells. This could occur if the shell swells and lifts away from the embryo or if the embryo
•'•., •*••• - *u*-; > <4 '
D» _ .._.- » i •»
» • f i j ^ •*
. , .,, - 1 « c • * • - *
:. >
Figure \. (A) Overview of a column of maturing oocytes showing the nuclei (germinal vesicles) with large nucleoli (Nu) and the increase in the number ot granules in the ooplasm with development. Ol = oocyte synthesizing Type A granules only: O2 = oocyte synthesizing Types A and B granules; O3 = oocyte containing Type A and B granules and lipid droplets. Bar = 15.0 ^m. (B) Cisternae of rough endoplasmic
56
SHELL DEPOSITION IN AN ACOEL
57
Type A
16
5 6 7 8 9 1C 11
Feret Diameter (x 100 nm)
13 H 15
TypeB
find
_n
567 8 9 1C 11 Feret Diameter (x 100 nm)
12 13 14 15
Figure 2. Feret diameter distribution of granule Types A and B. Each number on the abscissa represents the stated value ±0.5; n = 200.
loses some of its volume, shrinking away from the shell. Since the two possibilities affect the interpretation of the changes in volume densities of the granules, the absolute volumes of eggs prior to egg-laying and of embryos after egg-laying were determined. Serial 2 ^in-thick sections were viewed with a Zeiss Axioskop light microscope equipped with a Sony DXC-3000A color video camera and a Sony PVM-1343MD Trinitron color monitor. Each section of egg or embryo was traced from the monitor screen onto transparent plastic. The ZIDAS was used to calculate the area of each drawn section; the area, in turn, was multiplied by the thickness of the sections and these numbers summed for all sections to determine the volume of each egg or embryo. Four eggs with mature shells and four laid eggs were analyzed. ANOVA was used to com- pare the