THE JOURNAL OF EXPERIMENTAL ZOOLOGY 282:421���437 (1998) �� 1998 WILEY-LISS, INC. Microscopic Structure of the Gravid Uterus in the Little Skate, Raja erinacea T.J. KOOB1,2* AND W.C. HAMLETT1,3 1The Mount Desert Island Biological Laboratory, Salsbury Cove, Maine 04672 2Skeletal Biology, Shriners Hospital for Children, Tampa, Florida 33612 3South Bend Center for Medical Education, Indiana University School of Medicine, Notre Dame, Indiana 46556 ABSTRACT The morphology of the uterus in oviparous elasmobranchs has not been systemati- cally examined in any species. The uterus in oviparous sharks and skates houses the eggs during the crucial chemical events that result in the requisite polymerization of the assembled egg cap- sule materials. Capsule sclerotization involves a quinone tanning mechanism in which catechols are introduced in utero and subsequently oxidized to quinones prior to oviposition. The uterus in Raja erinacea is morphologically specialized for four well-defined functional activities associated with egg capsule formation: regionally distinct structural modifications, both in the uterine wall and the epithelial lining, for active movement of the capsule through the uterine lumen biosynthe- sis and secretion of materials into the lumen vascular facilitation for oxygen transfer to fuel the oxidation process involved in capsule tanning and intercellular spaces for removal of water from the uterine lumen. The first three activities are qualitatively similar to those operating in vivipa- rous species. The uterus throughout its length has longitudinal folds punctuated by secretory crypts. Proceeding from the anterior to the posterior end of the uterus there is a progressive decrease in the number of cilia and the depth of the lamina propria, and an increase in vascularity and the width of the muscularis. J. Exp. Zool. 282:421��437, 1998. �� 1998 Wiley-Liss, Inc. Despite the fact that nearly half of all elasmo- branch species and all chimaeroids utilize a spe- cialized oviparous mode of reproduction, little attention has focused on the physiology of the uterus during egg laying cycles. Only one report on the ultrastructure of the uterus in an ovipa- rous species (Cephaloscyllium umbratile) has ap- peared in recent literature, and the data consisted of a single light micrograph and a single trans- mission electron micrograph (Otake, ���90). This apparent lack of interest is likely the result of the often repeated hypothesis that oviparity, as exhibited by extant species, represents the primi- tive condition or pleisiomorphic reproductive mode of shark ancestors (Wourms, ���77, ���81 Wourms et al., ���88 Hamlett, ���89 Compagno, ���90 Otake, ���90 Callard et al., ���96). Yet oviparous elasmobranchs produce structurally and chemically complex cap- sules from a well developed, endocrine regulated, morphologically complex gland in the reproduc- tive tract (see Hamlett et al., ���98, this issue), and the chemical processes leading to the requisite physicochemical properties of the oviposited cap- sule take place in the uterus (Koob and Cox, ���90). Moreover, a recent cladistic analysis of evolution- ary transitions in reproductive modes concluded that oviparity in skates evolved from live-bear- ing ancestors (Dulvy and Reynolds, ���97). If ovi- parity in skates is the derived reproductive mode, we would predict that certain uterine specializa- tions might be related to those operating in vi- viparous species. The little skate (Raja erinacea) is typical of oviparous elasmobranchs in producing fertile eggs encapsulated in morphologically complex cap- sules. The capsule must endure protracted expo- sure to the corrosive action of seawater and protect the developing embryo from mechanical injury and attacks from predators. The process that produces such a stable matrix entails com- plex mechanisms for both the assembly and po- lymerization of the precursor materials. Capsule formation begins before ovulation with the secre- tion and assembly of capsule precursors in the Grant sponsors: Indiana University School of Medicine Shriners of North America. *Correspondence to: Dr. T.J. Koob, Skeletal Biology, Shriners Hos- pital for Children, Tampa, Florida 33612. E-mail: tkoob@earthlink.net

422 T.J. KOOB AND W.C. HAMLETT oviducal gland. The assembled portions of the cap- sule then move out of the gland lumen and into the uterus. Stabilization of the assembled precur- sors occurs first during the assembly process in the gland, in part via disulphide bond formation. The requisite sclerotization of the capsule to its final form then proceeds and is completed in the lumen of the uterus. Little skate egg capsules are stabilized by a quinone tanning process involving the introduc- tion and oxidation of catechols (Koob and Cox, ���90). Tyrosine residues in the capsule precur- sors are converted to 3,4-dihydroxy���phenylala- nine (DOPA) by tyrosine hydroxylase activity once the assembled materials leave the ovidu- cal gland (Koob, ���92). A portion of these DOPA residues subsequently is oxidized to quinone by a catechol oxidase (Koob and Cox, ���88). Both ty- rosine hydroxylation and DOPA oxidation are set in motion immediately after the assembled cap- sule moves out of the oviducal gland and enters the proximal portion of the uterus, suggesting that conditions in the uterine lumen promote the quinone tanning mechanism. Completion of the tanning process occurs entirely in utero. The present work was undertaken to determine whether morphological specializations of the uterus are centrally involved in capsule forma- tion and polymerization. The gravid uterus of Raja erinacea was exam- ined by light microscopy and scanning and trans- mission electron microscopy. Uterine modifications of the various segments of the reproductive tract in female elasmobranchs reflect the requirements for each type of reproductive mode (Hamlett and Hysell, ���98). Oviparous species have previously been considered primitive and unspecialized re- garding physiological maintenance of the uterine environment. We question this assumption. This study represents the first comprehensive ultra- structural description of the uterus in a gravid oviparous female. MATERIALS AND METHODS Gravid female little skates were obtained from Frenchman Bay (Mount Desert Island, Maine) and maintained in aquaria at the Mount Desert Island Biological Laboratory. Females were anes- thetized with MS-222, humanely sacrificed by pithing, opened by a longitudinal ventral incision, and the uteri isolated. The uteri were tied off an- terior to the oviducal gland, and near the entrance to the cloaca, before being removed. The uteri were opened with surgical scissors and small pieces of the uterus were removed and placed di- rectly into primary fixative. For scanning electron microscopy (SEM), tissues were immersed in a primary fixative consisting of: 3% glutaraldehyde in 0.1 M phosphate buffer with 0.4 M sucrose and 10 drops of 1% CaCl2 per 10 ml fixative. Fixation was carried out at room tem- perature for 6 to 8 hr. Tissues then were washed several times in fresh buffer. Samples were trans- ferred to a secondary fixative (1.0% osmium tetrox- ide in 0.1 M phosphate buffer) and post-fixed for 1.5 hr at room temperature. Specimens subse- quently were washed with distilled water and de- hydrated through a graded series of tertiary butanol from 30% to 100%. Specimens were fro- zen and then transferred to a lyophilizer. Dried specimens were mounted on double-sided plastic tape to which silver conductive paint was added. Samples were coated with a thin layer of gold and viewed on a JEOL JSM-T300 SEM at 10���25 kV. For transmission electron microscopy (TEM), the fixation and dehydration steps were the same as for SEM except the secondary fixative was 2.0% osmium tetroxide in 0.1 M phosphate buffer for 2 to 4 hr. Following dehydration, the tissue was transferred through three changes of propy- lene oxide of 5 min each at room temperature. Tissues were infiltrated in a 1:1 volumetric ratio of propylene oxide to catalyzed Embed 812, then 1:2, and, finally, pure resin overnight. Samples were cured at 60��C for 72 hr under vacuum. Blocks of tissue were sectioned either with glass or diamond knives on a LKB Ultratome IV. One to two micrometer thick sections were cut for light microscopy and stained with toluidine blue. Silver or gray sections were picked up on acid-cleaned copper grids. Sections on grids were stained with uranyl acetate and lead citrate. Stained grids were examined on a Hitachi H600 TEM at 75 and 100 kV. For light microscopy (LM), tissue samples were fixed in neutral buffered formalin. Samples were dehydrated through a graded series of alcohols and embedded in JB-4 glycol methacrylate (Poly- sciences). Sections, 2 microns thick, were made with glass knives and affixed to glass slides. Sec- tions were stained with either toluidine blue or methylene blue-basic fuchsin. For light micro- scopic histochemistry, tissues were stained for acid mucopolysaccharides by the periodic acid- Schiff (PAS) (Humason, ���79). Sections were pho- tographed on a Nikon Optiphot-2 light microscope equipped with a Microflex AFX-DX photomicro- graphic attachment.

GRAVID UTERUS IN THE LITTLE SKATE 423 Entire reproductive tracts containing eggs were immersed in a solution of 10% neutral buffered formalin. These specimens were retained in the author���s (WCH) possession at Indiana University School of Medicine. RESULTS The uteri of skates containing �� partially formed egg capsules were examined. The body of the capsule is nearly complete but the capsule lacks anterior tendrils (Fig. 1). The uterus is closely associated with the capsule, especially in the posterior region. The oviducal gland has ex- panded, as has the anterior portion of the uterus, to accommodate the egg capsule. We chose three segments of the uterus for examination (Fig. 1). The first (zone A) is the broadest portion of the uterus immediately adjacent to the posterior mar- gin of the oviducal gland, a zone we designated as the terminal zone (Hamlett et al., ���98, this is- sue). The second segment (zone B) corresponds to the posterior region of the capsule, excluding the posterior tendrils. The final segment (zone C) is in the region of the posterior tendrils. When the segments are examined by light mi- croscopy of glycol methacrylate-embedded tissue, striking differences are noted in the overall orga- nization of the organ. Zone A is characterized by an undulating surface epithelium and a very broad lamina propria (Fig. 2). The lamina pro- pria is a hydrated areolar connective tissue zone containing large blood vessels. The muscularis is modest and is composed of inner longitudinal and outer circular layers of smooth muscle bounded by a simple squamous serosa. Zone B has a dra- matically reduced lamina propria and a more prominent muscularis (Fig. 3). Zone C has an in- distinct lamina propria and a substantial mus- cularis (Fig. 4). In females actively secreting egg case compo- nents and assembling them into a nascent egg case, the uterus assumes differing states of mor- phology depending on the degree of tanning of the egg capsule. The region nearest the terminal portion of the oviducal gland is broadly dilated and imbibed with much fluid in preparation for the arrival of the egg case and egg from the ovary and oviducal gland. Scanning electron microscopy reveals the internal surface contour of the uterus is characterized by longitudinal and interlocking Fig. 1. Gross micrograph of the gravid skate uterus con- taining a �� formed capsule. 0 = oviduct g = oviducal gland line of arrowheads = posterior margin of the egg capsule ex- cluding posterior tendrils A, B, C = zones large arrowhead = posterior tendril. Scale bar in mm.