The Normal Genital Tract of the Female California Sea Lion Zalophus californianusCyclic Changes in Histomorphology and Hormone Receptor Distribution.код для вставкиСкачать
THE ANATOMICAL RECORD 292:1801–1817 (2009) The Normal Genital Tract of the Female California Sea Lion (Zalophus californianus): Cyclic Changes in Histomorphology and Hormone Receptor Distribution KATHLEEN M. COLEGROVE,1* FRANCES M.D. GULLAND,2 DIANE K. NAYDAN,3 AND LINDA J. LOWENSTINE1 1 Department of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, University of California, Davis, California 2 The Marine Mammal Center, Marin Headlands, Sausalito, California 3 Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California, Davis, California ABSTRACT Changes in reproductive tract histomorphology, and estrogen (ERa) and progesterone receptor (PR) expression throughout the breeding cycle were evaluated in free-ranging stranded female California sea lions (Zalophus californianus). Hormone receptor expression in the ovaries, uterus, cervix, and vagina was evaluated using an immunohistochemical technique with monoclonal antibodies. During a large portion of the cycle, ovaries contained both a corpora lutea (CL) and follicles in varying stages of development. In the periods of pupping and estrus during June and July, and in the spring morphologic features of the endometrium suggested estrogen influence. There were areas of squamous differentiation in the pseudostratified columnar epithelium of the cervix and vagina in some animals during estrus and in the spring. Estrogen receptor immunohistochemical scores were highest during pupping and estrus and in the spring and lowest during embryonic diapause. Cyclic changes in uterine PR expression throughout the cycle were minimal. Both ERa and PR were expressed in epithelial and stromal cells throughout the cervix and vagina, however, receptor expression was typically higher in the stroma. Stromal cell hormone receptors may play an important role in epithelial responses to circulating sex hormones. The results of this investigation add to the general knowledge of California sea lion reproduction and establish baseline information on reproductive tract hormone receptors that will aid in determining the factors involved in urogenital C 2009 cancer development in sea lions. Anat Rec, 292:1801–1817, 2009. V Wiley-Liss, Inc. Grant sponsor: National Institutes of Health (Postdoctoral Training in Environmental Pathology); Grant number: 5T32ES007055-30; Grant sponsor: West Coast Center of Excellence (NOAA Oceans and Human Health Initiative); Grant number: AB133F05SE5112. *Correspondence to: Kathleen M. Colegrove, Zoological Pathology Program, College of Veterinary Medicine, University of Illinois, Loyola University Medical Center Building 101 Room C 2009 WILEY-LISS, INC. V 0745, 2160 South First Avenue, Maywood, IL 60153. Fax: 708216-5934. E-mail: firstname.lastname@example.org Received 18 June 2008; Accepted 7 July 2009 DOI 10.1002/ar.21009 Published online 18 September 2009 in Wiley InterScience (www. interscience.wiley.com). 1802 COLEGROVE ET AL. Key words: California sea lion; embryonic diapause; estrogen receptor; progesterone receptor; reproductive tract; urogenital cancer; Zalophus californianus Evaluation of the normal reproductive cycle in wild marine mammals is often hindered by limited access, low sample numbers of healthy individuals, and lack of a thorough reproductive history. Currently, opportunistic sampling of deceased stranded animals is the primary means of studying normal anatomy, histomorphology, and diseases of free-ranging animals (Gulland, 1999). Gaining an understanding of normal reproductive physiology is crucial, however, for accurate evaluation of population demographics, causes of reproductive loss, and the potential effects of environmental contaminant exposure. In all pinniped species, the annual reproductive cycle includes synchronized parturition, estrus, embryonic diapause, placental pregnancy, and lactation. The most detailed information on the reproductive cycle of pinnipeds comes from investigations of wild northern fur seals (Callorhinus ursinus) and harbor seals (Phoca vitulina). Histologic evaluations of normal reproductive tracts in both species have shown cyclic morphologic changes corresponding to ovarian cycling and pregnancy (Craig, 1964; Bigg and Fisher, 1974). During pinniped embryonic diapause, conceptus development arrests at the blastocyst stage with the zona pellucida intact. The blastocyst remains unattached in the uterus throughout diapause until implantation (Boshier, 1981). Circulating reproductive hormones have been measured in fur and harbor seals in several studies (Daniel, 1974; Reijnders, 1990; Browne et al., 2006), however, the physiology and endocrine regulation of diapause and reproduction in pinnipeds remains poorly understood. Like other pinniped species, California sea lions (Zalophus californianus) have an annual reproductive cycle that is highly synchronized. Females give birth to a single pup in June on breeding rookeries on the Channel Islands of California and islands off of Baja California, Mexico. On San Nicolas Island, California, the majority of pups are born during the first week in June (Odell, 1975). Estrus and conception occur 3–4 weeks after parturition. Embryonic diapause lasts approximately 2 months, with implantation occurring by early October. Placental gestation lasts 8–9 months till the following June (Reeves et al., 1992). Relatively little is known about seasonal sex hormone fluctuations and corresponding changes in reproductive tract histology in California sea lions. A recent study investigated seasonal changes in circulating serum progesterone and estrogen in freeranging and captive female sea lions. Progesterone concentrations increased in the fall irrespective of pregnancy status and progesterone concentrations were higher in pregnant animals when compared with nonpregnant animals in the spring. In the majority of animals, there was no clear pattern in the estrogen concentration (Greig et al., 2007). Sex steroids exert their effects through interaction with intracellular steroid hormone receptors located in target tissues. Accordingly, estrogen (ERa), progesterone (PR), and androgen receptors (AR) are located throughout the reproductive tract of mammalian species, mediating the effects of steroid hormones on reproductive organs. The distribution of steroid hormone receptors in the human female reproductive tract, in relationship to cyclic changes and histomorphology, has been well documented (Press et al., 1986). Similar studies have documented hormone receptor distribution in the reproductive tract of animal species including the dog (De Cock et al., 1997; Vermeirsch et al., 2000a; Vermeirsch et al., 2002), cat (Li et al., 1992), mare (Watson et al., 1992), sow (Sukjumlong et al., 2003), and cow (Kimmins and MacLaren, 2001). To our knowledge, reproductive tract hormone receptors have not been examined in any marine mammal species. A high prevalence of aggressive urogenital carcinomas has been documented in California sea lions stranding along the central California coast. These tumors arise in the cervix and vagina of females and the penis, prepuce, and urethra of males. Widely disseminated metastases are often found in thoracic and abdominal viscera of affected animals (Gulland et al., 1996; Lipscomb et al., 2000). Some sea lions dying from other causes have histologic evidence of early neoplasia in genital epithelium consisting of epithelial hyperplasia, dysplasia, and carcinoma in situ. Differentiating between early neoplasia and epithelial changes due to normal cyclic hormone changes is sometimes difficult. The purpose of this investigation was to describe the normal cyclic changes in histomorphology and steroid hormone receptor distribution in the genital tract of normal free-ranging female California sea lions throughout the reproductive cycle. These observations will provide additional baseline information on the normal physiology and endocrine regulation of reproduction in sea lions and will allow for better differentiation between normal and precancerous epithelium. Additionally, defining the hormone receptor distribution in tissues prone to cancer will aid in identifying factors, such as endogenous hormones or environmental contaminants, which may play a role in cancer development. MATERIALS AND METHODS Animals Formalin-fixed whole reproductive tracts and archived paraffin-embedded tissues opportunistically sampled from California sea lions that stranded live on the central California coast (37 420 N, 123 050 W to 35 590 N, 121 300 W) were examined. Adult female sea lions (N ¼ 24) and cycling juvenile female sea lion (N ¼ 2) that died during rehabilitation at The Marine Mammal Center (TMMC) in Sausalito, California were evaluated. Animals dying during all months of the yearly reproductive cycle were represented. Causes of death included domoic acid toxicity (N ¼ 14), trauma (N ¼ 4), leptospirosis (N ¼ 4), pneumonia (N ¼ 1), enteritis (N ¼ 1), and peritonitis (N ¼ 1), pyelonephritis (N ¼ 1). Animals selected for the study had no significant reproductive 1803 FEMALE SEA LION REPRODUCTIVE TRACT tract lesions based on gross and histologic examination. Age class was estimated by standard length (measured from nose tip to tail tip), tooth dentin growth rings (Oosthuizen et al., 1998), and gross and histologic evaluation of the uterus for evidence of previous pregnancy. Histomorphology A gross necropsy was performed on all animals within 24 hr of death. Routine tissue samples, including the entire reproductive tract, were fixed in 10% neutral buffered formalin, processed routinely for paraffin-embedding, sectioned at 5 lm and stained with hematoxylin and eosin for histologic examination. Sections examined histologically included cross sections of both uterine horns and midsagittal sections through both ovaries, longitudinal sections through the cervix, and transverse sections of the cranial and middle vagina. Ovaries were evaluated for the presence of follicles, corpora lutea (CLs), and corpora albicans (CAs). Primary follicles had an oocyte surrounded by a single layer of cuboidal cells. Follicles were defined as secondary follicles if they had two or more layers of granulosa cells but without an antrum. Tertiary (antral) follicles contained a fluid filled cavity with multiple layers of granulosa cells surrounded by the theca interna, and theca externa. Folliculogenesis was defined as the presence of secondary and antral follicles within the ovary. In degenerating CLs, cells were variably shrunken with vacuolated cytoplasm, and were dissected by bands of fibrous connective tissue. Corpora albicans were defined as structures containing hyalinized eosinophilic fibrous connective tissue with no remaining luteal cells (Senger, 2005). Atretic follicles were present in most ovaries and were not quantified. The size, shape, and degree of cytoplasmic vacuolation of stromal interstitial cells were determined for each ovary. Uterine surface and glandular epithelia were evaluated based on the following features: shape and height of epithelial cells, cytoplasm vacuolation (low, moderate, heavy), secretion (absent, low, moderate, high), tortuosity (low, moderate, high), and the presence of subnuclear glycogen vacuoles (SNV). Uterine stromal edema and vascularity were also evaluated. In animals that were pregnant or post parturient at the time of death, the pregnant and nonpregnant horns were evaluated separately. Epithelium of the cervix and vagina were evaluated for the following features: thickness of the epithelium, height of epithelial cells, cytoplasmic vacuolation, degree of invagination, gland formation, and presence of goblet cells. Immunohistochemistry Immunohistochemistry was performed on serial sections of urogenital tissue from a subset of 15 females in which tissues were best preserved using a strepavidin-biotin method. Females evaluated had died during the time of pupping and estrus (N ¼ 4), diapause (N ¼ 3), fall (N ¼ 4), and winter to spring (N ¼ 4). After deparaffinizing, endogenous peroxidase activity was blocked by incubating sections in 0.2% H2O2 in methanol for 30 min. Antigen retrieval was accomplished for ERa and PR by heating sections to 95 C for 30 min in Citrate buffer (pH ¼ 6.2) (Dako Cytomation, Carpinteria, CA) in a rice steamer. Sections were incubated in 3% normal goat serum for 30 min. Sections were incubated with a monoclo- nal antibody to human ERa (1:125:1D5 Immunotech, Marseille, Cedex 9, France) and a monoclonal antibody to human PR (1:200:10A9 Immunotech, Marseille, Cedex 9, France) overnight at 4 C in a moist chamber. For Ki67, sections were incubated with the primary monoclonal antibody (1:75: MIB-1 Dako Cytomation, Carpinteria, CA). Slides were incubated with a biotinylated antimouse link reagent (Biocare Medical, Concord, CA) for 10 min and then incubated with streptavidin horseradish peroxidase (Biocare Medical, Concord, CA) for 10 min. Positive staining was visualized using 3-amino-9-ethylcarbazole (AEC) chromogen (Zymed Labs, San Francisco, CA). After all steps, sections were rinsed in phosphatebuffered saline (PBS) spiked with polyoxyethylenesorbitan monolaurate (TWEENV 20, SigmaAldrich, St. Louis MO). Sections of canine uterus known to be positive for ERa and PR were included in each procedure as positive controls. Antibodies utilized have been shown to be cross reactive to ERa and PR in an number of animal species (Vermeirsch et al., 2000a; Vermeirsch et al., 2000b; Martin de las Mulas et al., 2002; Vermeirsch et al., 2002; D’Haeseleer et al., 2006; D’Haeseleer et al., 2007). Negative controls were sections of sea lion uterus incubated with omission of the primary antibody. Immunostaining of all sections was evaluated by a single person without prior knowledge of the animal from which the tissue was sampled. Expression of ERa and PR was scored using a semiquantititative grading system identical to the grading system used in similar canine studies (De Cock et al., 1997; Vermeirsch et al., 1999; Vermeirsch et al., 2002). For each tissue examined, the representative section was dividend into five regions of approximately equal surface area and randomly selected areas evaluated within those regions were selected prior to observation of immunohistochemical stains. For each of the five sections, 100 of 500 cells were evaluated. A total score was calculated by the following formula TS ¼ PS þ IS, where TS is the total score, PS is the proportional score and IS is the intensity score. The proportional score corresponded to the percentage of 500 cell nuclei that stained positive. The intensity score reflected the subjectively evaluated intensity of positive, brown— red staining (Table 1). Proliferation index was calculated as a percentage of positively stained uterine epithelial nuclei out of 500 total evaluated cells. R Serum Hormone Analysis Serum archived at 80 C at the time of death was available from 11 animals for steroid hormone analysis. Serum samples were assayed for progesterone by enzyme immunoassay (EIA) and estradiol by radioimmunoassay (RIA) at the Clinical Endocrinology Laboratory, Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, CA (Shille et al., 1979; Munro and Stabenfeldt, 1984). Statistics The Kruskal-Wallis, nonparametric test was used to analyze the difference in total immunohistochemical scores between the four different times of the reproductive cycle. The Spearman rank correlation test was used to analyze potential correlations between receptor scores 1804 COLEGROVE ET AL. TABLE 1. Immunohistochemistry grading scores Intensity score Score Score Score Score 0 1 2 3 No staining Weak staining Moderate staining Strong staining Proportional score Score Score Score Score Score Score 0 1 2 3 4 5 No staining <1% positively stained nuclei 1–9% positively stained nuclei 10–32% positively stained nuclei 33–65% positively stained nuclei >65% positively stained nuclei Total scores were calculated by adding the intensity and the proportional scores. Fig. 1. Endometrial surface at a site of previous placental attachment in a California sea lion (Zalophus californianus). Scale bar ¼ 2.0 cm. and serum hormone levels. Statistical calculations were performed using MedcalcV statistical software, Version 22.214.171.124 1993 (Medcalc, Mariakerke, Belgium). A P-value < 0.05 was considered statistically significant. R RESULTS Gross Morphology The reproductive tract of female sea lions is similar to other phocids and otariids. The oblong ovaries are encased in a thick bursa that can become enlarged and contain small amounts of mucous in adult animals. Long arborizing oviduct fimbria extend over the surface of the ovary. The uterus is bicornuate and in a previous investigation of 52 archived specimens the uterine horns averaged between 11.0 and 12.0 cm in length after formalin fixation (Lowenstine, unpublished data). In multiparous adults, the uterine serosa is coarsely nodular, corresponding to the numerous thick-walled blood vessels that remain after pregnancy. The endometrial surface ranges from light pink to tan with prominent longitudinal folds extending along the length of the uterine horns. In post parturient animals, there is a well demarcated, circumferential, yellow brown band of variable width visible on the endometrial surface, corresponding to the site of previous attachment of the zonary placenta (Fig. 1). There are two internal cervical ostia and the internal cervical mucosa extends into the uterine horns such that there is no appreciable uterine Fig. 2. Anatomic features of the female California sea lion (Zalophus californianus) reproductive tract. Formalin-fixed. (a) uterine horns; (b) external uterine bifurcation, (c) internal cervical bifurcation; (d), external cervical fold; e, urinary bladder; f, hymenal fold; g, urethral protuberance; h, clitoris. body (Fig. 2). The length of the cervical canal from the bifurcation to the external cervical fold averaged 5.6 cm in a previous investigation (Lowenstine, unpublished data). There are no cervical rings and longitudinal folds extending along the length of the cervix. There are small transverse ridges in the cranial vagina and a prominent hymeneal fold in the middle portion of the vagina. A triangular urethral tubercle lies along one side of the hymeneal fold directly above the urethral opening. It has been postulated that the prominent hymeneal fold of pinnipeds is an adaptation to prevent entry of water FEMALE SEA LION REPRODUCTIVE TRACT into the urogenital canal during diving (Harrison, 1969). Longitudinal folds extend along the caudal vagina and the clitoris is sheathed in the clitoral fossa along the ventral commissure of the vestibule. During estrus the clitoris can become erect and protrude from the vestibule. Small, dark brown submucosal vestibular glands (Bartholin’s glands) can be noted on cut sections of the vestibule-vagina junction. Histomorphology Ovary. General histologic characteristics of the sea lion ovary included prominent subsurface epithelial structures and stromal interstitial cells. Similar to the bitch, the subsurface epithelial structures are invaginations of the surface (germinal) epithelium that extend into the superficial cortex forming cords and nests that often run parallel to the ovarian capsule. These structures generally increased in prominence with age and were occasionally cystic. Focal clusters of hyperplastic surface epithelium could occasionally be found along the ovarian capsule. Thick-walled hyalinized arterioles were scattered throughout the medulla and in some females there was segmental subintimal vascular mineralization. Thick, folded, eosinophilic membranous structures were often scattered in stroma. These structures were interpreted to be remnants of atretic follicles. Interstital cells in the stroma surrounding atretic follicles were often more prominent and clustered, with moderate to large amounts of eosinophilic cytoplasm. Cyclic changes in the presence of CLs and follicles, and the appearance of stroma interstitial cells were noted throughout the yearly reproductive cycle (Table 2). During pupping and estrus in June and July, follicles were present bilaterally in most animals. In some females folliculogenesis was asymmetric, with greater numbers of antral and larger preovulatory follicles in one of the ovaries. One animal (sea lion 4) dying in early July had a large cluster of polygonal cells with abundant pale eosinophilic cytoplasm. These cells resembled the lutenized cells of a CL suggesting recent ovulation and early CL formation. In post parturient animals, CAs were sometimes seen on the ovary from the gravid horn and CLs on the contralateral ovary, suggesting ovulation occurs in alternating ovaries each year. When folliculogenesis was present, ovarian interstitial cells were often large, polygonal, and had foamy to vacuolated eosinophilic cytoplasm (Fig. 3A). During diapause, in August and September, ovaries usually contained a CL. Interstitial cells were smaller than in estrus, and contained granular cytoplasm, often with abundant light brown granular pigment consistent with lipofuscin. Ovarian morphology was highly variable during the fall, with ovaries containing a CL and follicles in varying stages. Implantation in sea lions is thought to occur during late September and early October. During this period, all of the sea lions examined had an active CL in one of the ovaries and some follicular development. Folliculogenesis was common in ovaries of animals dying from late November till May regardless of pregnancy status, however, follicles rarely reached the size of the large, preovulatory follicles noted in July. During this time period, folliculogenesis typically occurred in both ovaries, even in the presence of a large CL (Fig. 3B). Some nonpregnant animals had a degenerating CL in one ovary sug- 1805 gesting prior fetal loss. In the spring, stromal interstitial cells were often larger than in diapause and contained foamy to vacuolated cytoplasm. Uterus. Distinct changes in the morphologic appearance of the uterine mucosa were also noted throughout the yearly reproductive cycle (Table 3). During the pupping and estrus, in late June and July, folliculogenesis was accompanied by changes indicative of estrogen influence on the endometerium. Endometrial glands were straight, epithelial cells of the surface lining and glands were cuboidal to medium columnar, and there was variable edema and congestion of the stroma (Fig. 4A). By late July, the endometrium was thicker and contained more developed glands. Several animals had tall columnar surface and glandular epithelium, increased amounts of foamy cytoplasm, and moderate to high levels of proteinaceous glandular secretion. Glands in the basal region of the endometrium appeared more tortuous and epithelium in the glands was often more than one cell layer thick. The endometrial stroma was edematous and congested in several animals sampled during July. Progestational changes were more common in animals with a CL (sea lions 6, 7) or animals with evidence of recent ovulation (sea lion 4). The endometrium was thickest and contained numerous gland profiles during embryonic diapause and in the fall. Endometrial glands were tortuous and the surface and gland epithelial cells were tall columnar with abundant foamy cytoplasm indicative of a response to progesterone secretion from the CL (Fig. 4B). Animals dying during the winter and spring commonly had similar endometrial features indicative of estrogen influence, regardless of pregnancy status. Surface and gland epithelial cells were cuboidal to low columnar, had small amounts of pale cytoplasm, and there were fewer gland profiles than in diapause or during the fall (Fig. 4C). Glands were often dilated and filled with proteinaceous secretion and glands in the basal region of the endometrium appeared more tortuous. Significant stromal infiltration by leukocytes was not found at any time period of the cycle. Mitotic figures were extremely rare in the epithelium of nonpregnant uterine horns. The average proliferation index for the surface and gland epithelium was 1.2% and 1.7%, respectively in the nongravid horn. In the pregnant uterine horn at the zone of placentation, the endometrium was thin and glands were widely separated by pale eosinophilic highly congested and often edematous stroma. Glands were lined by cuboidal to low columnar epithelium with minimal cytoplasm and were sometimes dilated by eosinophilic proteinaceous fluid. The superficial epithelium in contact with the placenta was composed of shallow folds of columnar to low cuboidal epithelial cells containing only small amounts of cytoplasm mixed with occasional syncytial trophoblastic cells. Trophoblastic cells were only noted in the superficial zone of endometrial-placental contact and did not invade deeply into the endometrium. There was only mild proliferation of endometrial epithelium and no evidence of a decidual reaction in any of the pregnant sea lions. At parturition, the superficial portion of the mucosa is partially torn away, leaving the basal glands intact. The endometrial stroma in the zone of placentation of early post-parturient sea lions was expanded by edema, a November 26 Pregnant left horn Pregnant right horn Pregnant left horn Pregnant right horn Pregnant left horn Pregnant right horn Post-parturient right horn 17 18 19 20 21 22 23 24 May 10 May 12 April 11 April 19 March 5 January 19 February 18 January 12 December 16 CL, corpora lutea; CA, corpora albicans. Post parturient left horn October 31 16 25 26 October October October October 12 13 14 15 10 11 19 23 August 8 August 11 September 4 September 14 July 30 8 9 10 11 7 July 9 July 22 Post-parturient left horn Post-parturient right horn 5 6 July 7 June 23 July 6 Date collected July 8 Post-parturient left horn Post-parturient right horn Condition 4 3 1 2 Sea lion # Folliculogenesis Folliculogenesis CL Degenerating CL Secondary follicles CL Secondary follicles Folliculogenesis CL CA Tertiary follicle CL Folliculogenesis CL CL Folliculogenesis CL CL Folliculogenesis CL CA Folliculogenesis Folliculogenesis CL Folliculogenesis Folliculogenesis CA Secondary follicles Folliculogenesis Right ovary Folliculogenesis CL Folliculogenesis Folliculogenesis Folliculogenesis CL Folliculogenesis Secondary follicles CL Folliculogenesis CL Folliculogenesis CL Folliculogenesis CL Tertiary follicle CA Folliculogenesis CL Folliculogenesis luteinized cells Folliculogenesis Folliculogenesis Folliculogenesis Folliculogenesis Secondary follicles CA Left ovary Large size, polygonal, foamy cytoplasm with lipid vacuoles Small size, cuboidal, granular cytoplasm with lipofuscin Small size, cuboidal, granular cytoplasm with lipofuscin Small size, spindle to cuboidal, granular cytoplasm with lipofuscin Moderate size, cuboidal, granular cytoplasm with lipofuscin Moderate size, polygonal, granular cytoplasm Moderate size, cuboidal, granular cytoplasm with lipofuscin Small size, spindle to cuboidal, granular eosinophilic cytoplasm with lipofuscin Small size, spindle to polygonal, granular cytoplasm with lipofuscin Moderate size, cuboidal, foamy cytoplasm with lipid vacuoles and lipofuscin Moderate size, polygonal, foamy cytoplasm with lipid vacuoles Moderate size, polygonal, foamy cytoplasm with lipid vacuoles Large size, polygonal, foamy cytoplasm Moderate size, polygonal, foamy to granular cytoplasm and lipid vacuoles Small size, cuboidal to spindle, granular eosinophilic cytoplasm Moderate size polygonal, foamy cytoplasm Large size, polygonal, foamy cytoplasm with lipid vacuoles and lipofuscin Small size, cuboidal, granular eosinophilic cytoplasm Moderate size, polygonal, foamy cytoplasm with lipid vacuoles Moderate size, cuboidal, granular cytoplasm Moderate size, cuboidal, granular eosinophilic cytoplasm Moderate size, cuboidal, granular eosinophilic cytoplasm Large size, polygonal, pale eosinophilic cytoplasm Large size, polygonal, pale eosinophilic cytoplasm Moderate size, polygonal foamy vacuolated cytoplasm Large size, polygonal, foamy vacuolated cytoplasm Interstitial cells TABLE 2. Histologic detail of the ovaries from 26 stranded California sea lions (Zalophus californianus)a 1806 COLEGROVE ET AL. 1807 FEMALE SEA LION REPRODUCTIVE TRACT Fig. 3. A, B: Ovarian structures noted in California sea lions (Zalophus californianus) collected (A) in July during pupping and estrus (Scale bar ¼ 500 lm) and (B) in the spring during pregnancy (Scale bar ¼ 1.0 mm). Note the developing follicles (FOL), subsurface epithelial structures (SES), stromal interstitial cells (INT), and the corpora lutea (CL). H&E. TABLE 3. Histologic features of the uterus of 26 stranded California sea lions (Zalophus californianus)a Sea lion # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Condition post-parturient post-parturient post-parturient post-parturient Pregnant Pregnant Post-parturient Pregnant Pregnant Post-pregnant Pregnant Post parturient Date collected Surface cell height and shape Gland cell height and shape June 23 July 6 July 7 July 8 July 9 July 22 July 30 August 8 August 11 September 4 September 14 October 10 October 11 October 19 October 23 October 31 November 26 December 16 January 12 January 19 February 18 April 11 April 29 March 5 May 10 May 12 Cuboidal Cuboidal Medium columnar Tall columnar Cuboidal Tall columnar Tall columnar Cuboidal Medium columnar Medium columnar Tall columnar Tall columnar Tall columnar Medium columnar Medium columnar Tall columnar Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Cuboidal Medium columnar Medium columnar Tall columnar Cuboidal Tall columnar Medium columnar Cuboidal Medium columnar Tall columnar Tall columnar Tall columnar Tall columnar Medium columnar Medium columnar Tall columnar Low columnar Low columnar Low columnar Cuboidal Low columnar Cuboidal Low columnar Cuboidal Cuboidal Cuboidal Gland cytoplasmic vacuolation Low Moderate, Moderate, High Low High Moderate, Low Moderate, Moderate, High High High Moderate Moderate, High Low Low Low Low Low Low Low, SNV Low Low Low SNV SNV SNV SNV SNV SNV Gland secretion Gland tortuosity High High Low Low Absent Low Low Low Moderate Moderate Low Moderate Moderate Moderate High Low High High High High High Moderate Moderate Moderate High High Low Moderate Moderate High Low High Moderate Low Moderate High High High High Moderate High High Moderate Low Moderate Low Moderate Low Moderate Low Low Low a SNV, subnuclear glycogen vacuoles. congested blood vessels, and dilated lymphatics. Endometrial glands and stromal cells were widely separated by pale staining, homogenous connective tissue, and the mucosa was infiltrated by small numbers of lymphocytes, plasma cells, and macrophages. Glands were lined by cuboidal epithelium and contained eosinophilic proteinaceous material. Mitotic figures were occasionally noted in the superficial epithelium. Later during involu- tion, the stromal connective tissue became thicker and more compact and there were increased numbers of lymphocytes, plasma cells, and hemosiderin laden- macrophages in the superficial mucosa. During diapause and in early fall, the endometrium at the previous placentation site contained large numbers of hemosiderin-laden macrophages, compact stroma, and numerous thickwalled hyalinized arterioles (Fig. 5). 1808 COLEGROVE ET AL. Fig. 5. Histologic features of the endometrium at a site of previous placental attachment in a California sea lion that died in early fall. Note the hemosiderin-laden macrophages (HM) and the hyalinized arterioles (HA). Scale bar ¼ 500 lm. H&E. Fig. 4. A–C: Histologic features of the endometrium in California sea lions (Zalophus californianus) collected (A) in July during estrus, (B) during diapause, and (C) in the spring during pregnancy. Scale bars ¼ 200 lm. H&E. Cervix and vagina. The morphologic features of the cervix and vagina varied by location and there was a lack of keratinization in all regions. In the endocervical canal, there were many long, complex epithelial infold- ings and cervical glands were often filled with large amounts of mucous (Fig. 6A). Epithelium ranged from pseudostratified columnar to stratified squamous. The endocervical stroma and smooth muscle were densely cellular and compact, resembling uterine horn stroma and myometrium. The ectocervix consisted of two portions. The prominent distal external cervical fold was lined at the tip by segmental, thin cuboidal to non-keratinized stratified squamous epithelium with few invaginations. The fornix, at the cervix–vagina junction, and ectocervix folds were lined predominately by pseudostratified columnar epithelium (Fig. 6B). There were many shallow invaginations in this region, some of which formed distinct glands. The epithelium of the cranial vagina was similar to the fornix, however, with fewer and less complex invaginations. Vaginal epithelium was predominantly pseudostratified columnar up to the middle vagina at the hymeneal fold where the mucosa transitions into non-keratinized stratified squamous epithelium (Fig. 6C). Subtle changes were observed in epithelial morphology during different times of the reproductive cycle. During pupping and estrus, the endocervical epithelium ranged from pseudostratified columnar to thicker, non-keratinized stratified squamous. Pseudostratified columnar epithelial cells had clear foamy cytoplasm and were interspersed with moderate numbers of goblet cells. In some animals, stratified squamous epithelium partially filled cervical glands and invaginations and in a few females the surface epithelium was entirely stratified squamous and cervical glands were completely filled with squamous epithelium (Fig. 7). Invaginations in the ectocervical and proximal vaginal epithelium were more numerous, prominent, and slightly deeper during June and July. Similar to the endocervix, some animals had regions of ectocervical and vaginal epithelium in which squamous epithelium was present within invaginations and glands, in addition to pseudostratified columnar epithelium. In a few animals, the columnar epithelium of the ectocervix and proximal vagina was nearly entirely replaced by well-differentiated stratified sqaumous epithelium. During diapause, FEMALE SEA LION REPRODUCTIVE TRACT 1809 Fig. 7. Ectocervical epithelium with stratified squamous epithelium in the deep aspects of the cervical glands. Scale bar ¼ 100 lm. the fall, winter, and spring was predominantly pseudostratified columnar, with a prominent basal layer, and foamy to pale eosinophilic cytoplasm. Goblet cells were more numerous in the fall and spring; however, in general, goblet cell numbers were highly variably during all times of the reproductive cycle. Several animals dying in the spring had segmental stratified squamous epithelium in the cervical glands and surface epithelium similar to that which was noted in some animals dying during pupping and estrus. Cervical and vaginal epithelium was generally thicker during pupping and estrus and during the spring, ranging from three to eight celllayers thick. Thickness was greatest in epithelium with squamous differentiation, with some invaginations containing up to 10–12 cell layers. In diapause and during the fall epithelium was notably thinner, averaged three to five cell-layers thick. Serum Hormone Analysis Serum estradiol was higher in the three animals collected during the pupping and estrus period compared to animals sampled during diapause and early fall (Table 4). Pregnant animals sampled during winter and spring, had estradiol concentrations that were as high, or higher, than those noted during pupping and estrus. Progesterone concentrations were variable but during the winter and spring were higher in pregnant animals than in a single nonpregnant female sampled. Immunohistochemistry Fig. 6. A–C: Histologic features of the (A) internal cervix (Scale bar ¼ 500 lm), (B) cervix and proximal vagina at the fornix (Scale bar ¼ 1.0 mm), and (C) middle vagina (Scale bar ¼ 200 lm). H&E. the endocervical epithelium was typically thinner and pseudostratified columnar with foamy cytoplasm. Similar to diapause, cervical, and vaginal epithelium during Positive dark red nuclear staining for both ERa and PR was observed in surface and glandular epithelial cells of the uterus, uterine stromal cells, uterine smooth muscle cells, epithelial cells throughout the cervix and vagina, stromal cells in the cervix and vaginal submucosa, distal urethral epithelial cells, and epithelium of the ovary. In all tissues, staining was multifocal, with positively stained cells interspersed with negative cells. In the ovaries, scattered surface and subsurface epithelial cells exhibited moderately intense positive ERa staining (Fig. 8). PR expression was typically absent or weak. 1810 COLEGROVE ET AL. TABLE 4. Serum estradiol (pg/ml) and progesterone (ng/ml) concentrations in 11 stranded female California sea lions (Zalophus californianus) Sea lion# 4 5 6 10 14 15 16 17 19 24 25 Condition Post-parturient Pregnant Post-parturient Pregnant Date collected Serum estradiol (pg/ml) Serum progesterone (ng/ml) July 8 July 9 July 22 September 4 October 19 October 23 October 31 November 26 January 12 March 5 May 10 10.2 5.6 10.7 3.4 4.9 3.5 4.5 14.4 27.5 10.4 4.7 3.2 0.2 12.3 2.3 2.7 5.8 9.1 2.8 5.8 10.4 0.7 In the uterus, the total ERa score was highest in the uterine glands and lowest in the surface epithelium throughout all stages of the reproductive cycle. Positive surface epithelial cells were typically clustered and the epithelial cells lining the base of the surface crypts often exhibited more intense staining. In the endometrium, ERa staining was most intense in the basal glands. In the stroma, receptor staining was more intense in stromal cells surrounding blood vessels. Cyclic changes were seen in the total uterine ERa score during the four different general time periods of the reproductive cycle; however, these differences were not statistically significant (Fig. 9A). The cyclic changes in stromal ERa staining between time periods were most dramatic. Total ERa scores in all layers of the uterus were highest during the spring and the pupping and estrous periods (Fig. 10A). Scores were lowest during embryonic diapause. In pregnant animals, ERa staining was similar in the pregnant and non-pregnant horns and there was no appreciable difference in staining between pregnant and non-pregnant animals. Except for the uterine gland epithelium score, the total PR scores were higher than total ERa scores in all layers of the uterus, cervix, and vagina. Progesterone receptor staining was most intense in the uteri in which columnar epithelium predominated (Fig. 10B). There was little variation in total uterine PR scores throughout the reproductive cycle (Fig. 11A). In the cervical and vaginal epithelium, intensity of ERa staining was higher in the basal and parabasal layers. Intensity of PR varied greatly among layers of the epithelium. Both ERa and PR positive nuclei were randomly scattered throughout the cervical and vaginal stroma. In all time periods except for the fall, ERa scores were higher in the cervical and vaginal stroma, than in the overlying epithelium; however, this difference was not statistically significant (Fig. 9B). The mid-vaginal epithelium had the lowest total ERa scores during all time periods. The lowest ERa scores occurred during embryonic diapause. Except for the middle vagina stroma (P < 0.05) the difference between total ERa scores were not significantly different during different time periods of the reproductive cycle. In regions of stratified squamous cervical and vaginal epithelium, ERa expression was high in the underlying stroma but usually low to moderate in the epithelium (Fig. 12A). Progesterone receptor expression was moderate to high in both the epithelium and underlying stroma (Fig. 12B). Differences in total PR scores were not statistically significant in dif- Fig. 8. Immunohistochemical localization of ER a in the nuclei of ovarian subsurface epithelial structures. Solid arrow ¼ positive staining White arrow ¼ negative staining. Scale bar ¼ 100 lm. ferent layers of the cervix or vagina or during different times of the reproductive cycle (Fig. 11B). Similar to the ERa staining patterns, the intensity of PR staining was often higher in the stroma than in the overlying epithelium. Estrogen receptor score was statistically positively correlated with serum estradiol concentrations for the uterine stroma and myometrium, endocervix epithelium, cervical stroma, and the cranial vagina epithelium and stroma (P < 0.05). There was no statistically significant correlation between ERa score and serum progesterone concentrations. Progesterone receptor scores were not significantly correlated with either serum estradiol or progesterone concentrations. DISCUSSION The primary goals of this study were to better define the normal changes in reproductive tract morphology and sex hormone receptor distribution that occur throughout the yearly reproductive cycle in female California sea lions. Although the number of animals examined was relatively low, general patterns in ovarian, uterine, and cervico-vaginal morphology were noted in the animals examined. In most mammals, the reproductive cycle is FEMALE SEA LION REPRODUCTIVE TRACT 1811 Fig. 9. A, B: Average total immunohistochemistry scores for ER a in the (A) uterus and (B) cervix and vagina. Error bars indicate standard error. divided into a follicular and luteal phase. The follicular phase is short and corresponds to the period from regression of the CL to ovulation, when the primary reproductive hormone is estradiol. The luteal phase corresponds to the presence of ovarian CLs and the primary reproduc- tive hormone is progesterone (Senger, 2005). Sea lions, however, have a complex reproductive cycle that includes a period of embryonic diapause and an extended period in which the ovaries can contain both developing follicles and a CL. 1812 COLEGROVE ET AL. Fig. 10. A, B: Immunohistochemical localization of hormone receptors in glandular epithelial (G) and stoma cell (S) nuclei in the endometrium. (A) Intense nuclear staining of ER a during estrus. (B) High PR expression in the progestational endometrium. Scale bars ¼ 100 lm. Cyclic Histomorphologic Changes Throughout the Reproductive Cycle During the period of pupping and estrus in June and July, ovaries commonly contained developing follicles and the uterine epithelium exhibited characteristics of estrogenic influence. In other species, estrogen concentration peaks just prior to ovulation (Senger, 2005) and in this study two of three animals evaluated had relatively high serum estradiol concentrations in July. Similar peaks in total serum estrogen also were reported in several captive sea lions sampled in July (Greig et al., 2007), further suggesting that an ovulatory peak in estrogen occurs during July. During mid to late July, in several females, a CL in the ovary and morphologic changes in the uterus were consistent with increasing progesterone concentrations. These animals likely had already ovulated prior to death. Similarly, marked progestational changes have been noted in uterine epithelium of fur seals shortly after ovulation (Craig, 1964). Subnuclear vacuoles are considered indicative of increased secretory activity of endometrial glands and, similar to sea lions, have been noted in fur seals and harbor seals starting in estrus and early diapause (Bigg and Fisher, 1974; Boshier, 1981; Roth et al., 1995). Histologic evaluations in fur seals during diapause suggested luteal regression and a corresponding uterine estrogenic appearance (Craig, 1964). There was no morphologic evidence of luteal regression in any of the sea lions during diapause and progestational uterine changes dominated. Similar to the sea lions examined during this study, small follicles were noted throughout diapause in harbor seals, with follicle numbers decreasing by implantation (Bigg and Fisher, 1974). Uterine secretory activity was moderate throughout diapause and in early fall and proteins or molecules elaborated by the uterus are thought to play an important role in control of diapause and blastocyst implantation. Uterine contents have been shown to vary significantly in the periimplantation period in a number of species exhibiting delayed implantation and uterine secretion is thought to activate the blastocyst in fur seals (Daniel, 1971, 1972). Recent molecular studies utilizing differential gene expression and mouse knockout models have shown that sets of genes involved in cell cycle, cell signaling, and energy metabolism regulation are likely involved in blastocyst activation and implantation (Dey et al., 2004; Hamatani et al., 2004). Further analysis of uterine protein content and gene expression studies will likely be needed to completely understand the physiology and regulation of embryonic diapause in pinnipeds. Several studies in northern fur seals found an increase in both estradiol and progesterone prior to implantation (Daniel, 1974). A more recent study found no significant increase in hormones during the peri-implantation period (Browne et al., 2006). Sea lions sampled during September and October, in this study, had relatively low serum estradiol concentrations; however, Greig et al. (2007) found increases in total serum estrogen in several California sea lions sampled in October and November. Frequent sampling of normal sea lions during diapause and implantation is needed to definitively determine if implantation associated hormonal surges occurs in sea lions. As free and conjugated estrone are thought to be the primary circulating estrogen in fur seals (Browne et al., 2006), determination of estrone as well as estradiol concentrations may be helpful in elucidating changes in circulating sex hormones. Pregnancy was not observed in females prior to late November; however, it was impossible to differentiate between animals in which early pregnancy was not detectable and non-pregnant animals. Accordingly, the histologic appearance of the ovaries and uterine epithelium was variable in the fall. The presence of a CL and the predominance of progestational changes in uterine epithelium, however, suggest either pregnancy or the existence of a pseudopregnant state, as has been suggested to occur in other pinnipeds species (Reijnders, 1990; Boyd, 1991; Greig et al., 2007). FEMALE SEA LION REPRODUCTIVE TRACT 1813 Fig. 11. A, B: Average total immunohistochemistry scores for PR in the (A) uterus and (B) cervix and vagina. Error bars indicate standard error. Ovarian Morphology During Pregnancy The finding of concurrent follicular activity and CLs in the ovary of pregnant sea lions is similar to studies in northern fur seals and harbor seals in which folliculogenesis was observed in the ovary contralateral to the ovary containing the CL during pregnancy (Craig, 1964; Bigg and Fisher, 1974). Unlike fur and harbor seals, however, some sea lions had follicular activity in the ovary containing the CL during pregnancy. Waves of follicular activity during pregnancy have also been noted in other pinnipeds species (Harrison, 1969). Follicular development during pregnancy varies among species. In species such as rats and primates, development of large antral follicles is suppressed during the luteal stage by 1814 COLEGROVE ET AL. Fig. 12. A, B: Region of squamous metaplasia in the cervix. Weak immunohistochemical staining in the epithelium and moderate staining in the stroma for (A) ERa and (B) PR. Solid arrows ¼ strong positive staining; Grey arrows ¼ weak positive staining. Scale bars ¼ 100 lm. the negative feedback of progesterone on GnRH secretion by the hypothalamus. Sea lions appear to be similar to cattle and horses, in which, follicle development to near ovulatory size can occur during both the follicular and luteal stages of the cycle (Fortune, 1994). The role of folliculogenesis in estrogen production during pregnancy in sea lions is unknown. Rising levels of estradiol have been noted in other pinniped species during late gestation (Reijnders, 1990; Boyd et al., 1999) and in three pregnant and recently post-parturient sea lions in this study, serum estradiol was as high or higher than the concentrations observed during the estrous period. In fur seals, steroidogenic enzymes have been localized to granulosa and theca cells of preovulatory follicles indicating the capability for estradiol synthesis with follicle development (Browne et al., 2006). Additionally, steroidogenic enzymes have been identified in the CL of fur seals, spotted seals (Phoca largha), ribbon seals (Phoca fasciata), and Stellar sea lions (Eumetopias jubatus) and the placenta of spotted seals, ribbon seals, and Stellar sea lions (Ishinazaka et al., 2001; Ishinazaka et al., 2002; Browne et al., 2006) indicating that the CL and placenta of pinnipeds may produce estrogen during pregnancy. In northern fur seals, regression of the CL begins during mid-gestation and progresses until parturition (Craig, 1964; Boyd et al., 1999). No morphologic evidence of CL regression was found in pregnant sea lions, in this study. The high estrogen receptor scores in pregnant and post parturient sea lions and morphology of the endometrium further suggests some estrogen influence during the later half of pregnancy. Craig (1964) described similar estrogenic features of the fur seal uterus during late gestation. In contrast, ERa expression is low to nondetectable in epithelial cells of the uterus in pregnant dogs and cows (Vermeirsch et al., 2000b; Kimmins and MacLaren, 2001). Interstitial cells are a feature of the ovary in some species and are prominent in the horse and cat (McEntee, 1990). The function of these cells is poorly understood, however, patterns of steroidogenic enzymes expression in cats suggests the cells are differentiated from theca interna and are associated with atretic follicles (Perez et al., 1999). The clustering of interstitial cells around atretic follicles in sea lions supports this theory. The increase in interstitial cell size and cytoplasmic vacuolation noted in this study during estrus and in the winter and spring suggests secretory and potentially hormone-responsive activity. Browne et al. (2006) found high expression of CYP 17 and cytochrome b5 in fur seal interstitial cells and hypothesized that, like some other species that undergo delayed implantation (MondainMonval et al., 1983; Stoufflet et al., 1989), androgens may play an important role in regulating reproductive events in pinnipeds. Future studies should include evaluation of circulating androgen levels, androgen receptor expression, and ovarian steroidogenic enzyme expression to better determine the role of androgens and interstitial cells in sea lion reproductive biology. Estrogen and Progesterone Receptor Expression Similarities between the ERa expression in sea lions and canines included ovarian surface and subsurface epithelial structures expression, clustering of ERa positive uterine surface epithelial cells and a higher expression of ERa in the basal uterine glands (De Cock et al., 1997). In the bitch, uterine ER a expression increases in the late secretory period of the estrous cycle and is highest during proestrous. Receptor levels then decrease during the late proliferative phase of estrus with rising progesterone concentrations (De Cock et al., 1997; Vermeirsch et al., 1999). The highest estrogen receptor scores in sea lions were noted during the pupping and estrus period, which likely encompassed the sea lion proestrus stage. Rising progesterone concentrations decrease estrogen receptor expression in humans and other species with and without high concurrent levels of estradiol (Lessey and Gorell, 1981; Boomsma et al., 1982; Li et al., 1992). Although serum progesterone varied greatly in this study, the high levels of uterine estrogen receptor noted in animals collected during the FEMALE SEA LION REPRODUCTIVE TRACT winter, spring, and estrus suggest that progesterone is lower during these time periods than during diapause and in the fall. Accordingly, Greig et al. (2007) found low levels of progesterone during April and May in both pregnant and nonpregnant sea lions. Progesterone receptor expression remained high in the uterus throughout the reproductive cycle. In most species, PR mRNA levels and protein synthesis are increased by estrogen via ER-mediated gene transcription and decreased by progesterone (Critchley and Healy). The relatively consistent expression of PR throughout the sea lion reproductive cycle is unusual, as cyclic changes in receptor expression similar to that observed with ERa have been noted in other species (Lessey and Gorell, 1981; Watson et al., 1992; Vermeirsch et al., 2000a; Kimmins and MacLaren, 2001). Low sample size may account for absence of statistically significant differences in hormone receptor expression during differing times of the sea lion reproductive cycle. Urogenital carcinoma frequently develops in the cervix and vagina of free-ranging female sea lions and dysplastic lesions are commonly found on histologic evaluation of stranded adult animals (Gulland et al., 1996). In order to better understand the factors involved in cancer development, characterization of the normal morphology, and hormone receptor distribution of the cervix and vagina was essential. The varying types of epithelium noted in the cervix and vagina further suggests that complex factors are likely involved in tumor development. Sea lions appear to be more similar to cows than to canines, in that the cranial portion of the vagina is lined by pseudostratified columnar epithelium. Of particular interest were the two squamous-columnar transition zones noted in the cervix and the middle vagina. In humans, the cervical transformation zone is located where columnar endocervical epithelium transitions to the squamous epithelium of the ectocervix and vagina. This zone shifts at puberty in response to hormonal influences, trauma, and infectious agents. Transformation zone epithelium is more susceptible to human papillomavirus (HPV) infection and estrogen-induced squamous metaplasia that can lead to cervical carcinoma (Elson et al., 2000). High levels of ER and PR are expressed in transformation zone epithelium in humans (Remoue et al., 2003). Both ERa and PR are expressed throughout the sea lion cervix and vagina surrounding these epithelial transition zones, and similar to dogs, there was little variation in receptor expression in the cervix and vagina throughout the cycle (Vermeirsch et al., 2002). Differences between receptor distribution at these transformation zones and other regions of the cervix and vagina were not noted; however, more detailed evaluation of this region in a larger sample population is needed. Infection with a gammaherpesvirus, Otarine Herpesvirus-1, has been associated with sea lion urogenital cancer (Lipscomb et al., 2000; Buckles et al., 2006). Given the susceptibility of the human transformation zone to viral infection, determining the specific regions of the sea lion reproductive tract susceptible to viral infection may be important in determining factors involved in carcinogenesis (Elson et al., 2000). The lower ERa expression in regions of cervical and squamous differentiation was an unexpected finding, as in most species, estradiol mediates epithelial proliferation and squamous differentiation. Progesterone receptor expression remained high in these areas, however, and 1815 both ERa and PR were often highly expressed in the underlying stroma. Recent studies utilizing receptor knockout mouse models have demonstrated that hormone-induced epithelial proliferation can be mediated by stromal receptors in a paracrine manner (Cunha et al., 2004). Similar to previous findings in harbor seals, squamous differentiation was most often noted during late estrus, further suggesting hormone involvement (Bigg and Fisher, 1974). Because of lack of epithelial dysplasia, these areas of squamous differentiation were not regarded as preneoplastic lesions. These squamous changes could potentially lead to cancer development; however, as cervical carcinogenesis in humans can begin as squamous metaplasia of uterine gland epithelium (Elson et al., 2000). The high expression of hormone receptors throughout the reproductive tract stroma and the cyclic variation throughout the cycle, suggests that stromal cells play a key role in mediating epithelial hormone responses in sea lions. All of the sea lions in this study were free-ranging, stranded animals that died or were euthanized during rehabilitation due to severe illness. Several animals (N ¼ 6) were moderately to severely underweight at the time of death. The significance of illness and malnutrition to the data presented in this study is unknown. Significantly lower serum progesterone and estradiol concentrations have been noted in starved rats when compared with nonstarved rats (Baranowska et al., 2001). Lower ERa and PR expression was noted in malnourished ewes when compared with ewes fed maintenance feed requirements (Sosa et al., 2006). The underweight sea lions evaluated had ER a and PR scores similar to animals with normal body condition scores that died during the same time period, indicating that if malnutrition affected hormone receptor expression the effect was likely not detected. Although sample numbers were small and immunohistochemical evaluation was partially subjective, the data presented in this study adds to the general understanding of the reproductive cycle in female California sea lions. Significant changes in uterine morphology were noted to occur throughout the reproductive cycle. Histologic evaluation of the entire reproductive tract should be included in routine postmortem evaluation of deceased individuals to further the understanding of normal sea lion reproductive biology and natural history. Additional characterization of the potential relationship between hormone receptors and cervico-vaginal squamous differentiation is needed to determine whether this change can lead to tumor formation. ACKNOWLEDGMENTS The authors thank Dr. Dennis Wilson and Dr. Chuck Mohr for advice and critical review of the manuscript. The authors are especially thankful to Dr. Alan Conely for his helpful comments on the manuscript and to the histology laboratory at the VMTH. Alejandro Vico, Irwin Liu, and Coralie Munro at the Clinical Endocrinology Laboratory, School of Veterinary Medicine, University of California, Davis, CA are acknowledged for assistance with serum progesterone and estradiol measurements. We thank Denise Greig, Tracey Goldstein, and Elizabeth Wheeler for help in sample organization and to the staff and volunteers of The Marine Mammal Center. 1816 COLEGROVE ET AL. LITERATURE CITED Baranowska B, Chmielowska M, Wolinska-Witort E, Roguski K, Wasilewska-Dziubinska E. 2001. The relationship between neuropeptides and hormones in starvation. Neuro Endocrinol Lett 22:349–355. Bigg M, Fisher HD. 1974. The reproductive cycle of the female harbor seal off southeastern Vancouver Island. In: Harrison RJ, editor. Functional anatomy of marine mammals. London: Academic Press. p 329–347. Boomsma RA, Jaffe RC, Verhage HG. 1982. The uterine progestational response in cats: changes in morphology and progesterone receptors during chronic administration of progesterone to estradiol-primed and nonprimed animals. Biol Reprod 26:511–521. Boshier DP. 1981. Structural changes in the corpus luteum and endometrium of seals before implantation. J Reprod Fertil Suppl 29:143–149. Boyd IL. 1991. Changes in plasma progesterone and prolactin concentrations during the annual cycle and the role of prolactin in the maintenance of lactation and luteal development in the Antarctic fur seal (Arctocephalus gazella). J Reprod Fertil 91: 637–647. Boyd IL, Lockyer C, Marsh HD. 1999. Reproduction in marine mammals. In: Reynolds JE, Rommel SA, editors. Biology of marine mammals. Washington, DC: Smithsonian Institution Press. p 218–286. Browne P, Conley AJ, Spraker T, Ream RR, Lasley BL. 2006. Sex steroid concentrations and localization of steroidogenic enzyme expression in free-ranging female northern fur seals (Callorhinus ursinus). Gen Comp Endocrinol 147:175–183. Buckles EL, Lowenstine LJ, Funke C, Vittore RK, Wong HN, St Leger JA, Greig DJ, Duerr RS, Gulland FM, Stott JL. 2006. Otarine Herpesvirus-1, not papillomavirus, is associated with endemic tumours in California sea lions (Zalophus californianus). J Comp Pathol 135:183–189. Craig AM. 1964. Histology of the reproduction and the estrus cycle in the female fur seal, Callorhinus ursinus. Bull Fish Res Board Can 21:773–811. Critchley H, Healy DL. 1998. Effects of estrogen and progesterone on the endometrium. In: Fraser IS, Jansen, RP, Lobo RA, Whitehead MI, editors. Estrogens and progestogens in clinical practice. New York: Churchhill Livingstone. p 145–171. Cunha GR, Cooke PS, Kurita T. 2004. Role of stromal-epithelial interactions in hormonal responses. Arch Histol Cytol 67: 417–434. D’Haeseleer M, Cornillie P, Simoens P, van den Broeck W. 2006. Localization of oestrogen receptors within various bovine ovarian cell types at different stages of the oestrous cycle. Anat Histol Embryol 35:334–342. D’Haeseleer M, Simoens P, Van den Broeck W. 2007. Cell-specific localization of progesterone receptors in the bovine ovary at different stages of the oestrous cycle. Anim Reprod Sci 98:271–281. Daniel JC, Jr. 1971. Growth of the preimplantation embryo of the northern fur seal and its correlation with changes in uterine protein. Dev Biol 26:316–328. Daniel JC, Jr. 1972. Blastokinin in the northern fur seal. Fertil Steril 23:78–80. Daniel JC, Jr. 1974. Circulating levels of oestradiol-17beta during early pregnancy in the Alaskan fur seal showing an oestrogen surge preceding implantation. J Reprod Fertil 37:425–428. De Cock H, Ducatelle R, Logghe JP. 1997. Immunohistochemical localization of estrogen receptor in the normal canine female genital tract. Domest Anim Endocrinol 14:133–147. Dey SK, Lim H, Das SK, Reese J, Paria BC, Daikoku T, Wang H. 2004. Molecular cues to implantation. Endocr Rev 25:341–373. Elson DA, Riley RR, Lacey A, Thordarson G, Talamantes FJ, Arbeit JM. 2000. Sensitivity of the cervical transformation zone to estrogen-induced squamous carcinogenesis. Cancer Res 60:1267–1275. Fortune JE. 1994. Ovarian follicular growth and development in mammals. Bio Reprod 50:225–232. Greig DJ, Mashburn KL, Rutishauser M, Gulland FMD, Williams TM, Atkinson S. 2007. Seasonal changes in circulating progester- one and estrogen concentrations in the California sea lion (Zalophus californianus). J Mammal 88:67–72. Gulland FM. 1999. Stranded seals: important sentinels. J Am Vet Med Assoc 214:1191–1192. Gulland FM, Trupkiewicz JG, Spraker TR, Lowenstine LJ. 1996. Metastatic carcinoma of probable transitional cell origin in 66 free-living California sea lions (Zalophus californianus), 1979 to 1994. J Wildl Dis 32:250–258. Hamatani T, Daikoku T, Wang H, Matsumoto H, Carter MG, Ko MS, Dey SK. 2004. Global gene expression analysis identifies molecular pathways distinguishing blastocyst dormancy and activation. Proc Natl Acad Sci USA 101:10326–10331. Harrison RJ. 1969. Reproduction and reproductive hormones. In: Andersen HT, editor. The biology of marine mammals. New York: Academic Press. p 253–348. Ishinazaka T, Suzuki M, Yamamoto Y, Isono T, Harada N, Mason JI, Watabe M, Tsunokawa M, Ohtaishi N. 2001. Immunohistochemical localization of steroidogenic enzymes in the corpus luteum and the placenta of the ribbon seal (Phoca fasciata) and steller sea lion (Eumetopias jubatus). J Vet Med Sci 63:955–959. Ishinazaka T, Suzuki M, Mizuno AW, Harda N, Mason JI, Ohtaishi N. 2002. Immunohistochemical localization of steriodogenic enzymes and prolactin receptors in the corpus luteum and placenta of spotted seals (Phoca larghna) during late pregnancy. J Vet Med Sci 64:329–333. Kimmins S, MacLaren LA. 2001. Oestrous cycle and pregnancy effects on the distribution of oestrogen and progesterone receptors in bovine endometrium. Placenta 22:742–748. Lessey BA, Gorell TA. 1981. Hormonal regulation of cytoplasmic estrogen and progesterone receptors in the beagle uterus and oviduct. Mol Cell Endocrinol 21:171–180. Li W, Boomsma RA, Verhage HG. 1992. Immunohistochemical analysis of estrogen and progestin receptors in uteri of steroid-treated and pregnant cats. Bio Reprod 47:1073–1081. Lipscomb TP, Scott DP, Garber RL, Krafft AE, Tsai MM, Lichy JH, Taubenberger JK, Schulman FY, Gulland FM. 2000. Common metastatic carcinoma of California sea lions (Zalophus californianus): evidence of genital origin and association with novel gammaherpesvirus. Vet Pathol 37:609–617. Martin de las Mulas J, Van Niel M, Millan Y, Ordas J, Blankenstein MA, Van Mil F, Misdorp W. 2002. Progesterone receptors in normal, dysplastic and tumourous feline mammary glands. Comparison with oestrogen receptors status. Res Vet Sci 72:153–161. McEntee K. 1990. Reproductive pathology of domestic animals. San Diego, CA: Academic Press. Mondain-Monval M, Bonnin M, Scholler R, Canivenc R. 1983. Plasma androgen patterns during delayed implantation in the European badger (Meles meles L.). Gen Comp Endocrinol 50:67–74. Munro C, Stabenfeldt G. 1984. Development of a microtitre plate enzyme immunoassay for the determination of progesterone. J Endocrinol 101:41–49. Odell DK. 1975. Breeding biology of the California sea lion, Zalophus californianus. Rapports et Proces-Verbaux des Reunions Conseil International pour l’Exploration de la Mer 169:374–378. Oosthuizen WH, Greyling FJ, Bester MN. 1998. Estimation of age from stained sections of canine teeth in the Cape fur seal (Arctocephalus pusillus pusillus). SADJ 53:47–51. Perez JF, Conley AJ, Dieter JA, Sanz-Ortega J, Lasley BL. 1999. Studies on the origin of ovarian interstitial tissue and the incidence of endometrial hyperplasia in domestic and feral cats. Gen Comp Endocrinol 116:10–20. Press MF, Nousek-Goebl NA, Bur M, Greene GL. 1986. Estrogen receptor localization in the female genital tract. Am J Pathol 123:280–292. Reeves RR, Steward BS, Leatherwood S. 1992. The Sierra club handbook of seals and sirenians. San Francisco, CA: Sierra Club Books. Reijnders PJ. 1990. Progesterone and oestradiol-17 beta concentration profiles throughout the reproductive cycles in harbour seals (Phoca vitulina). J Reprod Fertil 90:403–409. Remoue F, Jacobs N, Miot V, Boniver J, Delvenne P. 2003. High intraepithelial expression of estrogen and progesterone receptors FEMALE SEA LION REPRODUCTIVE TRACT in the transformation zone of the uterine cervix. Am J Obstet Gynecol 189:1660–1665. Roth TL, Munson L, Swanson WF, Wildt DE. 1995. Histological characteristics of the uterine endometrium and corpus luteum during early embryogenesis and the relationship to embryonic mortality in the domestic cat. Biol Reprod 53:1012–1021. Senger PL. 2005. Pathways to pregnancy and partuition. 2nd ed. Ames, Iowa: Current Concepts. Shille VM, Lundstrom KE, Stabenfeldt GH. 1979. Follicular function in the domestic cat as determined by estradiol-17 beta concentrations in plasma: relation to estrous behavior and cornification of exfoliated vaginal epithelium. Biol Reprod 21: 953–963. Sosa C, Abecia JA, Forcada F, Vinoles C, Tasende C, Valares JA, Palacin I, Martin GB, Meikle A. 2006. Effect of undernutrition on uterine progesterone and oestrogen receptors and on endocrine profiles during the ovine oestrous cycle. Reprod Fertil Dev 18:447–458. Stoufflet I, Mondain-Monval M, Simon P, Martinet L. 1989. Patterns of plasma progesterone, androgen and oestrogen concentrations and in-vitro ovarian steroidogenesis during embryonic diapause and implantation in the mink (Mustela vison). J Reprod Fertil 87:209–221. 1817 Sukjumlong S, Kaeoket K, Dalin AM, Persson E. 2003. Immunohistochemical studies on oestrogen receptor alpha (ER alpha) and the proliferative marker Ki-67 in the sow uterus at different stages of the oestrous cycle. Reprod Domest Anim 38:5–12. Vermeirsch H, Simoens P, Hellemans A, Coryn M, Lauwers H. 2000a. Immunohistochemical detection of progesterone receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 53:773–788. Vermeirsch H, Simoens P, Lauwers H. 2000b. Immunohistochemical detection of the estrogen receptor-alpha and progesterone receptor in the canine pregnant uterus and placental labyrinth. Anat Rec 260:42–50. Vermeirsch H, Simoens P, Lauwers H, Coryn M. 1999. Immunohistochemical detection of estrogen receptors in the canine uterus and their relation to sex steroid hormone levels. Theriogenology 51:729–743. Vermeirsch H, Van den Broeck W, Simoens P. 2002. Immunolocalization of sex steroid hormone receptors in canine vaginal and vulvar tissue and their relation to sex steroid hormone concentrations. Reprod Fertil Dev 14:251–258. Watson ED, Skolnik SB, Zanecosky HG. 1992. Progesterone and estrogen receptor distribution in the endometrium of the mare. Theriogenology 38:575–580.