Frequency of Different Estrous Stages in Purpose-bred Beagles: A Retrospective Study

Toxicologic Pathology, 36: 944-949, 2008 Copyright 2008 by Society of Toxicologic Pathology ISSN: 0192-6233 print / 1533-1601 online DOI: 10.1177/0192623308326150 Frequency of Different Estrous Stages in Purpose-bred Beagles: A Retrospective Study SUNDEEP A. CHANDRA AND RICK R. ADLER GlaxoSmithKline, Research Triangle Park, North Carolina, USA ABSTRACT The beagle is a monoestric, nonseasonal breeder with a long estrous cycle. Owing to lengthy stages in individual phases of the estrous cycle, limited group size, and typical group assignment focused on homogenized body weight, dogs in the same stage of the cycle can be inadvertently assigned to one treatment group in toxicity studies potentially leading to erroneous interpretation. This study was conducted to better understand the frequency of the different stages of the cycle and review the associated histological features. Histologic sections of reproductive tissues were reviewed from 102 control dogs from thirty-two GLP studies. The average age of dogs at necropsy was 14.38 months, and the mean terminal body weight was 6.87 kg. Based on histological classification, fifty-five dogs were in anestrus, twenty-eight in diestrus, nine in estrus, five in proestrus, and five were classified as immature. Mean ovarian weights were higher in the estrus stage. This review indicates that more than 80% of the dogs in this study were in the anestrus-diestrus stage, and a small percentage of dogs were immature. Interpretation of drug-induced effects on the morphologic changes in the reproductive tract should be performed with due consideration given to the stage of the cycle and the potential for nonuniform assignment to drug treatment groups. Keywords: beagle; canine; reproduction; estrus; immature; uterus; ovary. INTRODUCTION In comparison with other domestic and laboratory species, the female dog has several reproductive features that are unique. The estrous cycle of the bitch is considerably longer than in most other domestic animals. A nonseasonal anestrus of variable duration (two to ten months) follows each estrous cycle (Bouchard et al. 1991; Concannon 1993). There is little evidence of seasonality in most breeds, especially in purposebred beagles kept under controlled environmental conditions. The beagle is essentially monoestric (Anderson and Simpson 1973; Chastain and Ganjam 1986; Concannon 1993). Nonrodent toxicity studies are frequently performed in purpose-bred beagle dogs with a limited (three to five dogs) number of dogs assigned to individual treatment groups by body weight. Unless preliminary evidence of gonadal toxicity is available, in a vast majority of toxicity studies interpretation of drug-induced effects on the female reproductive tract are based solely on assessment of histopathological review of reproductive tissues coupled with organ weight data. Interpretations of drug-induced effects on the estrous cycle or the female reproductive tract become extremely ambiguous owing to small group size, long estrous cycle, absence of pretreatment information on the cycle status, and paucity of estrous-related literature on the female beagle used in toxicity studies. This retrospective study was undertaken to better Address correspondence to: Sundeep A. Chandra, Pathology, Safety Assessment, GlaxoSmithKline, RTP, NC 27709 USA; e-mail: sundeep.a.chandra@gsk.com. Abbreviations: CL, corpora lutea; FSH, follicle-stimulating hormone; LH, luteinizing hormone. understand the frequency distribution of the different phases of the estrous cycle based on the histological features in the reproductive tissues. MATERIAL AND METHODS This retrospective study was conducted using data and archival histology slides from previously conducted (2002 to 2007), good laboratory practice (GLP)-compliant toxicity studies. Data from 102 control (vehicle/water treated) dogs from thirty-two studies were reviewed. Each study usually had three dogs (or five if recovery was an endpoint) in the control group. The in-life portions of all studies were conducted at GlaxoSmithKline in Research Triangle Park, North Carolina, USA, with identical housing and husbandry conditions. The dogs were housed individually, although commingling with dogs in the adjacent cages was permitted during some hours. All studies were conducted in accordance with current guidelines for animal welfare (Guide for the Care and Use of Laboratory Animals 1996; Animal Welfare Act, 1966, as amended in 1970, 1976, 1985, 9 CFR Parts 1 3). Procedures used in these studies were reviewed and approved by the internal Institutional Animal Care and Use Committee. All dogs used in these toxicity studies were supplied by Marshall Farms (New York). At the end of the dosing period (two or four weeks duration), dogs were killed by exsanguination, and protocolrequired tissues were weighed and subsequently fixed in neutral buffered formalin for histopathological evaluation. In the context of this report, ovarian weights were obtained from all dogs (except decedents), and weights of other reproductive tissues (uterus, cervix, and vagina) were not obtained. Ovarian weights relative to the terminal body weight were calculated. 944

Vol. 36, No. 7, 2008 ESTROUS CYCLE IN BEAGLES 945 FIGURE 1. Typical morphological features of the ovary and uterus in anestrus (A), proestrus (P), estrus (E), and early diestrus (D). H&E. Original magnification 12.5X. For this study, hematoxylin and eosin-stained sections of both ovaries, uterine horns, vagina, cervix, and sections of skin containing the mammary gland were examined for histological identification of the phase of the estrous cycle. The slides were reviewed by a single pathologist (Dr. Chandra) to ensure consistency between dogs, and all tissues from a dog were reviewed simultaneously to ensure morphological changes in individual organs were in synchrony with the stage of the cycle. Histological classification of estrus was adapted and modified from published literature (Fowler et al. 1971; Harleman and Foley 2001; Rehm et al. 2007; Van Cruchten et al. 2002). The assigned stage of the cycle is based completely on the histological findings, since the estrous status prior to the animals being euthanized was unknown. Representative photomicrographs depicting the typical stages are shown in Figures 1 and 2. It should be emphasized that since tissue changes occur continuously, the photomicrographs are intended to show one good representative view and therefore represent images from multiple dogs; adjacent images do not reflect concurrent changes in the same dog. The salient features of each cycle used for classification are listed.

946 CHANDRA AND ADLER TOXICOLOGIC PATHOLOGY FIGURE 2. Typical morphological features of the vaginal mucosa and endometrium in anestrus (A), proestrus (P), estrus (E), and early diestrus (D). H&E. Original magnification 200X. Proestrus Moderately large ovarian antral follicles with liquor folliculi are usually present in both ovaries. These follicles are lined by multiple layers of granulosa cells. Small, shrunken (atretic) corporal lutea (CL) with vacuolated cells containing pigment from the previous cycle are generally present. The endometrial stroma is clear and edematous. There is proliferation of superficial and deep endometrial glands. The myometrium is thicker compared to that observed in anestrus, with hypertrophied eosinophilic smooth muscle cells. The vaginal mucosa is lined by five to seven layers of squamous epithelium. Estrus Histological sections of the ovaries are larger than those observed in anestrus and appear cystic to the naked eye. Microscopically, during early estrus, each ovary can have two

Vol. 36, No. 7, 2008 ESTROUS CYCLE IN BEAGLES 947 or three large tertiary follicles lined by stratified layers of elongated granulosa cells protruding into the antral space. In the bitch, luteinization of granulosa cells precedes ovulation, and early CL formation is observed in late estrus. In the late phase of estrus following ovulation, the follicles are lined by thick layers of rounded luteinizing cells amid loose stroma, and the cystic space contains strands of eosinophilic material. The endometrial stroma is thickened and eosinophilic owing to increased stromal collagen. Glandular and myometrial features are similar to those in proestrus. The vaginal mucosa is composed of five to seven layers of squamous epithelium covered by four to six layers of keratin (hyperkeratosis and parakeratosis). Diestrus Histological sections of the ovaries are larger than those observed in anestrus, and large CL can be seen with the naked eye. Each ovary in the early phase has up to two large CLs with closely packed luteal cells containing amphophilic to eosinophilic cytoplasm. In the late phase of diestrus, the luteal cells have rarefied to vacuolated cytoplasm. A striking feature of the uterus at this phase is the thick myometrium composed of hypertrophied smooth muscle cells in addition to a thick endometrium. Initially, the superficial epithelial cells are columnar and eosinophilic, followed by a fine vacuolated appearance in the later part of this phase. The vaginal mucosa is lined by three or four layers of cuboidal epithelium with neutrophils between the cell layers. Vaginal morphology was variable among dogs in the late phase of diestrus. Anestrus (Early and Late) In the early phase, CLs are small and irregular in outline, and luteal cells contain cytoplasmic vacuoles. There is relative prominence of the vasculature (arterioles) and stroma within the CL owing to regression of the luteal cells. In the late phase, CLs are shrunken (atretic) and contain lipofuscin pigment. Early follicular development may be observed simultaneously in the late phase, with many primary and secondary follicles. The uterus is atrophic and characterized by a small cross-shaped lumen, basophilic endometrial stroma, and compact myometrium. The vaginal mucosa is thinner than in diestrus and is generally two cell layers thick, with conspicuous absence of leukocytes. Immature/Noncycling The average age of dogs in this study was 14.38 months. Although by conventional definition immature might imply lack of sexual maturity because of age, in the context of this study it was used to designate dogs that had histomorphological features of immature dogs that had not cycled even once prior to necropsy. Two characteristic histological features in immature dogs were the striking absence of CL or its remnants in the ovary and the absence of glandular tissue in the mammary gland (Harleman and Foley 2001; Rehm et al. 2007). Rudimentary ducts were barely visible in sections of skin containing the nipple. Since histological features in the ovary and uterus of immature dogs have some similarities to those in anestrus, photomicrographs highlighting the differences are shown in Figure 3. RESULTS This retrospective study was conducted using data from 102 control dogs from thirty-two different toxicity studies conducted over a five-year period (2002 to 2007). These dogs were euthanized, and necropsies were conducted in every month of a calendar year with the exceptions of January and June for this data set (this does not reflect standard operating procedures). Roughly twice (twenty-one of thirty-two) the number of studies were completed in the second half of the year, with the overrepresentation mostly within the third quarter (July, August, and September). The average age of dogs at necropsy was 14.38 months (range: 11 to 22.5 months; SD 2.28), and the mean terminal body weight was 6.87 kg (range: 5.05 10.13 kg; SD 0.86). The frequency (absolute incidence and percentage) of the different phases of the estrous cycle and mean ovarian weights (absolute) for the respective cycle are listed in Table 1. As shown in Table 1, a vast majority of dogs (55/102; 53.92%) were in the anestrus phase of the estrous cycle, followed by diestrus with 28 dogs (27.45%) in this phase of the cycle. Group mean (absolute) ovarian weights were the highest in the estrus phase of the cycle. In this population, five dogs from five different toxicity studies were classified as immature/noncycling based on the histomorphology of the reproductive tissues. The mean age of these dogs was 13.6 months (range 12.5 to 15.5 months), and mean body weight was 7.22 kg. The ovaries from four of these dogs had multiple secondary antral follicles, but CL or remnants of CL from a previous cycle were not present. There was no evidence of follicular development in the fifth dog. DISCUSSION The stage of the cycle has a direct bearing on interpretation of toxicologic end points in studies. Because of the small group size (three to five per dose group) and study group assignment directed at homogenizing body weights, dogs in the same stage of the cycle can be inadvertently assigned to one treatment group, potentially leading to erroneous interpretation of druginduced effects on the female reproductive tract. This can be an especially challenging issue when the drug being tested has either direct or indirect influence on the reproductive tract. The preponderance of beagles in the anestrus stage is not unexpected, considering the very long anestrus (up to ten months) in dogs (Bouchard et al. 1991; Concannon 1993; Schaefers-Okkens 2005). Information on the frequency of the different phases of the cycle is lacking for laboratory beagles. In a study evaluating endometrial apoptosis in fifty-eight dogs of various breeds aged between 1 and 8.5 years, eight dogs were observed to be in proestrus, ten in estrus, twenty-four in diestrus (metestrus), and sixteen in anestrus (Van Cruchten et al. 2003). A direct comparison to the data presented in our study is inappropriate, since the age, breed, reproductive status (virgin and parous), and environmental conditions were significantly different. Nevertheless, a common observation between

948 CHANDRA AND ADLER TOXICOLOGIC PATHOLOGY FIGURE 3. Comparison of the typical morphological features observed in the ovary, uterus, vagina, and mammary gland of dogs in anestrus (A) to those noted in immature (I) dogs. Note the lack of corpora lutea in the ovary and mammary glandular tissue in the section of skin containing the nipple on the right compared to old CL in the ovary and quiescent glandular tissue in the anestrus stage on the left. H&E. Ovary and skin, original magnification 12.5X; uterus, original magnification 25X; endometrium and vaginal mucosa, original magnification 200X. TABLE 1. Frequency of the different phases of the estrous cycle in control beagles. Anestrus Proestrus Estrus Diestrus Immature Number of dogs in respective estrous cycle (%) 55 (53.92%) 5 (4.90%) 9 (8.82%) 28 (27.45%) 5 (4.90%) Mean absolute weight (g) of ovaries (SD) 0.77 (0.14) 0.88 (0.08) 1.81 (0.69) 1.17 (0.26) 0.71 (0.22) Mean relative (% of body weight) weight of ovaries (SD) 0.009 (0.0001) 0.012 (0.002) 0.027 (0.003) 0.020 (0.0007) 0.010 (0.002) Average age of dogs at necropsy (months) = 14.38 (range 11 to 22.5 months; SD 2.28). Average body weight of dogs at necropsy (kg) = 6.87 (range 5.05 to 10.13; SD 0.86). the data presented here and those reported by Van Cruchten is the fact that a majority (70% to 80%) of the dogs appeared to be in the diestrus-anestrus phase of the cycle. There are slight differences in the duration of anestrus between dog breeds (Feldman and Nelson 2004; Okkens and Kooistra 2006), but the data presented in our study would be a general reflection of the beagle population used in toxicity studies. Although these findings are for beagles used in toxicity studies at one institution (GlaxoSmithKline) and supplied by a single vendor (Marshall Farms), it is highly unlikely the distribution of the different phases of the estrous cycle will be significantly different for purpose-bred beagles housed indoors under controlled environmental conditions. Anestrus is considered to be a period of ovarian quiescence. In normal control dogs, the precise mechanism by which a new cycle is initiated between periods of anestrus and estrus is still poorly understood (Klein et al. 2003); it is thought that there is a dopaminergic influence involved in this transition. From an experimental perspective, termination of anestrus and initiation of a new cycle can be accomplished by different

Vol. 36, No. 7, 2008 ESTROUS CYCLE IN BEAGLES 949 methods. Estrus can be induced in anestrus bitches by the administration of a variety of hormones, including porcine FSH (Bouchard 1991), equine chorionic gonadotropin (Arnold et al. 1989), LH (Verstegen 1997), and GnRH analogs (Concannon 1989). Administration of the dopamine agonists bromocriptine and cabergoline shortens anestrus and is associated with a decrease in the plasma prolactin concentration, suggesting that the shortened anestrus is the result of suppression of prolactin secretion (Okkens and Kooistra 2006). There is some evidence that factors causing a decrease in opioidergic acitivity promote LH release and the termination of anestrus (Concannon 1993). Similar to anestrus, the length of diestrus can also be shortened by administration of the prostaglandin PGF2a and its analog, leading to regression of CL (Hori et al. 2002). Ovarian weights from this study varied greatly from 0.54 g to 3.46 g and correlate well with the stage of the cycle. The mean absolute ovarian weights were 0.77, 0.88, 1.81, 1.17, and 0.71 g for animals in anestrus, proestrus, estrus, and diestrus. For immature animals, the average ovarian weight was 0.71 g. Although the Society of Toxicologic Pathology does not recommend this procedure in nonrodents (Sellers et al. 2007), the ovarian weight data appear to be useful if the interpretation is performed in conjunction with the stage of the cycle. A few dogs (approximately 5%) were designated as immature/noncycling based on the histomorphological features in the ovary, uterus, vagina, and mammary gland (Harleman and Foley 2001; Rehm 2007). Two striking histological features are the lack of any CL or remnants thereof in the ovary and lack of substantial mammary glandular tissue in the nipple area. Four of these dogs had antral follicles, indicating these dogs might have been coming into their first cycle at the time of necropsy. Group mean body weights are generally uniform prior to the start of toxicity studies, but stage of the cycle for individual dogs or the group is generally unknown. However, if the test article has the potential to impact the estrous cycle, it is prudent to know the stage of the cycle for individual dogs prior to the start of the toxicity study. With some limitations, the stage of the cycle can be determined through a combination of vaginal smears and serum hormone levels prior to initiating toxicity studies (Fowler et al. 1971; Vermeirsch et al. 2001a, 2001b). The long duration of the cycle, monoestric behavior, and variable duration of individual stages can be confounding factors in interpreting xenobiotic-induced effects on the female reproductive tract. In conclusion, one should consider the overall endocrine system and unique reproductive features when interpreting data from toxicity studies. ACKNOWLEDGMENTS The authors acknowledge the generous help of personnel of the anatomic pathology laboratory. We thank Drs. Roger Brown and Sabine Rehm for reviewing the manuscript. REFERENCES Anderson, A. C., and Simpson, M. E. (1973). The ovary and reproductive cycle of the dog (Beagle). Geron-X, Inc., Los Altos, CA. Arnold, S., Arnold, P., Concannon, P.W., Weilenmann, R., Hubler, M., and Casal, M. (1989). Effect of duration of PMSG treatment on induction of oestrous, pregnancy rates and the complications of hyper-oestrogenism in dogs, J Reprod Fertil Suppl 39, 115 22. Bouchard, G., Youngquist, R. S., Vaillancourt, D., Krause, G. F., Guay, P., and Paradis, M. (1991). Seasonality and variability of the interestrous interval in the bitch. Theriogenology 36, 41 50. Bouchard, G., Youngquist, R., Clark, B., Concannon, P., and Braun, W. (1991). Estrous induction in the bitch using a combination diethylstilbestrol and FSH-P. Theriogenology 36, 51 65. Chastain, C.B., and Ganjam, V. K. (1986). Clinical Endocrinology of Companion Animals, pp. 439 85, Lea and Febiger, Philadelphia. Concannon, P.W. (1989). Induction of fertile oestrous in anoestrous dogs by constant infusion of GnRH agonist, J Reprod Fertil Suppl 39, 149 160. Concannon, P. W. (1993). Biology of gonadotrophin secretion in adult and prepubertal female dogs. J Reprod Fertil Suppl 47, 3 27. Feldman, E. C., and Nelson, R. W. (2004). Ovarian cycle and vaginal cytology. In Canine and Feline Endocrinology and Reproduction, 3 rd ed. (E. C. Feldman and R. W. Nelson, eds.), pp. 752-74, Elsevier, St. Louis. Fowler, E. H., Feldman, M. K., and Loeb, W. F. (1971). Comparison of histologic features of ovarian and uterine tissues with vaginal smears of the bitch. Am J Vet Res. 32, 327 34. Harleman, J. H., and Foley, G. L. (2001). Reproductive system. In Pathobiology of the aging dog, Vol. 1 (U. Mohr, W. W. Carlton, D. L. Dungworth, S. A. Benjamin, C. C. Capen, and F. F. Hahn, eds.), pp. 293 309, Iowa State University Press, Ames, IA.. Hori, T., Akikawa, T., Kawakami, E., and Tsutsui, T. (2002). Effects of administration of prostagrandin F(2)(alpha)-analogue fenprostalene on canine corpus luteum and subsequent recurrence of estrus and fecundity. J Vet Med Sci 64, 807 11. Klein, R., Schams, D., Failing, K., and Hoffmann, B. (2003). Investigations on the re-establishment of the positive feedback of oestradiol during anoestrus in the bitch. Reprod Domest Anim 38, 13 20. Okkens, A. C., and Kooistra, H. S. (2006). Anoestrus in the dog: a fascinating story. Reprod Dom Anim 41, 291 96. Rehm, S., Dinesh, J. S., and Williams, A. M. (2007). Estrous cycle-dependent histology and review of sex steroid receptor expression in dog reproductive tissues and mammary gland and associated hormone levels. Birth Defects Res B Dev Reprod Toxicol 80, 233 45. Sellers, R. S., Morton, D., Michael, B., Roome, N., Johnson, J. K., Yano, B. L., Perry, R., Schafer, K. (2007). Society of Toxicologic Pathology position paper: organ weight recommendations for toxicology studies. Toxicol Pathol 35, 751 5. Schaefers-Okkens, A. C. (2005). Ovaries. In: Clinical Endocrinology of Dogs and Cats (A. Rijnberk ed.), pp. 131 56, Kluwer Academic Publishers, Dordrecht, The Netherlands. Van Cruchten, S., Van den Broeck, W., Simoens, P., and Lauwers, H.(2002). Scanning electron microscopic changes of the canine uterine luminal surface during oestrus and late metoestrus. Reprod Domest Anim. 37, 121 26. Van Cruchten, S., Van den Broeck, W., Duchateau, L., and Simoens, P. (2003). Apoptosis in the canine endometrium during the estrous cycle. Theriogenology 60, 1595 608. Vermeirsch, H., Simoens, P., Coryn, M., and Van den Broek, W. (2001a). Immunolocalization of progesterone receptors in the canine ovary and their relation to sex steroid hormone concentrations. Reproduction 122, 73 83. Vermeirsch, H., Simoens, P., Coryn, M., and Van den Broek, W. (2001b). Immunolocalization of androgen receptors in the canine ovary and their relation to sex steroid hormone concentrations. Reproduction 122, 711 21. Verstegen, J., Onclin, K., Silva, L. and Concannon, P. (1997). Termination of obligate anoestrous and induction of ovarian cycles in dogs by administration of purified pig LH. J Reprod Fertil 111, 35 40