Maternal Recognition of Pregnancy in the Tasmanian Bettong, Bettongia gaimardi (Marsupialia : Macropodoidea)

Transcription

1 Reprod. Fertil. Dev., 1992, 4, Maternal Recognition of Pregnancy in the Tasmanian Bettong, Bettongia gaimardi (Marsupialia : Macropodoidea) R. W. Rose Department of Zoology, University of Tasmania, GPO Box 252C, Hobart, Tas. 7005, Australia. Abstract The fetus of the marsupial Bettongia gaimardi, the Tasmanian bettong, has both morphological and cytological effects upon the gravid uterus. Development of diapausing embryos or the initiation of oestrous cycles was achieved by removal of the pouch young (RPY). Increases in the diameter of the uterine basal endometrial glands were noted 3 and 4 days later. An animal at Day 5 after RPY had glands in the gravid uterus that were significantly larger than those in the non-gravid uterus (Pe0.01). This difference was also found in non-pregnant animals before Day 11 and was believed to be due to the local presence of the corpus luteum. Thereafter, significant differences were only found in pregnant animals until parturition at Day 17 or 18 after RPY. These differences were attributed to the local effect of the fetus and reflect a 'maternal recognition of pregnancy', as found in other macropodoid species. Extra keywords: embryo, uterine histology. Introduction Macropodoids (kangaroos, wallabies and rat-kangaroos) have paired uteri but are monovular; consequently it is possible to compare gravid with non-gravid uteri under the same peripheral endocrine milieu. Two early papers noted uterine changes associated with the presence of the macropodoid fetus: Owen (1834) recorded that the gravid uterus was thicker than the contra-lateral uterus in the grey kangaroo, Macropus giganteus, and Flynn (1930) demonstrated histological differences between the paired gravid and non-gravid uteri in the Tasmanian bettong, Bettongia cuniculus (=gaimardi). However, Flynn (1930) was unable to breed the bettong in captivity to estimate the age of embryos or stages of pregnancy in bettongs that he obtained; furthermore, Flynn was unware of the phenomenon of embryonic diapause, which is so widespread amongst members of the kangaroo family. The Tasmanian bettong, B. gaimardi, is a relatively small rat-kangaroo found only on the island of Tasmania. The gross anatomy of the reproductive tract has been described by Owen (1834) and Pearson (1945) and is similar to that of B. lesueur described more recently by Tyndale-Biscoe (1968). The reproductive biology of this species conforms to the usual macropodoid reproductive pattern (Rose 1987). The length of gestation is 21-3 days and the length of oestrous cycle is 22.6 days. A post-partum oestrus occurs and the fertilized egg that is produced remains in embryonic diapause for most of the period that the pouch is occupied. When a pouch young is accidentally lost or experimentally removed, the embryo in diapause (often referred to as the delayed blastocyst) resumes development and the young is born within days. Thus it is a relatively simple procedure to obtain accurately staged reproductive material from both pregnant and non-pregnant bettongs. One aim was /92/010035$05.00

2 R. W. Rose to determine whether the gravid uterus was different (in size and/or histological appearance) to a uterus of a similar stage from a non-pregnant animal and, if so, when these differences occurred. The data in this paper allow a chronological order to be placed upon some of the events reported by Flynn (1930) and Kerr (1934, 1936) in their work on the Tasmanian bettong. Maternal recognition of pregnancy (Tyndale-Biscoe 1979), in the form of an altered histological appearance of the gravid uterus in the Tasmanian bettong, Bettongia gaimardi, is demonstrated. Materials and Methods The 18 bettongs used in this study were housed as described in Rose (1982). In order to obtain a graded series of embryonic material, corpora lutea and reproductive tracts, the pouch young were removed (RPY Day 0) so that the delayed blastocyst (if present) would be activated. This technique provided pregnant animals on Days 0, 3, 5, 7, 11, 14 and 17 after RPY and non-pregnant animals on Days 2, 4, 9, 11 and 14 after RPY. In addition, one anoestrous animal without a pouch young was killed. After an intramuscular sedating injection of Ketalar (Ketamine hydrochloride, Parke-Davis, Caringbah, NSW) at 10 mg per kg bodyweight, animals were anaesthetized with a Halothane (ICI Australia, Villawood, NSW) and oxygen mixture. The ovaries and paired uteri were removed and the animals were allowed a post-operative recovery period of several days before being released. The uteri were either flushed with saline in order to collect small blastocysts, or opened and the larger embryos removed. After they had been measured, the embryos and reproductive organs were fixed in buffered formalin at ph 7.0. The ovaries and uteri were embedded in paraffin wax blocks and 8 pm embedded, serial sections were cut and stained with haematoxylin and eosin. In addition, measurements were taken in situ, at laparotomy, of the length and breadth of the uteri of four pregnant animals being operated upon for other experiments at Days 3, 4, 9 and 14 RPY and one Day-14 non-pregnant animal. Measurements The length and breadth of the paired uteri were obtained with a vernier caliper. The greatest diameter of the blastocyst or the yolk sac was measured up to Day 7; after this stage the overall length of the fetus was measured. In order to determine whether the corpus luteum and/or fetus had an effect, measurements were obtained of the diameters of the uterine basal endometrial glands, i.e, the region of the gland closest to the myometrium, and the diameters of the nuclei of the corpus luteum (cl) cells were measured as in Shaw and Rose (1979). These measurements (n = 30) were made with a micrometer eyepiece calibrated against a graduated microscope slide. Measurements of blastocysts were obtained in a similar manner but larger embryos were measured with a vernier caliper. The size of the corpus luteum was estimated by microscopic examination (magnification x40) of the histological slides. In each case the greatest diameter of the gland was measured. The Wilcoxon ranked-sums nonparametric procedure was applied to test for statistical differences between the uterine gland diameters within animals, as in Shaw and Rose (1979). All experimental procedures were approved by The Committee on the Ethical Aspects of Research Involving Animals convened at the University of Tasmania. Results The data were examined for differences between the uteri of individual animals that could be attributed to the local presence of the corpus luteum and also, in the pregnant animals, to determine whether the fetus itself had a local effect that could be distinguished from that of the corpus luteum. Uterine Glands Significant differences in the diameters of the uterine basal endometrial glands first appeared at Day 5 after RPY (PC 0.01) between glands from the two uteri in the one animal, the larger glands appearing in the uterus ipsilateral (i.e. same side) to the CL and embryo (Fig. la).

3 Pregnancy in the Tasmanian Bettong Fig. 1. Changes after removal of pouch young in (a) uterine basal endometrial gland diameters (Et s.e.) in the paired uteri of both pregnant and non-pregnant bettongs (e gravid; W non-gravid; 0 non-pregnant; same side as corpus luteum (CL); 0 non-pregnant, side opposite to CL); (b) mean nuclear diameter (Zks.e.) of the CL (0 nonpregnant, e pregnant); (c) embryonic growth. 161 C :I, I blastocysts Unildminar 0,/, Primlive. streak I Near term fetus Early fetus / Bilaminar blastocyst,,,,,,, Days after RPY The mean diameter of the basal endometrial glands was greatest in gravid uteri at Days 11 and 14 after RPY (and in the glands ipsilateral to the corpus luteum of the non-pregnant Day 9 bettong). At Day 14 after RPY the uterine basal endometrial glands of the non-pregnant animal were small and both uteri appeared similar to each other and to the non-gravid uterus of the Day 14 pregnant animal (Tables 1 and 2). The basal endometrial glands associated with the gravid uterus remained enlarged from Day 5 and throughout gestation and were significantly larger than the non-gravid side (P < for each animal, see Table 1). Uterine Size The size of the uteri varied throughout gestation (Table 2). By Day 4 there had been considerable expansion of the gravid uterus; at Day 5 the non-gravid uterus was enlarged

4 R. W. Rose but smaller than the uterus containing the embryo. From Day 9 onwards there was an obvious difference between the two uteri in pregnant animals although the two uteri were similar in a non-pregnant bettong at Day 14 after RPY. The epithelium of the gravid uterus (Day 11) was significantly thinner (16.5 pm) than that of the non-gravid uterus (23.3 pm, PC 0.01). The gravid uterus was stretched by the growing fetus at Day 14 and more so at Day 17 after RPY (one day pre-partum), so that the uterine epithelium became progressively thinner. At Day 17, the depth of the stretched uterine epithelium measured approximately 15 pm and there was an extensive blood capillary network underneath. Table 1. Diameters of uterine basal endometrial glands and embryonic vesicles at various stages of the reproductive cycle in bettongs *P<O.05, **P<O.01, ***P<O.001, Wilkinson ranked sums Uterine gland diameter ( ~ * m s.e.) Embryo Stage Ipsilateral to Contralateral to size corpus luteum corpus luteum c no estrus^ 53.6k k1.6 - Not pregnant Day k k pm Unilaminar blastocyst Day 2B 67.8k k1.8 - Not pregnant Day k k pm Unilaminar blastocyst Day 4B 86.5k k2.7 - Not pregnant Day k k1.8** 1.1 mm Bilaminar blastocyst Day k k2.1** 2.5 mm Primitive streak Day gb k k 2.3* - Not Pregnant Day llb 84.8k f 4.0* - Not pregnant Day k k 2.8*** 7.5 mm Early fetus Day 14~ 65.4k k1.2 - Not pregnant Day * &1.8*** 11 mm Early fetus Day k &2.1*** 14 mm Near-term fetus A No CL present. Data for right and left side respectively. Non-pregnant animal. Table 2. Uterine length and width at various stages of the reproductive cycle in bettongs Size of uterus (mm) Ipsilateral to Contralateral to corpus luteum corpus luteum noe estrus^ 7.0~ ~5.0 Day 0 7.0x ~5.0 Day 3 9.0x7.0 9~0~6.5 Day x ~6.9 Day ~ ~7.0 Day x ~ 10.0 Day gb 17.0~ ~5.5 Day ~ x5.9 Day llb 20.2~ x7.1 Day ~ ~8.0 Day 14~ 12.0~ x8.0 Day x x7.2 Day ~ ~6.0 A No CL present. Data for right and left side respectively. Non-pregnant animal.

5 Pregnancy in the Tasmanian Bettong Corpus Luteum Two and three days after RPY, the corpus luteum appeared similar to the quiescent gland at Day 0. By Day 4, mitotic figures were not apparent, and the now vascularized CL consisted of larger nucleated rounded cells with prominent nucleoli and a darker staining cytoplasm. There was an increase in the mean nuclear diameter of the cells of the corpus luteum from the quiescent state (Day 0 when the mother was lactating), through the early luteal phase (Fig. lb) until Day 7 after RPY when the levels reached a plateau. Although the 'line of best fit' indicates that the nuclear diameter decreased slightly after Day 11, the actual diameter of the whole gland reached a maximum of mm (n = 2) at Day 14 after RPY, measuring less than 4.0 mm on all other days. Hence, on the basis of these observations, the duration of the luteal phase after removal of the pouch young was approximately 14 days. Embryonic Growth Seven embryos of known age were obtained: two unilaminar blastocysts; one bilaminar blastocyst; an advanced primitive streak stage; two early fetuses; and a near-term fetus (Fig. lc). At quiescence (Day 0) the bastocyst measured 260 pm including the mucoid coat and shell membrane. No expansion had occurred by Day 3 but by Day 5 the blastocyst had increased to 1-1 mm in diameter. Seven days after RPY, a primitive streak stage was well developed, and by Days 11 and 14 after RPY a well formed fetus was present. Discussion This study demonstrates that there is a 'maternal recognition of pregnancy' (as defined by Tyndale-Biscoe 1979) in the Tasmanian bettong. Uterine basal endometrial glands initially increased in size after RPY in both pregnant and non-pregnant bettongs; however, later in pregnancy the glands were larger than at the same time after RPY in non-pregnant animals. This increased size seems to be due to the local presence of the embryo and not the corpus luteum. The histological changes in the reproductive tract of female B. gaimardi were similar to those found during the delayed cycle in other macropodoid marsupials, e.g. Setonix brachyurus (Sharman 1955; Tyndale-Biscoe 1963), Macropus eugenii (Renfree 1972; Renfree and Tyndale-Biscoe 1973) and Potorous tridactylus (Shaw and Rose 1979; Bryant and Rose 1986). The results presented also correspond to previous findings in B. cuniculus ( = gaimardi) (Flynn 1930). Changes from the quiescent state (Day 0) were apparent in the uterine histology at Days 3 and 4 RPY, when the diameters of the uterine basal endometrial glands were approximately 20-30% larger. From Day 5 there may have been local stimulation of the gravid uterus as there were significant differences between the basal gland diameters of the paired uteri of individual animals. Since this effect is seen in both pregnant and nonpregnant animals before Day 14 it must be associated with the local presence of the corpus luteum. Later in pregnancy (from Day 14) these differences are correlated with the presence of the feto-placental unit as they are not found in the non-pregnant animal. Similar differences have been described in M. eugenii (Renfree and Tyndale-Biscoe 1973) and in P. tridactylus (Shaw and Rose 1979) and the term 'maternal recognition of pregnancy' has been used by Tyndale-Biscoe (1979) to describe this phenomenon. The gravid uterus is stretched by the yolk-sac and distended by the growing embryo and this may in part be responsible for the altered histological appearance of the uterus, in particular the thinning of the uterine epithelium. Towards the end of pregnancy the thinning uterine epithelium and the associated increase in the sub-epithelial vascular network may provide a suitable respiratory surface for the developing fetus.

6 R. W. Rose The waxing and waning of the corpus luteum in the bettong is similar to that observed in the potoroo (Shaw and Rose 1979). The duration of the luteal phase obtained in this study (14 days) compare well with the estimate of 15 days (Rose 1987) calculated by subtracting the time to oestrus (after removal of the corpus luteum) from the gestation length, as described by Tyndale-Biscoe et al. (1974). Hinds (1990) reports that the interval between RPY and the progesterone peak is days in the related species, B. penicillata; the same interval has been observed in B. gaimardi (Jones and Rose 1992). Selenka (1892), Flynn (1930) and Kerr (1934, 1936) have described the embryos of Bettongia species at various stages of growth but were unable to accurately estimate the age of their specimens. For example, Selenka estimated that a 2-mm blastocyst was 2 days old, but the present study would suggest that the specimen was closer to 7 days old. Kerr (1934) gives considerable details on the growth of blastocysts and describes (Kerr 1936) how the primordium of the primitive streak was first noted in a 1-32-mm embryo (estimated to be 6 days old from the present study). However, the embryo (maximum diameter 1-1 mm) obtained from the gravid uterus at Day 5 in the present study was at an advanced primitive streak state. As in this study, Flynn (1930) found that there were obvious histological differences between the glands of the two uteri by the time a primitive streak was formed, and that these differences persisted throughout pregnancy. It seems certain that the fetus plays a number of roles during the reproductive cycle of macropodoids. Pregnancy leads to a more rapid return to oestrus than in non-pregnant macropodoids and this 'Merchant effect' has been demonstrated in a number of species including B. gaimardi and the tammar wallaby (Merchant 1979; Rose 1987). Tyndale-Biscoe et al. (1983) have shown that the differences between pregnant and non-pregnant cycles in the tammar reflect real differences in the profiles of progesterone, prolactin, LH and prostaglandin Fz,. Their findings led them to suggest that the conceptus is involved not only in the onset of parturition but also in the timing of post-partum oestrus and ovulation. Rose (1986) has also demonstrated a fetal role in the vacation of the pouch by its older sibling in B. gaimardi. Thus, in the bettong, the fetus affects at least four aspects of maternal reproduction; it results in changes in both morphology (expansion) and cytology in the pregnant uterus that allow gestation to continue past the luteal phase, it influences the duration of the oestrous cycle (Rose 1987) and it has an important role in the control of pouch vacation (Rose 1986). Acknowledgments I wish to thank Dr Roy Swain for his help throughout this project and Alan McFadyen for his surgical skills. References Bryant, S. L., and Rose, R. W. (1986). The growth and role of the corpus luteum throughout delayed gestation in the potoroo, Potorous tridactylus. J. Reprod. Fertil. 76, Flynn, T. T. (1930). The uterine cycle of pregnancy and pseudo-pregnancy as it is in the diprotodont marsupial Bettongia cuniculus. Proc. Linn. Soc. NS W 55, Hinds, L. A. (1990). Control of pregnancy, parturition and luteolysis in marsupials. Reprod. Fertil. Dev. 2, Jones, S. M., and Rose, R. W. (1992). Plasma progesterone levels in the pregnant rat-kangaroo (Bettongia gaimardi). Gen. Comp. Physiol. (in press). Kerr, T. (1934). Notes on the development of the germ-layers in diprotodont marsupials. Quart. J. Microsc. Sci. 77, Kerr, T. (1936). On the primitive streak and associated structures in the marsupial Bettongia cuniculus. Quart. J. Microsc. Sci. 78, Merchant, J. C. (1979). The effect of pregnancy on the interval between one oestrus and the next in the tammar wallaby Macropus eugenii. J. Reprod. Fertil. 56,

7 Pregnancy in the Tasmanian Bettong Owen, R. (1834). On the generation of marsupials with a description of the impregnated uterus of the kangaroo. Phil. Trans. Zoo. Soc. Lond. 1834, Pearson, J. (1945). The female urogenital systems of the Marsupialia with special reference to the vaginal complex. Pap. Proc. R. Soc. Tasm. 1944, Renfree, M. B. (1972). Influence of the embryo on the marsupial uterus. Nature (Lond.) 240, Renfree, M. B., and Tyndale-Biscoe, C. H. (1973). Intra-uterine development after diapause in the marsupial Macropus eugenii. Dev. Biol. 32, Rose, R. W. (1982). Tasmanian bettong Bettongia gaimardi: maintenance and breeding in captivity. In 'The Management of Australian Mammals in Captivity'. (Ed. D. D. Evans.) pp (Zoo. Board Vic.: Melbourne.) Rose, R. W. (1986). Control of pouch vacation in the Tasmanian bettong. Aust. J. Zool. 34, Rose, R. W. (1987). Reproductive biology of the Tasmanian bettong (Bettongia gaimardi:macropodidae). J. Zool. (Lond.) 212, Selenka, E. (1892). 'Studien iiber Entwickelungsgeschichte der Thiere. 5. Beutelfuchs und Kanguruhratte.' (C. W. Kriedel: Wiesbaden.) Sharman, G. B. (1955). Studies on marsupial reproduction Normal and delayed pregnancy in Setonix brachyurus. Aust. J. Zool. 3, Shaw, G., and Rose, R. W. (1979). Delayed gestation in the potoroo Potorous tridactylus (Kerr). Aust. J. Zool. 27, Tyndale-Biscoe, C. H. (1968). Reproduction and post-natal development in the marsupial Bettongia lesueur (Quoy and Gaimard). Aust. J. Zool. 16, Tyndale-Biscoe, C. H. (1979). Hormonal control of embryonic diapause and reactivation in the tammar wallaby. In 'Maternal Recognition of Pregnancy'. Ciba Found. Symp. 64, Tyndale-Biscoe, C. H., Hearn, J. P., and Renfree, M. B. (1974). Control of reproduction in macropodid marsupials. J. Endocrinol. 63, Tyndale-Biscoe, C. H., Hinds, L. A., Horn, C. A., and Jenkin, G. (1983). Hormonal changes at oestrus, parturition and post-partum oestrus in the tammar wallaby (Macropus eugenii). J. Endocrinol. 96, Manuscript received 4 January 1991; revised 16 September; accepted 22 October 1991