Inflammation in the endometrium of the tammar wallaby (Macropus eugenii) at the time of birth.

This research did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

The gravid and nongravid uteri differed in the numbers of all three evaluated types of immune cells ( Table 3 ). Quantification of the immune cells in the gravid and nongravid endometrium. Significance was rejected if P > 0.05. With a mean of 13/mm 2 , there were significantly ( P < 0.005) more myeloperoxidase (MPO)-positive granulocytes in the gravid endometrium than in the nongravid endometrium (4/mm 2 ) ( Table 3 ). The antibody used labeled the MPO within the granules of the cytoplasm. Neutrophils were characterized by a multilobed nucleus and a finely granulated cytoplasm. Some macrophages also showed MPO staining but were distinguished from MPO-positive granulocytes by size and appearance. The macrophages were larger and more elongated, with foamy cytoplasm and an eccentrically located oval nucleus. MPO-positive granulocytes were located within the blood vessels or in the tissue close to the vessels ( Fig. 1 ). In the gravid endometrium, this included the subepithelial capillaries. MPO-positive granulocytes were also present between the cells of the embryonic yolk sac and within the embryonic blood vessels. Light micrographs of MPO staining of the nongravid (A) and gravid (B) endometrium with inserts at higher magnification (1 × 63×, 1 × 100×). In the nongravid endometrium (A), the insert shows MPO-positive granulocytes with strong granular staining and a lobulated nucleus, which clearly distinguishes them from MPO-positive macrophages. MPO-positive macrophages are larger, less intensely stained, and have a single large, oval-shaped nucleus. In the gravid endometrium (B), the insert displays endoderm (End) and trophoblast (Troph) cells with intruding MPO-positive granulocytes. In the bottom insert, four intensely stained MPO-positive granulocytes within a blood vessel are shown. On average, we counted 95 CD3 + T cells/mm 2 in the nongravid and 28 CD3 + T cells/mm 2 in the gravid endometrium, which was a significant difference ( P = 0.01) ( Table 3 ). T cells were characterized by CD3-positive staining of the cytoplasm and cell membrane. They showed a round shape with a large nucleus. In the nongravid endometrium ( Fig. 2A ), the T cells were located within the cell-dense layer underneath the luminal epithelium and in proximity to blood vessels and glands. In the gravid endometrium, CD3 + T cells were overall rather scarce ( Fig. 2B ). Again, CD3 + T cells were found in close proximity to blood vessels or glands. Light micrographs of CD3 + staining of the nongravid (A) and gravid (B) endometrium. In the nongravid endometrium, abundant CD3 + T cells can be identified within the cell-dense layer underneath the luminal endometrium (top insert) and surrounding the blood vessels (bottom insert). In the gravid endometrium (B), few CD3 + T cells were found adjacent to the glandular cross sections and blood vessels (insert). Due to detachment, the embryonic part of the placenta is not visible in this image of the gravid endometrium. Macrophages were significantly ( P = 0.02) more abundant in the nongravid (mean = 247/mm 2 ) than in the gravid (mean = 131/mm 2 ) endometrium ( Table 3 ). Macrophages were initially identified in hematoxylin–eosin-stained samples according to their distinctive appearance as described above. For easier quantification, the samples were stained with Perls’ Prussian blue, which dyed the hemosiderin within the cytoplasm of the macrophages a distinctive blue. In the nongravid endometrium, macrophages were almost exclusively found in close proximity to the glandular cross-sections ( Fig. 3B ). In the gravid endometrium, macrophages were likewise located around the glands ( Fig. 3A ). Light micrographs of Berlin blue staining of the nongravid (A) and gravid (B) endometrium. The overview shows the original image, with the inserts displaying the segmentation (green) conducted by the image analysis software. The trophoblast cells of the BOM were generally large in size and are hereinafter referred to as trophoblast giant cells (TGCs). TGCs exhibited an irregular shape with fringed cell margins. The cytoplasm often appeared vacuolated. The likewise large nuclei of the TGCs seemed to be frayed in some cells. Condensation of chromatin could be observed in most of the TGC nuclei. In most areas, the trophoblast was detached from the maternal luminal epithelium ( Fig. 4 ). (A) Light micrograph of the bilaminar omphalopleure with endoderm (End) and trophoblast giant cells (Troph) in HE stain. (B) Light micrograph of a HE-stained sample of the bilaminar omphalopleure with endoderm and trophoblast giant cells in proximity to the maternal luminal epithelium (Me). The stars indicate vacuolization of the cytoplasm.

Animals used in this study were kept at the research field station of Melbourne University, and samples were collected in the course of another study ( Drews et al. 2013 , Buentjen et al. 2015 ). All experimentation was approved by the University of Melbourne Institutional Animal Ethics Committees and conformed to the National Health and Medical Research Council of Australia (2004) guidelines. The experiment took place during the breeding season and lactational quiescence. The pouch young of the females were removed (day 0 RPY) in order to reactivate the dormant blastocyst from lactational diapause. Pregnancy was diagnosed and further monitored by ultrasonography (Voluson ‘i’, GE, Austria) ( Drews et al. 2013 ). The eight animals in the experimental group were treated with indomethacin to prevent birth, as described previously ( Renfree et al. 1994 ). Briefly, indomethacin powder (Sigma-Aldrich, Australia) was dissolved in 2 mL ethanol and 17 mL peanut oil, resulting in a solution of 60 mg/mL. Indomethacin solution was administered by intramuscular injection from midnight on day 24 RPY every 8 h until fetal death was detected, which resulted in a minimum of four and a maximum of six injections. No female in the experimental group gave birth. By ultrasonography, fetal death was diagnosed after a mean pregnancy length of 26.3 days. This was in line with the expected point of parturition and, therefore, has been considered the ‘calculated time of birth’. Against initial expectations, fetal lifespan could not be elongated by suppressing birth. In only one female, the fetus was still alive on day 27 RPY. Every female was euthanized by intramuscular injection of 60 mg pentobarbitone (60 mg/mL; 3–4 mL/kg) in the lateral tail vein as soon as fetal death was diagnosed. The reproductive tract was removed immediately following euthanasia, and strips of uterine corpus tissue were collected for further processing and analysis ( Table 1 ). The untreated control group comprised eight animals that gave birth normally after a mean pregnancy length of 26.4 days. Collected tissue samples of pregnant and non-pregnant uteri. BOM, bilaminar omphalopleure. The endometrium of the uteri hosting an embryo will hereinafter be referred to as ‘gravid’, while the endometrium of empty uteri will be addressed as ‘nongravid’. After collection, the samples were immediately placed in 4% paraformaldehyde in PBS for 12 h and then washed and dehydrated in 70% ethanol. Samples were embedded in paraffin for light microscopy. The paraffin blocks were cut into sections of 2 μm and placed on color-frosted microscope slides (Jiangsu Huida Medical Instruments Co., Ltd, Yancheng, China) for hematoxylin–eosin (HE) and Perls’ Prussian blue staining, and on Superfrost Plus adhesion microscope slides (Epredia Netherlands B.V., DA Breda, Netherlands) for immunohistochemistry. To identify macrophages in the endometrium based on their hemosiderin content, we used Perls’ Prussian blue staining. For comparison, consecutive sections of the Perls’ Prussian blue-stained slides were stained with filtered Haemalaun Meyer and Phloxin/Eosin solution. To detect T cells, we used a polyclonal rabbit anti-human Cluster of differentiation (CD) 3 primary antibody (41635582, Agilent Technologies Denmark, Denmark), and for neutrophil detection, a polyclonal rabbit anti-human myeloperoxidase (MPO) primary antibody (A0398, Dako Agilent, USA) ( Table 2 ). Antibodies, dilutions, and controls used in this study to label T cells and neutrophils in the uterine tissue of the tammar wallaby. Derived from the archival tissue bank of the Division of Veterinary Anatomy (Vetsuisse Faculty, University of Bern, Bern, Switzerland). Incubation overnight with antibody diluent without primary antibody. After deparaffinization and rehydration, antigen retrieval was performed in 10 mM EDTA buffer at pH 9 (E9) in a PELCO BioWave ® Pro microwave (Ted Pella, Canada), and the slides were transferred into a tris-buffered saline solution (TBS) for washing. For protein blocking, TCGA (9 mL demineralized water, 1 mL tris-buffered saline, 100 μL NaN 3 , 50 mg NaCaseinat, 0.5 mL 5% normal goat serum) was applied for 5 min before another washing step. The primary antibodies diluted in TCGA were added for incubation at 4°C overnight. Primary antibody concentrations and controls are outlined in Table 2 . The next day, after washing off the primary antibody with TBS, the protocol proceeded with 3% HCl in demineralized water to block the peroxidase before flushing with TBS again. A PolyQ Stain two-step detection system goat anti-mouse/rabbit AP kit (44353OT, Quartett, Germany) was then used according to the instructions. After another washing step, the targeted cells were stained with PermaRed/AP (W723-M, Diagnostic BioSystems, USA) and counterstained with hematoxylin. The slides were dehydrated through increasing ethanol concentrations and mounted with CVUltra (Leica Biosystems Richmond Inc., USA). Tile scan light micrographs of the whole tissue sample on each slide were taken with a Zeiss Axio Imager with Apotome (Carl Zeiss AG, Switzerland) with a 40× objective for the assessment of CD3 and MPO expression and a 20× objective for the assessment of macrophages. For every animal, we evaluated three slides for each cell type. For each cell type, the slides to be evaluated were chosen at least 20 μm apart. In total, we analyzed 24 samples per cell type. The analysis was conducted using NIS-Elements AR software (Nikon, Japan). T cells and neutrophils were manually counted in an endometrial area chosen to be as large as possible without artifacts. The mean evaluated area was 2.5 mm 2 (min. 1.09 mm 2 , max. 6.02 mm 2 ) for MPO counts and 2.92 mm 2 (min. 1.11 mm 2 , max. 4.93 mm 2 ) for CD3 counts. Based on that total number, the mean number of cells/mm 2 was calculated for each section. As the macrophages were often clustered and overlapping, they were hard to distinguish manually. Therefore, for better reproducibility, the threshold function of the NIS-Elements software was used for quantification. In the red channel of the image, stained cells were counted if they fulfilled the following criteria: pixel intensity 0–200, cell size 5–45 μm, separation of individual cells within a distance of 0.15 μm, and cell circularity 0.05–1. In the secretion of the endometrial glands of the tammar wallaby, ferric iron accumulates ( Freyer et al. 2007 ), which also reacts with the Perls’ Prussian blue stain. Therefore, we checked the automatic cell count and removed false-positive segmentations within the glandular lumen. Again, this analysis was performed in an intact area as large as possible (mean 4.07 mm 2 , min. 0.93 mm 2 , max. 9.14 mm 2 ). Based on the corrected total cell count, the average number of macrophages/mm 2 was calculated. All statistical analyses were performed in NCSS 2024 Statistical Software (2024). We tested whether there were significant differences between the nongravid ( n = 4) and gravid uteri ( n = 4) with regard to the number of macrophages, CD3 + T cells, and neutrophils, respectively. The Shapiro–Wilk test was conducted to assess normality. At a significance level of 0.05, normality was rejected. Statistical significance was evaluated according to data distribution: normally distributed data (neutrophils and T cells) were analyzed using an independent-samples t -test, whereas non-normally distributed data (macrophages) were analyzed using a two-sample Wilcoxon rank-sum test. A P value below 0.05 was considered statistically significant.

The inflammatory process observed around the time of birth is likely to be less dependent on an adverse reaction toward the foreign fetus but rather triggered by the aging and deteriorating embryonic trophoblast. The marsupial placenta may constitute a primitive form of yolk sac placenta, independent of a hemochorial placental ancestry ( Vogel 2005 ), and immune cell functions might therefore differ from those in eutherian placentation. In our study, prostaglandin synthesis and birth were blocked by indomethacin treatment. However, fetal lifespan could not be elongated (B Drews, BR Menzies, K Roellig, T Hildebrandt, I Buentjen, MB Renfree, personal communication), indicating severe placental insufficiency as a consequence of trophoblast degeneration. Because the tammar wallaby has, compared with other marsupials, a relatively long gestation and clear maternal recognition of pregnancy ( Renfree 1972 ), the initial immune response after loss of the embryonic coats may act as a signal to increase vascularization of the subepithelial capillary network – the maternal component of the yolk sac placenta. Our results suggest that this occurs through inflammation mediated by angiogenic neutrophils. In conclusion, our study emphasizes the role of the marsupial tammar wallaby as a model for unraveling the evolution of the immune system at the embryo–maternal interface and its role in the birthing process.

Our study is the first to describe cells of the innate immune system in the endometrium of the gravid and nongravid uteri of the macropodid tammar wallaby at the calculated time of birth by immunohistochemistry and histological staining. The observed differences in immune cells between the pregnant and nonpregnant endometrium support the hypothesis that the female tammar wallaby is not immunologically ignorant of its developing embryo, as could be suggested by the short pregnancy length. Neutrophils, characterized by the expression of MPO, were significantly more abundant in the gravid endometrium than in the nongravid endometrium. Neutrophils are considered first responders to foreign antigens and tissue damage. Their classical functions are phagocytosis, production of cytotoxic granules, and release of inflammatory mediators ( Riaz & Sohn 2023 ). The role of endometrial neutrophils during pregnancy is not clear. Their numbers are low in the gravid endometrium of mice ( Kaitu’u-Lino et al. 2007 ) and humans ( Salamonsen & Woolley 1999 ). Toward the end of pregnancy, a moderate rise in endometrial neutrophil numbers has been reported. However, a significant increase is only observed postpartum and is associated with endometrial involution ( Shynlova et al. 2013 ). During early pregnancy in the tammar wallaby, maternal capillaries are small and located in the deeper parts of the endometrium, whereas closer to term, the capillary network significantly expands and is located directly underneath the endometrial epithelium ( Freyer et al. 2002 ). In the nonpregnant endometrium, where neutrophil numbers were significantly lower, the subepithelial capillary network was almost invisible. The angiogenic effect of neutrophils has been described in the endometrium during early pregnancy in mice and humans ( Heryanto et al. 2004 , Amsalem et al. 2014 ). Therefore, the angiogenesis in the subepithelial network, which is part of the maternal yolk sac placenta in the tammar wallaby ( Freyer & Renfree 2009 , Muenzenberger et al. 2025 ), is likely to be induced by the neutrophils in the gravid endometrium. Since the tammar wallaby embryo comes into physical contact with the endometrium only later in pregnancy, after loss of the shell coat, the high number of neutrophils might in fact correspond to the inflammatory state observed in eutherians around the time of implantation. This idea is supported by the fact that marsupial attachment is associated with similar gene expression changes to those observed in eutherian implantation ( Dudley et al . 2024 ). In contrast to eutherians, where inflammation is suppressed after implantation and neutrophils do not play a major role ( Chavan et al. 2017 , Zhang et al. 2020 ), the initial inflammation triggered by embryo–maternal contact is only moderately suppressed in the tammar wallaby ( Dudley et al. 2024 ) and might finally contribute to the induction of birth. The abundance of neutrophils at term in the tammar wallaby endometrium might further be triggered by damage-associated molecular patterns (DAMPs) released from the senescent trophoblast ( Muzio et al. 2000 ). In our study, we also observed neutrophils in the yolk sac membrane. In humans, infiltration of the embryonic membrane, in this case the amniochorion, has been associated with both term and preterm labor and may initiate inflammation, remodel uterine tissues, clear degenerating placental cells, and promote the physiological processes required for labor and delivery ( Steel et al. 2005 , Bredeson et al. 2014 ). Indeed, genes associated with neutrophil activation are upregulated at term in the tammar wallaby ( Dudley et al. 2024 ). Therefore, we suggest that neutrophils may also play a role in the induction of birth in the tammar wallaby. However, the definitive role of neutrophils in macropodid pregnancy and delivery requires further study. In our study, CD3 + -positive T cell numbers were significantly lower in the pregnant endometrium than in the nonpregnant endometrium. Since CD3 is a pan-T-cell marker, different subsets of T cells were not distinguished. In humans, total T cell numbers are high in the proliferative phase of the cycle but low during the secretory phase and early pregnancy ( Vassiliadou & Bulmer 1996 ). For maintaining pregnancy in species with invasive placentation, increasing numbers of regulatory T cells play a pivotal role. They are activated by paternal antigens and contribute to immune tolerance by direct killing of cytotoxic CD8 + cells ( Suvas et al. 2003 ) and inhibition of inflammatory cytokine production ( Veldhoen et al. 2006 ). In ruminants and pigs, total T cell numbers significantly increase during implantation, dependent on direct embryo–maternal contact, and then decrease with progressing pregnancy ( Ohta et al. 2013 ). The scarcity of T cells in the marsupial term endometrium might be comparable to the early pregnant eutherian endometrium due to the comparatively shorter marsupial pregnancy length. From our results, it seems likely that T cells do not play a major role in late marsupial pregnancy or birth induction. Macrophages constituted the greatest proportion of leukocytes. Quantification of macrophages was performed via thresholding by image analysis software after Perls’ Prussian blue staining. This method stains ferric iron blue, indicating its storage in the form of hemosiderin in the cytoplasm of macrophages. Hemosiderin-laden macrophages or siderophages are usually observed perivascularly or near sites of hemorrhage. In eutherians, to our knowledge, hemosiderin-laden macrophages in the endometrium have not been described under physiological conditions. In humans, macrophages stained with Perls’ Prussian blue have been associated with endometriosis ( Bedaiwy et al. 2012 ). In the tammar wallaby and the gray short-tailed opossum, positive Perls’ Prussian blue staining of macrophages has been described in the gravid endometrium. It was hypothesized that macrophages take up ferric iron from the fenestrated maternal capillaries and transfer it across to the uterine glands, which secrete it into the uterine lumen, where it is taken up by the yolk sac placenta ( Freyer et al. 2002 ). This possible role of macrophages in iron transport might be supported by our findings, where macrophages were often in close vicinity to uterine glands. However, in our samples, their number was significantly higher in the nongravid than in the gravid endometrium, which suggests additional functions during the estrous cycle. Due to the noninvasive nature of marsupial placentation, there is less tissue turnover and, therefore, less need for the classical phagocytic function. A higher abundance of macrophages in the nongravid uterus of the wallaby suggests a role in preparation for future pregnancy. In our endometrial samples, the giant trophoblast cells of the yolk sac exhibited morphological signs of senescence. Degeneration of the BOM trophoblast cells has likewise been observed in previous studies by Denker & Tyndale-Biscoe (1986) and Jones et al. (2014) . Cellular senescence is a state of terminal proliferative arrest elicited by stress. It is characterized by morphological changes and a shift toward a proinflammatory phenotype ( Cox & Redman 2017 ). We observed perinuclear vacuolization, which is caused by an accumulation of unfolded proteins in the endoplasmic reticulum, eventually leading to increased osmotic pressure and subsequent water influx ( Shubin et al. 2016 ). Fetal membrane senescence has been proposed as a fetal signal with an important role in labor initiation ( Behnia et al. 2015 ). When the placenta can no longer supply sufficient nutrients and oxygen to the embryo, the embryo experiences stress and begins to produce cortisol ( Ingram et al. 1999 ). This hormone signals the mother’s body to release prostaglandins, which play a key role in initiating labor. Locally, trophoblast degeneration results in the formation of DAMPs, which in turn activate endometrial neutrophils and macrophages. Activation of these immune cells leads to the release of prostaglandins, which cause uterine contractions and induction of birth ( Renfree et al. 1994 ). We hypothesize that ancestral intrauterine development is limited by the aging and degenerating trophoblast, which triggers a systemic maternal as well as local endometrial response. Consequently, the embryo is born in an altricial state, and further development into the fetal stage occurs in the pouch. Interpreting our results, one must keep in mind that all animals in this experiment were treated with the nonsteroidal anti-inflammatory drug indomethacin. This inhibits COX-2 synthesis, in turn decreasing prostaglandin synthesis. Prostaglandins play an important role in the inflammatory response, mediating inflammatory responses such as vasodilation, edema, and pain. There have been several studies suggesting that COX-2 inhibitors have an influence on tumor-associated immune cells ( Jin et al. 2023 ), while other studies show no effect of COX-2 inhibitors on immune cell numbers and composition ( Blomgran et al. 2017 ). Further, sex hormones have a strong influence on immune cell numbers in the reproductive tract ( Wira et al. 2015 ). All samples, gravid as well as nongravid endometrium, were treated with indomethacin and evaluated under the same circumstances, reflecting true differences in number and distribution. Therefore, even though a possible influence of the treatment cannot be ruled out, the study is designed to provide a direction and tendency rather than absolute numbers. Furthermore, the fetuses were dead at the time of tissue sampling. This could raise the question of whether the immune cells could have migrated in reaction to the dead fetuses rather than for the induction of birth of the living fetus. We think this to be unlikely since the pregnancy was closely monitored and samples were taken as soon as fetal death was diagnosed, so there was a maximum of a few hours between fetal death and sampling. In one instance, the fetus was still alive at the point of dissection from the reproductive tract, with similar results. Our study only compares the immune cell population of nongravid endometrium versus gravid endometrium at one time point and therefore does not describe the temporal fluctuation of immune cells in the course of pregnancy. Furthermore, the total sample size is small when compared with studies in classical experimental animals such as mice, and the statistical results need to be interpreted with caution. It would be interesting to validate and expand our results in a follow-up study, including a greater number of tissue samples. However, it needs to be considered that samples from undomesticated animals are far more difficult to obtain due to ethical and experimental reasons. In our study, we did not distinguish between further subsets of immune cells, e.g. T helper cells or regulatory T cells and different types of macrophages. For in-depth investigations, IHC protocols for antibodies used in human medicine and in murine models need to be established first. Despite the shortcomings mentioned, we show and quantitatively determine for the first time the leukocyte population in a marsupial, which is a solid basis for further investigations.

With the exception of birds and monotremes, placentation has evolved many times in every vertebrate class ( Vogel 2005 ). In mammals, placentation is thought to have evolved from an egg-laying ancestor. With the transition from oviparity to viviparity, the amount of yolk was reduced, and the eggshell was lost. Subsequently, the embryo was placed into direct physical contact with the maternal tissues, and a placenta developed to meet the nutritional needs of the developing embryo ( Warren et al. 2021 ). Although the intimacy of contact varies depending on the type of placentation, the maternal immune system must be regulated to avoid immunological rejection of the embryo, which expresses paternal antigens ( Medawar 1953 ). In a very simplified model, the first physical contact at implantation is considered a pro-inflammatory event. To facilitate the prolonged pregnancy of eutherians, this initial immune response is followed by an anti-inflammatory phase. A second inflammatory event at the end of gestation is connected to the preparation for labor, finally leading to birth ( Chavan et al. 2017 ). Marsupials provide a valuable opportunity to study immunological embryo–maternal interactions and parturition in a lineage separate from that of eutherian mammals, which diverged roughly 160 mya ( Luo et al. 2011 ). They have a very short gestation and, due to the presence of a marsupial-specific shell coat, an even shorter period of actual placental attachment with physical embryo–maternal contact. These conditions might require different demands on the regulation of the maternal immune system during pregnancy compared with eutherians. The marsupial tammar wallaby ( Macropus eugenii ) is often used for comparative research, as its genetics, reproductive cycle, and development have been extensively studied. Experimental cohorts can be synchronized as the embryo enters diapause once it reaches the blastocyst stage and can be artificially reactivated by removal of the pouch young (day 0 RPY). After RPY, the dormant blastocyst continues development within 3 days ( Gordon et al. 1988 ). In the first two-thirds of gestation, the embryo is surrounded by a mucoid shell coat, preventing cell-to-cell contact with the mother. Following dissolution of the shell coat on day 18 RPY of the 26.5-day-long pregnancy, embryonic and maternal tissues become apposed via a noninvasive epitheliochorial yolk sac placenta ( Denker & Tyndale-Biscoe 1986 , Freyer et al. 2002 ). The embryonic part of the yolk sac placenta comprises the bilaminar (BOM) and trilaminar omphalopleure (TOM). The TOM consists of an endoderm layer toward the embryo, a trophoblast layer toward the mother, and a vascularized mesoderm layer spread out in between. The BOM is composed of only endoderm and trophoblast ( Renfree & Tyndale-Biscoe 1973 ). The maternal component of the placenta encompasses the luminal epithelial layer of the endometrium together with its underlying subepithelial capillary network ( Jones et al. 2014 , Muenzenberger et al. 2025 ). Whereas the morphological features of the term placenta are well characterized, immune cells at the embryo–maternal interface have only been described anecdotally ( Jones et al. 2014 ). From eutherians, we know that the endometrium is populated by a range of immune cells. Their number, function, and localization vary depending on pregnancy status, hormone levels, species, and invasiveness of the placenta ( Vargas et al. 1993 , Vassiliadou & Bulmer 1996 , Tekin & Hansen 2004 , Oliveira et al. 2013 ). For this reason, it is very difficult to disentangle the interplay between the different immune cells and their specific roles in the maintenance of pregnancy and the birthing process. The major immune subsets in the endometrium comprise cells of the innate immune system, such as macrophages, neutrophils, and natural killer cells, and cells of the adaptive immune system, namely T cells ( Weng et al. 2023 ). In our study, we identify and quantify immune cells of the tammar wallaby in the gravid and nongravid endometrium at term pregnancy by immunohistochemistry and histological staining. We morphologically describe the aging embryonic trophoblast cells and discuss our findings with respect to possible implications for the immunological status and induction of birth at the end of pregnancy.

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.

The data that support this study will be shared upon reasonable request to the corresponding author.

MKM and BD conceptualized the study. MKM carried out the preparation, processing, and analysis of the samples and results. MKM and BD drafted the manuscript. BRM critically revised the manuscript and contributed valuable expertise.