Unique expression patterns of the embryonal stem cell marker SOX2 and hormone receptors suggest the existence of a subpopulation of epithelial stem/progenitor cells in porcine and bovine endometrium

Abstract Background There are currently insufficient data on the population of endometrial epithelial stem/progenitor cells in farm animals. Objectives With the aim of identifying a potential population of epithelial stem/progenitor cells in the porcine and bovine endometrium, this study immunohistochemically examined the expression patterns of the oestrogen and progesterone receptors, as well as that of the embryonal stem cell marker SOX2. Methods A total of 24 endometrial tissue samples obtained from cycling pigs (n = 12) and cows (n = 12) were included in our study. Each endometrium was divided into basal, middle and luminal portions. The percentage of marker‐positive cells and the intensity of the immunoreaction in each portion of the endometrium were determined. Results Inverse expression patterns of SOX2 and progesterone receptors were found in both animal species throughout the oestrous cycle. Strong diffuse SOX2 expression was detected in the basal portions of the glands, while a significant decrease in positivity and a weak immunoreaction were found in the luminal two thirds of the glandular epithelium. Strong progesterone receptor expression was observed in at least 90% of glandular cells in the middle and luminal portions, whereas weak staining and significant decrease in positivity were detected in the basal portions of the glands. One oestrogen receptor expression pattern resembled that of progesterone receptors. Conclusion The inverse expression patterns of SOX2 and hormone (especially progesterone) receptors suggest that endometrial epithelial stem/progenitor cells represent a subset of cells that reside in the basal portions of the endometrial glands in both the bovine and porcine endometrium.


INTRODUCTION
Stem cells are undifferentiated cells characterized by the ability to selfrenew and differentiate into multiple cell types (Garget, 2004;He et al., 2009). Several types of stem cells are currently recognized. Totipotent stem cells, which have the highest differentiating potential, have the ability to develop into any cell within the whole organism. By contrast, pluripotent stem cells do not have the ability to form extraembryonic tissue. Multipotent stem cells have a limited differentiation capacity, but can produce cells of a specific cell lineage (e.g., mesenchymal stem cells in the endometrium can differentiate into several types of connective tissue cells). The remaining types are oligopotent stem cells and unipotent stem cells. Oligopotent stem cells retain a relatively broad differentiation capacity; for example, myeloid stem cells can produce erythrocytes, platelets and various white blood cells. By contrast, unipotent stem cells, such as muscle stem cells, have the most limited differentiation potential and can only produce one cell type. Progenitor cells are early progeny of stem cells, but unlike unipotent stem cells, they do not have the ability to self-renew (Mitalipov & Wolf, 2009;Mutalibov & Totipotency, 2009;Trounson, 2006;Ulloa-Montoya et al., 2005).
The enormous regenerative capacity of the human endometrium, together with its bilayer structure, in which the stratum functionalis is sloughed off during the menstrual cycle and is regenerated from the stratum basalis, has prompted researchers to investigate the existence of endometrial stem/progenitor cell population (Padykula, 1991). The endometrium of farm animals, including pigs and cows, undergoes specific morphological changes during the reproductive cycle, known as the oestrous cycle (Noseir, 2003;Soede et al., 2011). Despite many similarities, there are several substantial differences between the menstrual and oestrous cycles (i.e., the endometrium is reabsorbed during the oestrous cycle but shed during the menstrual cycle). In relation to our study, the differences that result from the microscopic structure of the endometrium itself are crucial. Unlike that in animals with oestrous cycles, the endometrium of humans undergoing menstrual cycles is divided into two structurally and functionally distinct layers: the stratum functionalis (upper layer) and the stratum basalis (lower layer).
Both the morphological appearance and the thickness of the upper functional layer differ markedly during specific phases of the menstrual cycle. The lower basal layer abuts the myometrium, has a thickness of 0.5 to 1 mm and contains the bottoms of the uterine glands, capillaries and connective tissue cells, the proliferation of which leads to the restitution of the competent functional layer. Thus, it can be assumed that the persistent glands and connective tissue in the stratum basalis contain subpopulations of both epithelial progenitor cells and multipotent mesenchymal stem cells (Ferenczy, 1976;Salamonsen, 2003;Spencer et al., 2005). However, there are currently very few publications describing the identification of epithelial progenitor cells and their markers in the endometrium of both humans and animals. In particular, the localization of endometrial epithelial stem/progenitor cells in animal species undergoing the oestrous cycle remains unknown.
With the aim of identifying a potential population of endometrial epithelial stem/progenitor cells in the bovine and porcine endometrium, this study examined the expression patterns of the oestrogen receptor alpha (ER) and progesterone receptor (PR), as well as that of the embryonal stem cell marker SOX2 using immunohistochemical staining methods. Regarding the association between hormone receptors and epithelial progenitor cells in the endometrium, a lower content of hormone receptors is thought to indicate a less differentiated cell phenotype, which is a typical feature of progenitor cells (Valentijn et al., 2013). We also examined the expression of the embryonal stem cell marker SOX2, a member of the sex determining region Y (SRY)-related HMG box family of transcriptional factors that plays a key role in mammalian development. SOX2 is essential for maintaining pluripotency in undifferentiated embryonic and neural stem cells and is considered a promising marker in the field of induced pluripotency (Bunina et al., 2020;Holmes et al., 2020).

Animals and tissue specimens
A total of 24 resection specimens consisting of uterine cervix, uterine corpus, uterine horns, fallopian tubes and ovaries were obtained from healthy cycling pigs (n = 12) and cows (n = 12) slaughtered in an animal abattoir. Tissue fragments measuring approximately 2 × 1,5 × 1 cm was dissected from the middle parts of both uterine horns of each sample. In addition, both ovaries were separated from the resection specimens and cut in a longitudinal plane into two equal parts. The obtained tissue fragments were then fixed in 10% neutral buffered formalin for 36 h (1.5 days). Further tissue processing was performed in a standard manner. In brief, the samples were dehydrated by immersion in ethanol solutions of increasing concentrations, and then cleared with xylene and wax infiltration. In the final step, thin tissue sections (3-4 μm) were stained with haematoxylin-eosin.
The phase of the oestrous cycle was determined by gross examination of both ovaries according to the criteria described by Ireland et al. (Ireland et al., 1980) and confirmed microscopically by evaluating the histological appearance of endometrial tissue and folliculogenesis and/or luteogenesis of both ovaries (Ginther et al., 1989 For immunohistochemical analyses, the endometrium was divided into basal, middle and luminal portions. While the thicknesses of the middle and luminal portions varied depending on the presence of mucosal folds, that of the basal portion of the endometrial mucosa was limited to 0.5 mm (similar to the stratum basalis in the human endometrium). As the endometrium of farm animals is not organized into functionalis and basalis layers, we replace the term basalis glandular epithelium with a descriptive term: the basal portion of the endometrial glands.

Immunohistochemistry
Three antibodies were used in immunohistochemical assay, namely

Evaluation of immunostaining
For ER and PR, nuclear staining in the endometrium and myometrium was considered positive. For SOX2, cytoplasmic and nuclear staining in endometrial stroma and cytoplasmic staining in endometrial glands (Perry et al., 2013) were considered positive. ER and PR expression was The assessment of immunohistochemical staining was evaluated independently by two histopathologists (Jiri Lenz and Frantisek Tichy).
Discrepancies were resolved by a consensus.

Statistical analysis
Using statistical analysis, differences in the percentages of markerpositive cells between the basal portions and the luminal two thirds (i.e., middle and luminal portions) of the endometrial glands were determined in each endometrial sample. Differences were compared using the McNemar's test for paired nominal data. α = 0.05 was used as the level of statistical significance in all analyses.

Microscopic findings of porcine and bovine endometrium and ovaries
In all cases, the morphological appearance of the endometrium was For bovine tissue samples, three cases were classified as proestrus, two cases as oestrus, three cases as early metestrus, two cases as mid/late metestrus and two cases as dioestrus.

Analysis of SOX2 expression in porcine and bovine endometrium
Identical SOX2 expression patterns were found in porcine and bovine endometrium regardless of the phase of the oestrous cycle (Tables 1   and 2). In both animal species, SOX2 expression in glands was detected mainly in the basal portion of the endometrial tissue and was char- In the endometrial stroma, strong cytoplasmic and nuclear positivity was found in less than 1% of cells. Distribution of these markerpositive stromal cells was random, and clustering was not apparent.
Most SOX2-positive stromal cells were located just below the surface epithelium ( Figure 4).

Analysis of PR expression in porcine and bovine endometrium
Immunohistochemical staining of PR was similar in both porcine and bovine endometrium regardless of the phase of the oestrous cycle (Tables 1 and 2 (range 60%-80%) in the middle portion and >95% in the luminal portion. Regarding the intensity of the reaction, the only difference (compared with pigs) was the moderate expression in the middle portion of the endometrium observed in two cases (Figure 6a-d).

Analysis of ER expression in bovine endometrium
Three different ER expression patterns were found in bovine endometrial samples. The first pattern was characterized by virtually diffuse nuclear positivity in more than 90% of glandular cells across the endometrium (Tables 1 and 2

Analysis of ER expression in porcine endometrium
ER staining revealed two different expression patterns in porcine endometrial samples (Tables 1 and 2). A gradual slight increase in expression from 70%-80% of glandular cells in the basal portion to approximately 90% of the surface epithelium was observed in two cases classified as proestrus and one case classified as dioestrus.
This expression pattern was characterized by a weak intensity of immunoreaction (Figure 8a Only a few publications have analyzed stem cell markers in the porcine and bovine endometrium using immunohistochemical staining methods. Our current study focused on the pluripotency marker SOX2, which is crucial for the survival and self-renewal of undifferentiated embryonic stem cells. A search of the literature identified three studies reporting SOX2 positivity in the bovine endometrium (Cabezas et al., 2014;Lara et al., 2017;Łupicka et al., 2015) but no studies reporting SOX2 immunohistochemical expression in pigs. The first bovine finding was published by Cabezas et al. in 2014(Cabezas et al., 2014. The basalis glandular epithelium is the postulated site of endometrial epithelial progenitor cells in humans (Garget). Unfortunately, no progress has yet been made in identifying a specific marker of this glandular cell population in humans or animals. Based on the results of our current study, we believe that SOX2 could be a promising marker for identifying basal portions of endometrial glands in pigs and cows.
It is speculated that undifferentiated endometrial stem cells are less sensitive to sex hormones than their terminally differentiated daughter cells due to a lack of hormone receptor expression (Garget et al., 2008).
Regarding the immunohistochemical analysis of hormone receptors in bovine and porcine endometrium, the following results were obtained in our current study. Staining with PR revealed an inverse expression pattern to that of SOX2 in both the bovine and porcine endometrium. In general, the differences in ER expression may be due to technical problems or functional reasons. Given that ER immunohistochemistry was performed under identical methodological procedures, we believe that the differences between the individual ER expression patterns found in our current study were due to functional reasons.
Regarding the bovine endometrium, patterns 1 and 3 were very similar, while in the second pattern, the most striking difference was the decrease in the percentage of ER-positive cells and the intensity of the immunoreaction in the basal endometrial portions. In pigs, the differences between the two expression patterns were relatively discrete. Minor changes in ER expression may reflect slight interindividual differences in hormone receptor status in bovine and porcine endometrium. The age of the animals could theoretically be another reason for the differences between ER expression patterns. There are currently no data comparing the number of endometrial epithelial stem cells in animals of different ages. Thus, the question is whether the proportion of endometrial stem/progenitor cell population, which is characterized by a lesser amount of receptor content, is age-dependent.
In our study, the second ER expression pattern, which was similar to that of the PR, was found mainly in metestrus and dioestrus (corresponding to the luteal phase of the oestrous cycle). In one recent study, the authors reported different hormonal expression patterns in eutopic and ectopic endometrium in humans during the menstrual cycle (Lenz et al., 2021). As in our current study, a significant decrease of ER expression was found in the secretory phase of the menstrual cycle.
Inverse correlation of SOX-2 and hormone (especially progesterone) receptors in both the porcine and bovine endometrium found in our current study could be related to the FOXA1 gene. This gene, also known as hepatocyte nuclear factor 3α, is involved in regulating the embryogenesis of various tissues as well as playing an important role in the post-natal development of hormone-dependent tissues such as prostate or mammary gland (Costa et al., 1989). Recently, attention has been focused on investigating the role of FOX1A gene in the pathogenesis of certain cancer types. In breast cancer, a positive correlation between FOXA1 and ER expression has been reported (Badve et al., 2007). Regarding the association between SOX2 and FOX1A, one recent study found a negative regulation of FOX1A by SOX-2 in human breast and lung cancer (Li et al., 2014). Thus, the question is whether the inverse correlation of SOX-2 and hormone receptors in bovine and porcine endometrium is functionally linked to the FOXA1 gene.
Overall, our study demonstrates an inverse correlation between the expression patterns of SOX2 and hormone receptor expression in both bovine and porcine endometrium. SOX2-positive glandular cells in the basal portions of the endometrium expressed lower levels of hormone receptors (especially PR) than in the middle and luminal portions of the endometrial mucosa. Down-regulation of hormone receptors has been used to indicate a less differentiated cell phenotype. Therefore, our results support the existence of epithelial stem/progenitor cells in the porcine and bovine endometrium, and also suggest their possible localization in the basal portion of the endometrial mucosa. These findings in farm animals are surprising for two reasons. First, the endometrium of pigs and cows is not structurally and functionally divided into the basalis and functionalis, and second, the endometrial tissue is resorbed (not shed) during the oestrous cycle. By contrast, putative epithelial stem cells in the human endometrium are thought to reside in the basalis, allowing the glandular epithelium of the functional layer to be replenished and regenerated during the proliferative phase of the menstrual cycle. This hypothesis is also supported by some immunohistochemical investigations (Fayazi et al., 2016). Thus, from the endometrial epithelial stem cell perspective, the data obtained in our study point to a similarity between the human endometrium and that of pigs and cows.

CONCLUSION
Our data support the presence of two stem cell populations in porcine and bovine endometrium, one of epithelial origin and one of mesenchymal (stromal) origin. As far as we know, our current study is the most thorough investigation of the porcine and bovine endometrium,

CONFLICT OF INTEREST
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.