Characterization of ovarian progenitor cells for their potential to generate steroidogenic theca cells in vitro

In brief Progenitor cells with ovulation-related tissue repair activity were identified with defined markers (LGR5, EPCR, LY6A, and PDGFRA), but their potentials to form steroidogenic cells were not known. This study shows that the cells can generate progenies with different steroidogenic activities. Abstract Adult mammalian ovaries contain stem/progenitor cells necessary for folliculogenesis and ovulation-related tissue rupture repair. Theca cells are recruited and developed from progenitors during the folliculogenesis. Theca cell progenitors were not well defined. The aim of current study is to compare the potentials of four ovarian progenitors with defined markers (LY6A, EPCR, LGR5, and PDGFRA) to form steroidogenic theca cells in vitro. The location of the progenitors with defined makers was determined by immunohistochemistry and immunofluorescence staining of ovarian sections of adult mice. Different progenitor populations were purified by magnetic-activated cell sorting (MACS) and/or fluorescence-activated cell sorting (FACS) techniques from ovarian cell preparation and were tested for their abilities to generate steroidogenic theca cells in vitro. The cells were differentiated with a medium containing LH, ITS, and DHH agonist for 12 days. The results showed that EPCR+ and LGR5+ cells primarily distributed along the ovarian surface epithelium (OSE), while LY6A+ cells distributed in both the OSE and parenchyma. However, PDGFRA+ cells were exclusively located in interstitial compartment. When the progenitors were purified by these markers and differentiated in vitro, LY6A+ and PDGFRA+ cells formed steroidogenic cells expressing both CYP11A1 and CYP17A1 and primarily producing androgens, showing characteristics of theca-like cells, while LGR5+ cells generated steroidogenic cells devoid of CYP17A1 expression and androgen production, showing a characteristic of progesterone-producing cells (granulosa- or lutea-like cells). In conclusion, progenitors from both OSE and parenchyma of adult mice are capable of generating steroidogenic cells with different steroidogenic capacities, showing a possible lineage preference.


Introduction
The ovary has two major functions: generating reproductive cells (eggs) and producing sex hormones, two essential components for successful reproduction.Ovarian hormone disorders are often accompanied by irregular menstruation, amenorrhea, infertility, and other symptoms that affect not only reproduction but also the general health of the individuals (Miller & Auchus 2011).Three major endocrine cells in the ovary are theca cells, granulosa cells, and luteal cells.Theca cells provide androgen precursors for estrogen synthesis by the granulosa cells, while luteal cells produce progesterone to support pregnancy and fetal growth.All three cells are necessary for maintaining healthy sex steroid hormone profile within the ovary and across the body during different reproductive stages (Kotsuji & Tominaga 1994, Dewailly et al. 2016).Unlike steroidogenic cells in the testes, ovaries undergo reproductive cycles regularly that involve steroidogenic cell regenerations and transformations.The dynamics of theca steroidogenic cells, especially their progenitor cells, have been not well studied.
The ovarian theca layer surrounds the developing follicles and contains various cell types, including steroidogenic, vascular endothelial, immune, and fibroblast-like cells (Yamamoto et al. 1997, Young & McNeilly 2010).Since theca cells are not found in the early follicles and are only observed once a follicle developed with two or more layers of granulosa cells, it suggests that theca cells have to be recruited or newly developed from progenitors during folliculogenesis even in the adult.Isolation and culture of potential steroidogenic theca progenitor cells was carried out for various species, including mouse (Honda et al. 2007), porcine (Lee et al. 2013), sheep (Adib & Valojerdi 2017), monkey (Chen et al. 2021), and human (Dalman et al. 2018, Asiabi et al. 2020).These studies consistently demonstrated that mammalian ovaries of adults contain theca cell progenitors.However, these in vitro studies adopted procedures similar to the culture of general mesenchymal stem cells without prepurification of the cells.With proper inducing conditions, although these cells can express some theca cell-specific markers and produce androgens, they were also frequently found to express other ovarian cell markers, such as granulosa cells and/or oocytes, suggesting the heterogeneity of the cells in culture.
In addition to these studies focused on steroidogenic theca progenitor cells, researchers have also identified a group of general progenitor cells in both the ovarian surface epithelium (OSE) and interstitial compartments for ovulation-related tissue repair and homeostasis maintenance (Gamwell et al. 2012, Ng et al. 2014, Schindler et al. 2018, Wang et al. 2019).During fetal development, progenitors from both the OSE and interstitial perivascular locations contribute cells to granulosa and theca cell populations (Hummitzsch et al. 2015, Li et al. 2022).The OSE is a single layer of poorly differentiated epithelial cells that contains progenitors to repair the ovarian rupture during ovulation (Ng et al. 2014, Schindler et al. 2018, Wang et al. 2019).Pluripotent stem cells were also reported in the OSE that have the potential to form oocytes recently (Sharma & Bhartiya, 2021).The markers used to identify the progenitor populations included LY6A (Gamwell et al. 2012), leucinerich repeat-containing G protein-coupled receptor 5 (LGR5) (Ng et al. 2014, Schindler et al. 2018), and endothelial cell protein C receptor (EPCR, coded by gene Procr) (Wang et al. 2019).However, the relationships between these tissue repair progenitors with defined markers and those steroidogenic progenitors identified by in vitro approaches (Honda et al. 2007, Lee et al. 2013, Adib & Valojerdi 2017, Dalman et al. 2018, Asiabi et al. 2020) are not clear.Also, it is still unknown whether these ovulation-related tissue repair progenitors with defined markers are capable of forming steroidogenic cells.
In the present study, we compared the ability of the three ovarian progenitors with the distinguished markers (Ly6A, LGR5, and EPCR) to generate steroidogenic theca cells in vitro.As a comparison, ovarian cells expressing testicular stem Leydig cell markers platelet-derived growth factor receptor alpha (PDGFRA) and CD51 (Ge et al. 2006, Jiang et al. 2014, Shao et al. 2023) were also tested.

Chemicals and reagents
DMEM/F12 medium, fetal bovine serum (FBS), insulin/ transferrin/selenium (ITS), dexamethasone, and BSA were obtained from Sigma-Aldrich.β-Mercaptoethanol, N2 and B27 supplements, and fibroblast growth factor 2 (FGF2) were purchased from Thermo Fisher.Epidermal growth factor (EGF) was from PeproTech (Rocky Hill, NJ, USA).Chicken embryo extract was from USBiological (Salem, MA, USA).Leukemia inhibitory factor (LIF) was from Millipore.Smoothened agonist (SAG) was purchased from Cayman Chemical.Oncostatin M and platelet-derived growth factor BB (PDGFBB) was from ProSpec (East Brunswick, NJ, USA).Human LH was from MyBioSource (San Diego, CA, USA).Anti-R-phycoerythrin (PE) Magnetic Particles and BD IMag Buffer (10×) were from BD Biosciences (Franklin Lakes, NJ, USA).The manufacturers and the dilutions of the antibodies are summarized in Supplementary Table 1 (see section on supplementary materials given at the end of this article).The primers for qPCR can be found in Supplementary Table 2.The information for ELISA kits for progesterone and estradiol assays can be found in Supplementary Table 3.

Experimental animals
Female C57BL/6 mice were purchased from the Shanghai Laboratory Animal Center (Shanghai, China) at 9 weeks of age.Mice were maintained in the experimental animal https://doi.org/10.1530/REP-23-0407X Wen, J Wang and others center of Wenzhou Medical University under controlled light (14 h light:10 h darkness) and temperature (22°C), with free access to water and mice chow.All animal procedures were performed according to the National Research Council's Guide for the Care and Use of Laboratory Animals, with the protocol approved by the Wenzhou Medical University Animal Care and Use Committee.

Immunohistochemistry and immunofluorescence staining of ovarian sections
For immunohistochemistry staining of ovary sections, the tissues were fixed with 4% paraformaldehyde for 24 h, dehydrated with sucrose (20% for 12 h followed by 30% for 12 h), and finally embedded in OCT at −20°C.Frozen sections (8 μm thick) or isolated/cultured cells were washed with PBS three times and then incubated with a mixture of first antibodies (1:100-1:500) overnight at 4°C, followed by a mixture of fluorescentlabeled secondary antibodies (1:1000) at room temperature for 45 min in darkness.Before examination by microscope, a drop of cover-slide sealing solution containing DAPI was used to reveal the cell nucleus.Negative control was performed by replacing primary antibodies with 1% serum of normal (unimmunized) animals (Supplementary Fig. 1).The specificity of some antibodies used (positive control) was confirmed by staining the testis, uterus, or intestine (Supplementary Figs 2 and 3).The studies that used the antibodies previously were also cited in Supplementary Table 1.
For immunohistochemical staining of ovarian tissue, the labeled dextran polymer method was used.The ovarian sections were treated with 3% hydrogen peroxide to quench endogenous peroxidase activity.After treatment with citrate buffer (pH 6) using microwave heating to retrieve antigens, the sections were incubated with the PDGFRA primary antibody at 4°C overnight, followed by the secondary antibody conjugated to horseradish peroxidase (1:3000, Cat# GAR007, MultiSciences Biotech, China) for 1 h at room temperature.The bound antibodies were then visualized using a diaminobenzidine substrate, and the nuclei were counter-stained with hematoxylin.

Isolation of ovarian progenitor cells
For each ovarian progenitor isolation, 20 randomly cycling adult female mice (40 ovaries) were used.The dissected ovaries were washed with PBS, minced with sterile scissors for 20 s, and digested in 10 mL DMEM/ F12 medium containing type IV collagenase (1 mg/mL) at 37°C for 30 min, with slow shaking (90 cycles/min).In total, 300 animals were used for 15 isolations for the entire study.Digested cells were filtered through a 70-μm nylon mesh, washed, and then purified via magnetic/fluorescence-activated cell sorting (MACS/ FACS) procedures.The progenitor markers used for cell isolation included LY6A, EPCR, LGR5, PDGFRA, and CD51.
For the isolation of progenitors by MACS procedure, the tagged cells were selected by BD IMag protocol according to the manufacturer's instructions.The cell pellets were suspended in cold BD IMag (BI) buffer at a density of 1 × 10 7 cells/mL and stained with PE-conjugated primary antibodies (1:500) at 4°C in darkness for 40 min.After washing and labeling with anti-R-PE magnetic particles (1:100) for 30 min at 4°C, the cells were transferred to a collection tube and immediately placed onto the BD IMag Cell Separation Magnet holder (BD Biosciences) for 8 min.The negative fraction (supernatant) was collected, while the positive fraction (adhered to the wall of the tube) was resuspended with BI buffer.The sorting procedure was repeated two to three times depending on the antibodies used.The final positive and negative cell fractions were combined, and the percentage of positive cells was determined by flow cytometer or cultured in vitro for differentiation.Some of the positive and negative fractions were collected and had their expressions of steroidogenic genes/proteins (Cyp11a1 and Cyp17a1) checked by qPCR and/or immunofluorescence staining.
The medium was changed every other day until cells reached approximately 60% confluence.To examine the cell proliferating activity, cells were cultured for 3 days and then labeled with ethynyl-2′-deoxyuridine (EdU, 2 μM) for 24 h.Incorporated EdU was revealed by Click-iT reaction, and the positive cells were counted and expressed as a percentage of total cells as previously reported (Wang et al. 2019, Shao et al. 2023).
After cells reached about 60% confluence, they were transferred to a differentiation-inducing medium consisting of DMEM/F12, 1× ITS, 0.5 μM SAG, and 2 ng/mL LH and were differentiated for up to 12 days.The medium was changed every other day, and the spent medium was saved for steroid hormone assays by UPLC-MS/MS or ELISA kits (Shao et al. 2023).RNA was isolated from the cells, and the expressions of Cyp11a1, Cyp17a1, Lhcgr, and Fshr were assayed by qPCR.

Immunofluorescent and enzymatic staining of marker proteins
Cells isolated or cultured were washed with PBS three times and then incubated with a mixture of first antibodies (1:100-1:500, see Supplementary Table 1) overnight at 4°C, followed by a mixture of fluorescentlabeled secondary antibodies (1:1000) at room temperature for 45 min in darkness.Before examining by microscope, a drop of cover-slide sealing solution containing DAPI was used to reveal the cell nucleus.
For HSD3B enzymatic staining, freshly isolated cells or cultured cells were allowed to dry on slides for about 10 min.The cells were then stained for 40 min with a solution containing 5β-androstan-3β-ol-17-one steroid substrate (0.4 mM), NAD (1 mg/mL), and tetranitro blue tetrazolium (0.2 mg/mL).After staining, cells were washed with HBSS and fixed with 10% formalin/ HBSS for 5 min.

Steroid hormone assays
Culture medium was collected from four individual experiments.
Androstenedione and testosterone were measured by XEVO TQD triple quadrupole mass spectrometer (MS) (Waters Corp, MA, USA) following Acquity UPLC separation as previously reported (Shao et al. 2023).Androgen-free cell culture media was used for internal control.The androgen-d3 stock solutions were used to create the internal standard (IS) working solution.The collected cell medium (50 µL) and 100 µL acetonitrile were combined with the IS working solution (5 µL).After vibration for 3 min, the mixture was centrifuged at 12,000 g for 15 min.The supernatant (10 µL) was loaded into the system through a selfsampling protocol.
Progesterone and estradiol were assayed by ELISA kits (Elabscience Biotech, Shanghai, China).Fifty microliters of culture medium were used for the assay.For experimental accuracy, all samples are assayed in duplicates, with an intra-assay coefficient of variation less than 5%.The detailed information about the progesterone and estradiol ELISA kits can be found in Supplementary Table 3.

RNA extraction and real-time qPCR analysis
After cell sorting or immediately after culture, total RNA was extracted by RNeasy Mini Kit (Qiagen) according to the manufacturer's instructions.Extracted RNA was quantified by NanoDrop 2000 Spectrophotometer (Thermo Fisher) and reverse-transcripted to cDNA using the iScript cDNA synthesis kit (Bio-Rad).The qPCR amplification was carried out with a SYBR Green PCR Master Mix Kit, using a protocol consisting of 95°C for 5 min, followed by 40 cycles of 95°C (10 s), and 60°C (30 s).The total 15 μL reaction volume contained 7.5 μL SYBR Green mix, 1.5 μL forward and reverse primer mix, and 0.02 μg diluted cDNA.A universal expression gene ribosomal protein S16 (Rps16) was used as an internal control.The expression levels were calculated using the Delta-Ct method and adjusted to Rps16.The primer sequences can be found in Supplementary Table 2.

Statistical analysis
Student t-test was applied to detect a significant differences for experiments involving two groups.Oneway analysis of variance (ANOVA) was used for multiple group comparisons.Differences between groups were considered significant if P < 0.05.The significant differences among individual groups were determined by the Student-Newman-Keuls test, using SPSS (IBM) statistical software package.Significances with different confidence levels were defined at P < 0.05, 0.01, 0.001, or 0.0001.

Distribution of progenitor cells in adult mouse ovaries
Before establishing the differentiating potential of various ovarian progenitors, we first determined their locations in the adult mouse ovary.Ovarian sections were stained with antibody against the three progenitor cell markers, LGR5 (Ng et al. 2014, Schindler et al. 2018), EPCR (Wang et al. 2019), or LY6A (Gamwell et al. 2012).For comparison, we also included two wellestablished testicular stem Leydig cell markers, PDGFRA (Ge et al. 2006) and CD51 (Jiang et al. 2014, Shao et al. 2023) (Fig. 1).As reported previously, LY6A, EPCR, and LGR5 were all expressed by cells primarily associated with the OSE (red arrow), with very few (EPCR) or none (LGR5) positive cells being detected in the interstitial compartment.However, significant numbers of LY6A+ cells were found in the interstitial compartment (yellow arrow), including a few in the theca layer (white arrow).PDGFRA+ cells were exclusively detected in the interstitial compartment (yellow arrow) without any in the OSE (red arrowhead) or theca layer (white arrowhead).CD51, on the other hand, labeled almost all cell types across the whole interstitial compartment, including a few inside follicles (green arrow).When the primary antibodies were replaced by 1% unimmunized animal sera, no staining was detected (Supplementary Fig. 1), confirming the specificities of the staining.

Isolation and expansion of progenitor cells in vitro
To compare the potential of these progenitors in generating steroidogenic cells, we isolated the https://doi.org/10.1530/REP-23-0407X Wen, J Wang and others positive cells specifically expressing each marker.The fluorescence-labeled or magnetic particle-tagged cells were purified through MACS and/or FACS procedures (Fig. 2, Supplementary Figs 4, 5, and 6).Taking LY6A as an example, the raw cell suspension before antibody labeling formed a cluster below the diagonal line (Fig. 2A and Supplementary Figure 4; LY6A untagged).For cells labeled with PE-LY6A antibody, about 7.1% were tagged and moved above the diagonal line (Fig. 2A and B, LY6A-tagged).After the FACS procedure or four rounds of MACS purifications, the cell purity reached above 99% (Fig. 2A and B and Supplementary Fig. 4; LY6A purified).Similarly, CD51+, PDGFRA+, EPCR+, and LGR5+ cells were all enriched significantly to about 70-96% (Fig. 2A and B, Supplementary Figs 5 and 6).
To further examine whether the cell preparations isolated by each marker were contaminated with steroidogenic cells, the unsorted, positively and negatively sorted cells were compared for their expressions of the key steroidogenic genes Cyp11a1 (a marker for all ovarian steroidogenic cells) and Cyp17a1 (a marker for theca steroidogenic cells only) (Fig. 2C).Both Cyp11a1 and Cyp17a1 mRNA levels were significantly enriched in LY6A− preparation, and the levels became undetectable in the LY6A+ fraction, suggesting successful exclusion of all steroidogenic cells by LY6A marker (Fig. 2C).Similar enrichments were also achieved for LGR5+ and PDGFRA+ cells.Interestingly, the expressions of the two marker genes were also enriched by CD51, but in exactly the opposite direction to those of LY6A (Fig. 2C).The CD51 positive, but not negative, fraction was enriched for steroidogenic cells.Similarly, for an unknown reason, EPCR also failed to exclude either of the two steroidogenic genes examined, damping the usefulness of CD51 and EPCR in isolating progenitors without contamination of steroidogenic cells.
Since LY6A and CD51 behaved in an opposing manner, differences between the cells isolated by the two markers were further compared for their expressions of steroidogenic enzymes HSD3B and CYP11A1 (Fig. 3).Consistent with the qPCR results, LY6A−, but not LY6A+ cells, expressed both HSD3B and CYP11A1, while the opposite was true for CD51 cells.

Differentiation of progenitor cells into steroidogenic cells in vitro
To further characterize the differentiating potential of LY6A+ cells, we cultured both LY6A+ and LY6A− cells in vitro and assessed their abilities to proliferate and to differentiate into steroidogenic cells.The proliferative activity was assayed by incubating cells with EdU for 24 h during the expanding period.About 87% of LY6A+ cells were labeled with EdU, while only 14% of LY6A− cells were positive for EdU (Fig. 4A).
To determine the potential of LY6A+ cells to form steroidogenic theca cells, a protocol for differentiating testicular stem Leydig cells was adopted (Li et al. 2016, Shao et al. 2023), with a differentiation-inducing medium containing LH (2 ng/mL), SAG (a DHH agonist, 0.25 μM), and 1× ITS.During the first 4 days of differentiation, no testosterone or androstenedione was detected in the culture medium (Fig. 4B).However, both androgens became detectable by day 6, and the concentrations kept increasing through day 10.To confirm the theca cell identity, expressions of the key steroidogenic genes (Cyp11a1 and Cyp17a1) were assayed for the cells before and after their differentiation (Fig. 4C).No expression was detected for the two genes before the differentiation, but significant levels were detected after the differentiation.Also, the cells were negative for CYP11A1 and HSD3B before the differentiation (d0, Fig. 4D) and became positive for the two enzymes after differentiation for 10 or 12 days (d10 or d12, Fig. 4D).

Difference in the progenies generated by progenitors in vitro
Since LY6A+ cells were distributed in both the OSE and parenchyma, the LY6A+ cells that gave rise to steroidogenic cells could come from one or both locations.To further confirm the differences between the progenitors of the two locations, cells isolated by LGR5 (exclusively OSE-associated) and PDGFRA (exclusively parenchyma-associated) were compared for their abilities to generate steroidogenic cells (Fig. 5).The isolated LGR5+ cells and PDFRA+ cells did not express either CYP11A1 or CYP17A1 (Fig. 5, d0).After differentiation in vitro for 10 days, the majority of LGR5+ cells gained the ability to express CYP11A1 (Fig. 5, d10), while none of the cells expressed CYP17A1.However, after similar differentiation in vitro for 10 days, PDGFRA+ cells were able to generate cells expressing both CYP11A1 and CYP17A1 (Fig. 5, d10), suggesting their theca-like steroidogenic cell characteristics.
To further confirm the functional differences of the cells generated by different progenitors, steroid hormones were assayed (Fig. 6A).Cells generated from LGR5+ progenitors produced only progesterone but not testosterone, while cells generated by LY6A+ or PDGFRA+ progenitors were able to produce both progesterone and testosterone.Estradiol was also assayed but was undetectable in all the cells tested.To examine whether there was a difference in the ability of cells to respond to pituitary hormones, Lhcgr and Fshr were assayed (Fig. 6B).Unexpectedly, cells derived from LGR5+ progenitors expressed Lhcgr but not Fshr, while cells from LY6A+ or PDGFRA+ progenitors expressed both genes (Fig. 6B).

Discussion
Ovary somatic progenitor cells have been studied over the years for their abilities to generate functional theca steroidogenic cells.Putative theca stem cells have been reported from various species based on their abilities to give rise to steroidogenic theca-like cells in vitro (Honda et al. 2007, Lee et al. 2013, Adib & Valojerdi 2017, Dalman et al. 2018, Asiabi et al. 2020, Chen et al. 2021).However, except for the few studies that used the theca layer to isolate cells, the true identities and locations of these cells in other studies are difficult to determine since the studies cultured cells without prepurification.In addition to these studies focused on theca precursors, progenitors involving ovulation-related tissue rupture repairs were also identified along the OSE or interstitial compartment with defined markers (LGR5, EPCA, LY6A, and PDGFRA) (Gamwell et al. 2012, Ng et al. 2014, Schindler et al. 2018, Wang et al. 2019).However, the abilities of these progenitors with defined markers to form steroidogenic cells were not known.In the present study, we tested a hypothesis that in addition to their tissue-repairing activities, these progenitors with defined markers may also play roles in forming steroidogenic theca cells.The results confirmed our hypothesis that the progenitors expressing LGR5, LY6A, or PDGFRA were all able to form steroidogenic cells in During every ovulation, the OSE and the underlying tunica albuginea are damaged.The progenitor cells in the remaining OSE start to proliferate and repair the damaged surface (Flesken-Nikitin et al. 2013).Various markers (LGR5, EPCA, and LY6A) were identified for OSE progenitors involved in ovulatory rupture repairs (Gamwell et al. 2012, Ng et al. 2014, Rastetter et al. 2014, Schindler et al. 2018, Wang et al. 2019).LGR5+ cells were found to be broadly expressed during ovary organogenesis but limited to the OSE in neonatal life (Ng et al. 2014, Schindler et al. 2018).In adults, the expression of Lgr5 is predominantly restricted to proliferative regions of the OSE.Using in vivo lineage tracing tools, the Lgr5+ population was demonstrated to play important roles in OSE homeostasis maintenance and ovulatory regenerative repairs (Ng et al. 2014, Schindler et al. 2018).In the present study, we not only confirmed the specific OSE distributions of the LGR5+ population but also demonstrated their activities to form progesteroneproducing cells, suggesting their potential contribution to progesterone-producing cells in the adult ovary.
In addition to LGR5, EPCA is another protein identified recently that can specifically label OSE cells with ovulatory rupture repair activity (Wang et al. 2019).EPCA and LGR5 may not mark the same population along the OSE because EPCA cells had distinct lineage-tracing behavior in OSE homeostasis maintenance (Wang et al. 2019).In the current study, we also confirmed that the EPCA population is distributed along the OSE.Since the antibody used for cell isolation was not able to exclude the steroidogenic cells from the EPCA+ population, the differential activity of progenitors was not tested in the current study.
LY6A is a glycosylphosphatidylinositol-linked cell surface protein known to be capable of enriching Differentiation of LGR5+ and PDGFRA+ ovarian progenitor cells by steroidogenic cell-inducing medium in vitro for 10 days.Expressions of CYP11A1 and CYP17A1 were compared between the two cell types before (d0) and after differentiation for 10 days (d10).Both cells did not express any of the two steroidogenic proteins before the differentiation but expressed CYP11A1 after differentiation (d10).However, LGR5+ cells did not express CYP17A1 while PDGFRA+ cells did after differentiation for 10 days.Scale bars represent 10 μm in length.Reproduction (2024) 168 e230407 https://doi.org/10.1530/REP-23-0407X Wen, J Wang and others somatic stem cells in other tissues (Spangrude et al. 1988, Welm et al. 2002, Xin et al. 2005).The LY6A+ population was also isolated from the mouse ovary and expanded in vitro (Gamwell et al. 2012).The cells showed stem/ progenitor properties with extensive cell proliferation potential and sphere formation capacity.Unlike LGR5 or EPCA, LY6A only labeled a very rare OSE population, since the LY6A+ cells only represented 2% of the total OSE cells (Gamwell et al. 2012).The current study confirmed that the LY6A+ cells were distributed primarily in the interstitial compartment instead of the OSE.PDGFRA, however, is a receptor expressed by mesenchymal cells to respond to PDGF signaling from the epithelium compartment in many tissues.The protein was found useful in labeling and isolating stem Leydig cells from the testes (Ge et al. 2006).In the current study, we confirmed its interstitial location in the adult ovary and further demonstrated that the positive cells can be differentiated into theca-like cells in vitro.
Since LY6A+ cells are distributed in both the OSE and parenchyma, the isolated progenitors based on the marker could originate from one or both locations.To further study the potential difference in the cells between the two locations, we compared LGR5+ OSE progenitors and PDGFRA+ parenchymal progenitors for their potential to form different steroidogenic cells in vitro.The results showed that while both LGR5+ and PDGFRA+ progenitors formed steroidogenic cells, the progenies generated were different.Although the cells generated by progenitors from both locations expressed CYP11A1 and produced progesterone, only cells derived from PDGFRA+ progenitors expressed CYP17A1 and produced androgens, indicating that OSE progenitors may not be able to generate theca-like cells as interstitial PDGFRA+ progenitors do.These results suggest a potential lineage preference between the cells from the two locations.This is consistent with early observations that during fetal development, progenitors from the interstitial perivascular location can contribute cells to both granulosa and theca populations (Hummitzsch et al. 2015), while progenitors associated with the OSE only generate granulosa cells (Rastetter et al. 2014, Li et al. 2022).This preference apparently is maintained up to adulthood.
Interestingly, there is an important difference between the theca-like cells generated in the current study and the real theca cells in vivo.Theca cells express LHCGR and respond to LH, but not to FSH (Sharma & Bhartiya, 2021).However, the theca-like cells generated in the present study expressed both Lhcgr and Fshr.The reason for this inconsistency is unknown but may well be related to the simple inducing medium used.The differentiation medium used in the current study was based on previous observations that LH and DHH signaling were enough to induce stem Leydig cells to differentiate (Chen et al. 2019, Shao et al. 2023).It is well known that theca and Leydig cells share many characteristics, including origins, major products produced (androgens and INSL3), and responses to LH but not to FSH.More importantly, their development depends on a common hedgehog signaling (Yao et al. 2002, Wijgerde et al. 2005, Spicer et al. 2009, Liu et al. 2015).The present results indicated that a combination of LH and DHH was also effective in inducing ovarian progenitors into theca-like cells.The differentiated cells of interstitial origin produced androgens but not estrogen, confirming their theca cell identity.However, the unexpected expression of Fshr suggests that the simple formula may not be good enough to induce 'perfect' theca cells in vitro.Improvement of the medium would be needed in the future.
Isolation and culture of theca progenitor cells was done in different species before (Honda et al. 2007, Lee et al. 2013, Adib & Valojerdi 2017, Dalman et al. 2018, Asiabi et al. 2020, Chen et al. 2021).Cells cultured by these studies were not pre-purified, so the in vivo identity of the cells was not always clear.Characterization of the cells, after they were stably established in vitro, showed that the cells shared general mesenchymal stem cell properties most of the times.Differentiations to theca lineage were carried out by complex formulas containing three to nine growth factors in addition to ITS, LH, and/or FSH.The growth factors used included IGF1, SCF, BMP6, TGFB, HGF, KGF, GDF9, EGF, FGF2, LIF, GNDF, and PDGF.Although androgen-producing cells were differentiated, estradiol was also detected in the culture media or was not tested at all, so the true theca cell identity was not always proved.Since these growth factors were shown to play roles in theca cell development, they become the best candidates to be tested with the present differentiation medium.Similarly, the cells established by these procedures also deserve to be tested by the inducing medium established in the present study.
In summary, we have examined the location of ovarian progenitors using four previously reported markers and also have determined their potential to form steroidogenic progenies in vitro.LGR5-marked progenitors were exclusively located along the OSE, while PDGFRA-labeled progenitors were found to be located in the interstitial compartment.LY6A-labeled cells, on the other hand, were found in both locations.In vitro, LGR5+ OSE progenitors formed CYP11A1+ steroidogenic cells devoid of CYP17A1 expression and testosterone production, while PDGFRA+ interstitial progenitors generated steroidogenic cells expressing both CYP11A1 and CYP17A1 and produced androgens, suggesting a lineage preference of the progenitors from the two locations in forming steroidogenic progenies (progesterone-vs androgen-producing cells).Future studies are needed to confirm such activities in vivo.

Figure 1
Figure 1 Immunofluorescent staining of five potential stem cell markers.LY6A and PDGFRA primarily labeled ovarian interstitial cells without any staining of granulosa cells, while EPCR and LGR5 primarily labeled OSE cells.LY6A also labeled OSE cells.CD51, however, stained almost all of the interstitial cells and also a few granulosa cells in the follicles.Red arrow: positive OSE cells; red arrowhead: negative OSE cells; white arrowhead: positive theca cells; white arrowhead: negative theca cells; yellow arrow: positive interstitial cells; and green arrow: positive granulosa cells.Scale bars represent 50 μm in length.

Figure 2
Figure 2Purification of LY6A and CD51 positive cells by FACS/MACS procedures.(A) Ovarian cells with LY6A-PE antibody staining were isolated by MACS and analyzed by flow cytometer.LY6A+ cells were purified to above 99%, while CD51+ cells were purified to above 95%.Positive cells were within the blue (LY6A) or red (CD51) boxes.LY6A untagged: cells before LY6A PE-antibody staining; LY6A-tagged: cells after LY6A PE-antibody staining but before isolation; LY6A+: LY6A+ cells isolated by MACS procedure; CD51+: CD51+ cells isolated by MACS procedure.(B) PE-tagged LY6A+ and CD51+ cells before and after purification.(C) qPCR analysis of the two steroidogenic genes for ovarian cells before and after enrichments by LY6A or CD51.Data are expressed as mean ± SE of three to four individual experiments.*, **, ***, ****: significantly different from the positive cells at P < 0.05, 0.01, 0.001, and 0.0001, respectively.

Figure 3
Figure 3Expression of steroidogenic markers (HSD3B and CYP11A1) by the cells isolated with LY6A and CD51 markers.LY6A− and CD51+ cells exclusively expressed the two steroidogenic proteins, while LY6A+ and CD51− cells did not express the two proteins.Scale bars represent 50 μm in length.

Figure 4
Figure 4 Proliferation and differentiation of LY6A−/+ cells in vitro.(A) EdU labeling of LY6A−/+ cells in culture.(B) Androgen (androstenedione and testosterone) production by LY6A+ cells after their steroidogenic cell-inducing differentiation for 12 days.(C) Expression of two steroidogenic genes by LY6A+ cells before and after their differentiation in vitro for 6 days.(D) CYP11A1 immunofluorescent and HSD3B enzymatic staining of LY6A+ cells before (d0) and after (d10 or d12) their differentiations in vitro for up to 12 days.

Figure 6
Figure 6 Steroid hormone production and gonadotropin receptor gene expressions.(A) Progesterone and testosterone production by steroidogenic cells generated from LGR5, PDGFRA, or LY6A progenitors.(B) Lhcgr and Fshr expressions by steroidogenic cells generated from LGR5, PDGFRA, or LY6A progenitors.Data are expressed as mean ± SE of three to four individual experiments.*, **, ****: significantly different from the day 10 differentiated cells at P < 0.05, 0.01, and 0.0001, respectively.