Isolation and characterization mesenchymal stem cells from red panda (Ailurus fulgens styani) endometrium

In this work, we, for the first time, isolated and characterized the endometrial mesenchymal stem cells from red panda. It could be beneficial not only for the germ plasm resources conservation, but also for basic or pre-clinical studies (e.g. the study of interspecies somatic cell nuclear transfer and embryonic diapause) in the future.


Introduction
Red pandas (Ailurus fulgens), world-famous wild animals, are mainly distributed in the west of China, the Himalaya mountain ranges of Nepal, India, Bhutan and Myanmar (Li et al., 2005). The red pandas are recently classified as two separate species Ailurus fulgens fulgens and Ailurus fulgens styani. In China, both subspecies are found and were separated by Yalu Zangbu River (Hu et al., 2020). Due to the destructed habitats and declining population, the red panda was listed as endangered by the International Union for Conservation of Nature (IUCN) Red List of Threatened Species in 2015 (Glatston et al., 2015). Nowadays, some measures such as ex situ and in situ conservations were taken to protect red pandas. In addition, the efficient preservation and utilization of stem cell resources is also an important way to protect endangered wildlife (Stanton et al., 2019). Assisted reproduction combined with embryonic stem cell technology also played a valuable role in the effort to protect endangered species (Hildebrandt et al., 2018;Saragusty et al., 2020).
Mesenchymal stem cells (MSCs), firstly identified in the bone marrow, are a population of pluripotent cells with high proliferative rate, low immunogenicity, self-renewal and multi-directional differentiation potential throughout the entire stage of life (Friedenstein et al., 1966;Li et al., 2015;Mason et al., 2014). In humans, MSCs have been identified in many tissues including umbilical cord blood and adipose tissues (Lv et al., 2014). As for wild animals, earlier studies by our group have identified the giant panda bone marrow MSCs (Liu et al., 2013) and umbilical cord MSCs (Liu et al., 2021), as well as the red panda bone marrow MSCs for the first time .
Recently, in vitro gametogenesis was proposed as the ultimate solution for infertility caused by loss or compromised function of gametes (Makar & Sasaki, 2020). Reconstitution of primordial germ cell (PGC) specification from pluripotent cells is an essential first step for in vitro gametogenesis. Previous study reported that PGC-like cells were successfully derived from canine adipose mesenchymal stem cells (Wei et al., 2016). Human amniotic membrane MSCs could be induced to express PGC gene markers and have enough potential to PGC specification (Alifi & Asgari, 2020). During uterine organogenesis, cell communications were closer and polypotential germ cells differentiated and grew into myometrium and endometrial layers (Makiyan, 2017). Moreover, endometrium is the site for embryo implantation and accompanies all stages of post-implantation embryo development and has a direct intercellular communication with the embryo (Massimiani et al., 2019). Early studies in bovine have found that eMSCs ensured the maternal immunomodulation required for embryo survival (Calle et al., 2019). Therefore, the eMSCs would have the potential to PGC specification, in vitro gametogenesis and embryo implantation regulation.
For endangered wild animals, cell germ plasm resources are extremely precious, which is one of the key factors to protect the genetic diversity of species. In addition, the research on the reproductive mechanism of wild animals is not well explained and the potential regulation should be further studied.
Therefore, the aim of this study was to isolate and characterize MSCs from red panda endometrium. The new type of cells will be beneficial for germ plasm resources conservation of red panda, as well as for basic or pre-clinical studies in the future.

Isolation of eMSCs from red pandas
The samples used in this study were obtained postmortem from six red pandas (Supplementary Table S1), which were raised in the Chengdu Research Base of Giant Panda Breeding. All eMSCs isolation performance of the six red pandas were performed as follows. Briefly, the harvested uterus samples were washed five times in phosphate-buffered saline (PBS, Gibco) with 5% antibiotic-antimycotic solution (Gibco). Uteruses were cut into 1 × 1 × 1 mm pieces without fatty tissues under sterile condition, and then washed twice in PBS. The uteruses were mechanically minced and dissociated into single cells with collagenase type IV (1 mg/ml; Gibco) for 15 min at 37 • C, then centrifuged at 600 g for 3 min. Discarding the supernatant, the tissue was further digested with 0.25% trypsin (Gibco) for 15 min at 37 • C. Cell suspensions were filtered through a 40-μm sieve (BD Falcon). The filtrates were centrifuged at 600 g for 5 min at room temperature. The cell pellets were resuspended and cultured in culture dishes (diameter, 10 cm; Corning) containing lowglucose Dulbecco's modified Eagle's medium (LG-DMEM, Gibco) supplemented with 10% foetal bovine serum (Gibco), 10 ng/ml basic fibroblast growth factor (Peprotech) and 1× antibiotic-antimycotic. The cells were incubated at 37 • C with 5% CO 2 , and the medium was changed every 2 days. At ∼10 days, the primary eMSCs were passaged with 1× TrypLE Express (Gibco) when the cells reached ∼80% confluence.

Cellular proliferation assay
Red panda eMSCs at passages 4-7 were seeded in 24-well plates at a density of 1 × 10 4 cells per well to establish the growth curves. The numbers of cells per three wells were counted every day for eight successive days. Cells were counted by automatic cell counter (Countstar) after acridine orange (AO)/propidium Iodide (PI) staining (Countstar).

Cell surface antigen analysis
Red panda eMSCs at passage 4 were digested with Try-pLE Express (Gibco), then washed twice and incubated in buffer with the relevant antibody or with the corresponding isotype control IgG for 40 min. Then cells were washed three times and analysed by using flow cytometry (NovoCyte, ACEA). Compensation and data analysis were performed using FlowJo software (Tree Star, Inc., Ashland, OR, USA). Corresponding antibodies used for flow cytometry analysis were listed in Table 1.

Karyotype analysis
For karyotype analysis, red panda eMSCs at passage 8 were exposed to 10 ug/ml colcemid (Beyotime) for 4 h, then digested and resuspended in 0.075 M KCl (Sigma) at 37 • C for 40 min. After that, eMSCs were fixed in acetic acid and methanol (1:3) (Sigma). The numbers of chromosomes were counted by an inverted fluorescence microscope (BX53, Olympus) with an oil immersion objective. Chromosome images were analysed by ImageJ software (National Institutes of Health, Bethesda, MD, USA).

RT-PCR
Red panda eMSCs at passage 4 were used for reverse transcription-polymerase chain reaction (RT-PCR) analysis.
Total RNA of cultured cells was extracted with the RNAprep Pure Cell Kit (TIANGEN) in accordance with the manufacturer's instruction. Then, the samples were treated with DNase to remove possible contamination by genomic DNA and reverse transcribed into cDNA using PrimeScript RT reagent Kit (Takara). The specific primer sequences were shown in Table 2, and β-actin was used as reference gene.

RNA-seq and differentially expressed genes enrichment analysis
Red panda eMSCs isolated from three different individuals and corresponding red panda skin fibroblast cells (skin FCs) at passage 4 were seeded in culture dishes (diameter, 10 cm) and treated with normal growth medium. When the cells reached ∼80% confluence, the cells were collected and treated with Trizol (Thermo Fisher) as manufacturer's protocol to extract total RNA and then for RNA-seq (Novogene). The RNA-seq data were assembled and analysed as no reference genome sequences. Differential expression analysis was performed with DESeq2. For functional enrichment analysis, DEGs were mapped to terms in the GO database, and then searched for significantly enriched GO terms (P < 0.05). DEGs were mapped to the KEGG database, and searched for significantly enriched KEGG pathways (P < 0.05).

Multilineage differentiation
For adipocytic differentiation, eMSCs at passage 4 were seeded in 6-well plates and treated with adipogenic medium (Cyagen) as per the manufacturer's protocol. The medium was changed three times per week. After 8 days, adipogenesis was evaluated by Oil red O staining (Sigma). Staining was assessed by bright-field inverted microscopy (IX73, Olympus). Red panda eMSCs cultured in normal growth medium served as control.
For chondrogenic differentiation, eMSCs at passage 4 at a density of 4 × 10 5 cells were cultured in chondrogenic differentiation medium (Cyagen) as per the manufacturer's protocol. The medium was changed three times per week. After 21 days, chondrogenesis was detected by the staining of toluidine blue. Red panda eMSCs cultured in normal growth medium served as control.
For hepatogenic differentiation, eMSCs were cultured in hepatogenic differentiation medium (Cyagen) as per the manufacturer's protocol. The medium was changed three times per week. After 16 days, hepatogenic differentiation was evaluated by cytokeratin 18 (CK18) (ab181597, Abcam) immunofluorescence staining, detection of CK 18, ALB and DKK1 mRNA expression, ALB protein expression, Periodic Acid-Schiff (PAS) staining and Indocyanine Green (ICG) uptake. Red panda eMSCs cultured in normal growth medium served as control.

PAS staining
Red panda eMSCs after hepatogenic differentiation were stained by PAS staining kit (Solarbio). Following the manufacturer's instructions, hepatogenic differentiated cells were washed twice with PBS, then fixed with 4% paraformaldehyde (Sigma) for 15 min. Cells were incubated in 1% periodic acid for 20 min and then washed four times with distilled water. After incubating with Schiff regent for 20 min, the cells were counterstained with Mayer haematoxylin solution for 2 min. Then, the stained cells were rinsed in distilled water and photographed under an inverted microscope (IX73, Olympus).

ICG uptake
Red panda eMSCs after hepatogenic differentiation were determined through detecting the cellular uptake of ICG. Briefly, hepatogenic differentiated cells were washed twice with PBS, then incubated with LG-DMEM supplemented with 1 mg/ml of ICG (Sigma), and incubated at 37 • C with 5% CO 2 for 1 h. Subsequently, the cells were washed three times with PBS, and then photographed under an inverted microscope (IX73, Olympus).

Statistical analysis
The results of growth curves were analysed by using Graph-Pad Prism 5 software (GraphPad Software, San Diego, CA, USA). Data were expressed as the mean ± SEM. All results were generated from at least three independent experiments.

Isolation and culture of red panda eMSCs
After 9 days culture, the primary cells were fibroblast like and epithelial like, presenting triangular, fusiform, ovoid or polygonal shapes, and the overall percentage of eMSCs was ∼50% (Fig. 1A). Red panda eMSCs at passage 4 were fibroblast like and presented polygonal or long spindle shapes (Fig. 1B). After seven generations, eMSCs also have the ability to maintain their morphological characteristics (Fig. 1C). The growth curves of red panda eMSCs from passage 4 to 7 were established. Results showed that eMSCs entered the exponential phase after 4 days culture and then reached the stationary phase after 7 days culture (Fig. 1D). After culturing eight passages, the chromosome number of red panda eMSCs are still normal (2n = 36) ( Fig. 1E and F).

Surface antigens and transcription factors in red panda eMSCs
The expressions of pluripotency and MSC marker genes, including Thy1, Klf4, Sox2 and CD44, were confirmed by RT-PCR. Genes Thy1, Klf4, Sox2 and CD44 were all highly expressed in red panda eMSCs ( Fig. 2A). Additionally, SOX2 protein was also detected in eMSCs, but not in skin FCs (Fig. 2B). According to the flow cytometry analysis, the red panda eMSCs were positive for MSCs phenotype CD44, CD49f and CD105 and negative for endothelial cell marker CD31 and haematopoietic cell marker CD34 (Fig. 2C). antibodies. Blue areas, signal from isotype controls; red areas, signal from the specific cell surface marker; grey areas, unstained black control.

Differential gene expression analysis between red panda eMSCs and skin FCs
A total of 79 311 and 55 016 predicted expressed genes were identified in eMSCs and skin FCs, respectively. Of these, 47 981 genes were common expressed in eMSCs and skin FCs (Fig. 3A). Compared with skin FCs, 5034 genes upregulated in eMSCs and 4070 genes downregulated (Fig. 3B), and DEGs could significantly separate the samples into the eMSCs group and skin FCs (Fig. 3C). DEGs were analysed by GO annotation, and the significantly enriched GO terms mainly contained G-protein coupled receptor signalling pathway, carbohydrate derivative binding, nucleoside binding, purine nucleoside binding, ribonucleoside binding, motor activity, etc. (Fig. 3D). The signal pathways of DEGs were analysed by KEGG pathway analysis. As shown in Fig. 3E, the top 20 mainly enriched KEGG pathways were ribosome biogenesis, cell cycle, DNA replication, Ras signalling path-way, purine metabolism, etc. In addition, some representative genes about promoting MSCs differentiation and proliferation were upregulated and promoting fibroblasts proliferation were downregulated in eMSCs group when compared skin FCs. Also, all these genes significantly separated the samples into the eMSCs group and skin FCs ( Fig. 3F; Table 3).

Differentiation of red panda eMSCs
For adipocytic differentiation, cells were cultured in adipocytic induction medium. In the control group, no positive staining signal of Oil-red O was detected (Fig. 4A). In the induced group, many lipid droplet accumulations were detected by the staining of Oil-red O (Fig. 4B). sulfated proteoglycans of the cartilage matrices. The results showed positive staining signals (Fig. 4C).
To determine whether red panda eMSCs have the ability to endoderm differentiation, cells were cultured in hepatogenic differentiation medium. As shown in Fig. 5A, the cells after differentiating became flat, presenting polygonal morphology when compared with the control group. Immunofluorescence staining revealed strong positive signals of the hepatogenic marker CK18 in the hepatogenic-induced group, but no positive staining was detected in the control group (Fig. 5B). The results of PAS staining showed that extensive cytoplasmic positive staining signalling (red to purple) were only observed in the hepatogenic-induced group (Fig. 5C). The hepatogenic differentiated group successfully showed an indocyanine green uptake staining and the control group could not upake ICG (Fig. 5D). In addition, the results of RT-PCR further showed that the hepatic and liver progenitor marker genes CK18, ALB and DKK1 were highly expressed in the hepatogenic-induced group when compared with the control  FBLN5 Downregulated −4.630 0.000 Affect adhesion and proliferation of human fibroblast-like cells (Furie et al., 2016) group (Fig. 5E). And the result of western blot also showed that the ALB was highly expressed in the hepatogenic-induced group when compared with the control group (Fig. 5F).

Discussion
In the present study, we successfully isolated eMSCs in red panda endometrium. The primary cells were fibroblast like and epithelial like, presenting triangular, fusiform, ovoid or polygonal shapes, because primary cells were a mixed-cell group, containing epithelial cells, basal layer cells and eMSCs. After passage cultivations, fibroblast-like eMSCs gradually became the major cells, suggesting that the currently used culture method would greatly benefit eMSCs growth and reproduction. During proliferation in eight successive days, eMSCs at passage 4 experienced a short lag phase from Days 1 to 4 and subsequently a logarithmic rise from Day 4, then reached the stationary phase at Day 7 (Fig. 1). However, eMSCs at passages 5-7 did not reach the plateau phase at Day 8. It may be that the cells did not reach the maximum growth density. The eMSCs at passage 7 still had a strong proliferative potential, suggesting eMSCs had strong proliferation stability. These results were similar to the growth curves of mesenchymal stem cells isolated from red panda bone marrow . diploid karyotype (2n = 36), suggesting that this type of cell had a stable and normal growth and reproduction, as well as further confirmed that the currently used culture method was appropriate.
To characterize the red panda eMSCs isolated in present study, we examined some MSCs biomarkers, such as CD44, CD49f, CD90 and CD105. The red panda eMSCs highly expressed CD44, CD49f, CD105, but not expressed CD31 and CD34 when compared with the isotype control. The results of RT-PCR also confirmed CD90 (Thy1) was highly expressed in red panda eMSCs. These results suggested that the red panda eMSCs were neither endothelial cell nor haematopoietic cell and was in accord with the mesenchymal cells standards of the International Society for Cell Therapy (Dominici et al., 2006). Pluripotency genes Sox2 and Klf4 also expressed in red panda eMSCs. However, other markers THY1 and OCT4 were not detected with western blot analysis, this may be because there are no specific antibodies available for red panda or the expression of these markers in red panda eMSCs is too low to be detected. In the future studies, more specific antibodies to recognize red panda should be selected and further supplement the current study.
In order to analyse the red panda eMSCs characters in gene expression profile, RNA-seq analysis between eMSCs and skin FCs was performed. To date, the complete gene expression profile of red panda has not yet been established, thus the present study adopted a no reference transcriptome analysis, which might lead to an incomplete functional genes annotation. We used Trinity software (Grabherr et al., 2011) to identify 86 346 unigenes (genes), and the number of genes was different from referenced mammalian animals. We identified 9104 genes that were differentially expressed. Moreover, the top 20 enrichment pathways of DEGs in GO and the KEGG mainly associated with G-protein coupled receptor signalling pathway, carbohydrate derivative binding, nucleoside binding, ribosome biogenesis, cell cycle, DNA replication, Ras signalling pathway, purine metabolism and cell cycle. These results suggested that eMSCs had a high frequency of cellular activity and proliferative capacity. Among the DEGs, some genes about promoting MSCs differentiation and proliferation were upregulated and promoting fibroblasts proliferation were downregulated in eMSCs group (Table 3), which further confirmed the eMSCs pluripotency.
MSCs have the ability to multiple differentiation, and earlier studies had reported that MSCs could differentiate into adipocytes, osteoblasts, chondrocytes, neural cells and smooth muscle cells (Liechty et al., 2000;Wu et al., 2007). In the present study, we confirmed the red panda eMSCs could be differentiated into adipocytes and chondrocytes, from which were mesoderms. Additionally, the red panda eMSCs also had the ability to differentiate into hepatocytes, from which were endoderms. This result revealed that the red Red panda eMSCs after hepatogenic differentiation were stained with anti-CK18 (hepatogenic marker, green), and the nucleus were stained with DAPI (blue). Untreated cells were used as control. Scale bar, 50 μm. (C) Red panda eMSCs after hepatogenic differentiation were stained with PAS staining (purple signal). Untreated cells were used as control. Scale bar, 100 μm. (D) Red panda eMSCs after hepatogenic differentiation were detected by ICG uptake (green signal). Untreated cells were used as control. Scale bar, 100 μm. (E) The expressions of liver-specific genes in hepatogenic-induced (H) and control (C) red panda eMSCs were detected by using RT-PCR. β-actin was used as a reference gene. (F) The expressions of ALB in hepatogenic-induced (H) and control (C) red panda eMSCs were detected by using western blot. GAPDH was used as the loading control. panda eMSCs have the capacity for differentiation potential across embryonic lineage boundaries. Interestingly, previous research had shown that MSCs from bone marrow could differentiate into hepatocytes in vitro (Schwartz et al., 2002). Moreover, some studies revealed that eMSCs are likely to be derived from bone marrow (Cervello et al., 2012;Cervello et al., 2015), but the real origin of red panda eMSCs needs further studies.
The red panda eMSCs with high proliferation in vitro, stable karyotype and multipotential differentiation have more potential applications in wildlife protection and future research and/or clinical practice. Firstly, somatic cell nuclear transfer (SCNT) or interspecies SCNT (iSCNT) may be a potential tool for aiding the conservation of endangered animal species, although accompanied with low efficiency and mitochondrial heterogeneity, which could compromise the energy-making process of embryo, leading to its death (Loi et al., 2011). To date, iSCNT has been successfully performed in many endangered wild animals, such as African wildcat (Felis lybica) (Gomez et al., 2004), Bactrian camel (Camelus bactrianus) (Wani et al., 2017) and even the extinct species Pyrenean ibex (Capra pyrenaica pyrenaica) (Folch et al., 2009). A previous study tried to construct interspecies cloned embryos by using red panda fibroblasts and rabbit enucleated oocytes (Tao et al., 2009). It is well known that fibroblasts were common donor cells used in SCNT, but MSCs were also suitable. In 2018, horse MSCs from bone marrow were used as nuclear donor cells and produced healthier cloned horses compared with fibroblasts (Olivera et al., 2018). Red panda eMSCs derived from endometrium, which directly interacted with early embryos, may be more suitable and further increase the SCNT embryo development. Secondly, to date, embryonic  (Deng et al., 2018) including red panda (Macdonald et al., 2010;Miles et al., 1979), but the potential mechanism is still unclear. Embryo implantation is a key step in the establishment of pregnancy, and the eMSCs is a great cell model for studying embryonic diapause. On one hand, endometrium of mammal secretes cytokines and growth factors that influence the development of early embryo (Cha et al., 2012). It is likely that some of growth factors control the arrested growth that occurs in diapause (Renfree & Fenelon, 2017). It is also clear that MSCs secrete a variety of cytokines and growth factors, such as insulin-like growth factor-1, vascular endothelial growth factor, epidermal growth factor, fibroblast growth factor, interleukin-6, leukaemia inhibitory factor and transforming growth factor-β (Feng et al., 2009). Therefore, it will be a new angle to study embryonic diapause from the eMSCs secretion. On the other hand, during embryo implantation, there is intercellular communication between the embryo and maternal eMSCs and peripheral blood MSCs (pbMSCs), which could chemotax to embryonic trophectoderm secretome (Calle et al., 2021a, Calle et al., 2021b. In bovine embryo implantation, the migratory capacity of eMSCs was increased towards an inflammatory niche and then reduced by the expression of implantation cytokine by the embryo, which are necessary for immunorepression to prevent embryo rejection by the maternal organism (Calle et al., 2019). Therefore, deeply to study the communication between eMSCs and embryo will be beneficial for clarifying the regulation mechanism of embryo implantation. Finally, early embryo mainly develops in the maternal oviduct and uterus, which contains a large number of factors that can promote early embryo development (Kolle et al., 2020). Recent studies found that coculture of mouse embryos and mesenchymal stem/stromal cells derived from menstrual blood enriched the embryonic microenvironment and promoted embryo development (Goncalves et al., 2020). Coculture with extracellular vesicles from endometrialderived MSCs could increase the quality of aged mouse embryos and presumably by modulating the expression of antioxidant enzymes and promoting pluripotent activity (Marinaro et al., 2019). Therefore, paracrine regulation of eMSCs may be a feasible way to promote in vitro embryo development, which will benefit endangered wild animals like the red panda.

Conclusion
In this study, we, for the first time, isolated and characterized the red panda eMSCs in endometrium. Red panda eMSCs were fibroblast like and highly expressed the pluripotency genes including Thy1, Klf4 and Sox2. Additionally, red panda eMSCs were positive for MSC markers CD44, CD49f and CD105 and negative for endothelial cell marker CD31 and haematopoietic cell marker CD34. The red panda eMSCs also had the pluripotent differentiation capacities of adipocytes, chondrocytes and hepatocytes. Using RNA-seq, significant DEGs were identified, which further demonstrated the eMSC gene expression characters. The new multipotential stem cell could not only benefit the germ plasm resources conservation of red panda, but also basic or pre-clinical studies in the future.

Funding
This work was supported by the Sichuan Science and Technology Program (2020JDJQ0074); the Chengdu Research Base of Giant Panda Breeding (2020CPB-B07); and the Chengdu Giant Panda Breeding Research Foundation (CPF2017-16).