Long-term maintenance of human endometrial epithelial stem cells and their therapeutic effects on intrauterine adhesion

Human endometrial SSEA-1+ and SUSD2+ cells isolation and purification

Based on previous findings regarding the survival and growth requirements for adult and mouse mammary gland progenitor cells in vitro, the candidates of the transition and expansion medium (TEM) were selected from a pool of growth factors and bioactive small molecules [17, 18]. All of the selected molecules are identified to date that provide enhancements of somatic cell reprogramming [19]. This led us to ask whether certain combinations of these molecules were an optimum culture medium for the survival and growth of SSEA-1+ cells. The whole experimental procedure for SSEA-1+ and SUSD2+ cells isolation and purification were schematically shown in Fig. 1a. To test this idea, we firstly isolated human primary EpCAM+ cells from human endometrial tissues using MACS-based sorting to exclude EpCAM− stromal cells (Additional file 1: Figure S1a). Contrary to the general notion that endometrial EpCAM+ cells do not proliferate in vitro, we observed the proliferation of endometrial EpCAM+ cells in the presence of multiple combinations of these factors (Fig. 1b) [20]. The results shown in Additional file 1: Figure S4b indicate that the EpCAM+ cells expressed EpCAM, Cytokeratin and E-cadherin while Vimentin was only observed in EpCAM− cells.

Fig. 1figure 1

Generation of human SSEA-1+ cells in vitro. a Overview of the protocol used to obtain SSEA-1+ and SUSD2+ cells. b Light microscopy images of primary endometrial cells, P3 SUSD2+ and SSEA-1+ cells cultured in TEM at day 1 and day 5. Blue arrow, EpCAM+ cells; red arrow, EpCAM− cells. Scale bars, 100 µm. c Immunofluorescence analyses demonstrating the expression of SUSD2 and SSEA-1. Scale bars, 50 µm in 20X, 20 µm in 60X. d Light microscopy images show typical 6-well plates of clone formation of SUSD2+ and SSEA-1+ cells by crystal violet staining. Scale bars, 1 cm in 1X, 100 µm in 10X. e SUSD2+ cells showed higher colony-forming unit numbers than SSEA-1+ cells (Error bars represent s.d.; n = 3 technical replicates; two-tailed unpaired t-test, **P < 0.01). f CCK-8 analyses demonstrating higher cell proliferation of SUSD2+ cells. (Error bars represent s.d.; n = 3 technical replicates from one donor; two-way ANOVA, P < 0.001). g Doubling time calculated for SUSD2+ and SSEA-1+ cells. Error bars represent s.d.; n = 3 technical replicates, *P < 0.05. h Flow cytometric analysis showing the proportion of positive-cells in the populations. Red, positive-cells; blue, negative controls

SSEA-1 is a cell surface glycan which indicates an undifferentiated state of the committed endometrial progenitors [8, 10]. With increase in number of passages, the SSEA-1 expression of the sorted EpCAM+ cells was gradually elevated when cultured in TEM (Additional file 1: Figure S1b). At the 3rd passage, we used FACS-based sorting to achieve purified SSEA-1+ cells. Next, we optimized the culture condition for maintenance and expansion of the SSEA-1+ cells. Upon the sequential withdrawal of individual factors from TEM, the proliferative capacity decreased accordingly, which suggested that each factor had a crucial influence on the growth of SSEA-1+ cells (Additional file 1: Figure S2a,b). Notably, proliferating cells were scarcely observed without Y2732 or CHIR99021 treatment, which were considered to be key factors for stable culturing of adult mammary progenitor cells (Additional file 1: Figure S2a) [16, 21, 22]. Consistent with this pattern, lower numbers of colonies were shown in cells cultured with additional Nicotamide or PD0325901 (Additional file 1: Figure S2a,b). These results suggest that the synergistic relationship among these factors in TEM has a strong effect on stably expanding SSEA-1+ cells. Supporting this suggestion, similar proliferative capacity was found in three donor-derived SSEA-1+ cells cultured in TEM (Additional file 1: Figure S2c).

The purified SSEA-1+ cells were positive for SSEA-1 both by immunofluorescence staining and flow cytometry (Fig. 1c, h). Flow cytometry analysis revealed that the SSEA-1+ cells purity was 97%, accompanied by a strong expression of CD24 (81.3%) and they were negative for N-cadherin, CD31, CD34, CD45, CD90, and CD105 (Fig. 1h). Typical 3D confocal laser scanning image of SSEA-1+ and SUSD2+ cells were presented in supplemental material. The SSEA-1+ cells exhibited a homogeneous whorled or polyhedral morphology and formed tight complex clone structures (Fig. 1b). The immunohistochemistry of the proliferative endometrial tissues showed that all endometrial luminal and glandular epithelial cells were positive for Cytokeratin. However, only a small proportion of endometrial epithelial cells were positive for SSEA-1, whereas all epithelial cells were negative for N-cadherin (Additional file 1: Figure S1c).

According to Masuda et al., the SUSD2 was identified as a novel single marker for purifying human eMSCs [7]. After expansion in TEM, almost all P3 endometrial EpCAM− cells were positive for SUSD2 both by flow cytometry and immunofluorescence staining (Fig. 1c, h). To confirm the eMSCs phenotype, FACS analyses revealed that the SUSD2+ cells were positive for CD 90 and CD105, and negative for SSEA-1, N-cadherin, CD31, CD34 and CD45 (Fig. 1h). These results were consistent with previous reports about eMSCs’ surface markers [23]. The SUSD2+ cells presented a characteristic fibroblast-like morphology and were arranged in a spiral pattern (Fig. 1b). These results indicate that the SUSD2+ cells, or human eMSCs, were successfully isolated and identified in the present culture system.

Although both SSEA-1+ and SUSD2+ cells displayed great colony formation potential when cultured in TEM, the SUSD2+ cells had higher colony-forming unit numbers (223 ± 6) than SSEA-1+ cells (134 ± 6) (Fig. 1d, e). In addition, the cell proliferative capacity of SUSD2+ cells was significantly higher than SSEA-1+ cells (Fig. 1f), accompanied by a slightly lower mean population doubling time (21.4 ± 0.2 h vs. 22.6 ± 0.2 h, Fig. 1g). We then asked whether SSEA-1+ cells could undergo long-term culture without losing proliferative capacity. Previous studies demonstrated that OCT-4, NANOG, SOX2 and SOX9 are expressed in adult stem cells under a less-differentiated state [24,25,26]. We found that the late-passage SSEA-1+ cells displayed reduced proliferative capacity and decreased mRNA expressions of OCT-4, NANOG, SOX2 and SOX9 (Additional file 1: Figure S3a,c). The morphology of P10 SSEA-1+ cells became apical, and the colonies were found scattered with an increase in apoptosis (Additional file 1: Figure S3e, f). The apoptotic rate of P10 SSEA-1+ cells was much higher than that of P1 SSEA-1+ cells (71.3% vs. 5.6%). In contrast, the SUSD2+ cells showed no signs of senescence with passages (Additional file 1: Figure S3d). There were no significant differences of mRNA expressions of OCT-4, NANOG, SOX2 and SOX9 among the P1, P6 and P10 SUSD2+ cells (Additional file 1: Figure S3b). The karyotype analysis revealed that all the three independently established P10 SSEA-1+ and SUSD+ cells maintained normal diploid karyotypes (Additional file 1: Figure S3g).

Gene expression profiles between SSEA-1+ and SUSD2+ cells

To further characterize the function of SSEA-1+ and SUSD2+ cells, RNA-seq analysis was performed to compare the gene expression levels in these two cell populations from three donors. Pearson correlation analysis revealed that the SSEA-1+ and SUSD2+ cells exhibited a high degree of phenotypic homogeneity (Fig. 2a). The volcano plot for hierarchical clustering showed that 2460 genes were up-regulated, and 2136 genes were down-regulated in SSEA-1+ cells among differentially expressed 4596 genes (twofold change, P < 0.05, Fig. 2b). Next, the global gene expression profiles from SSEA-1+ and SUSD2+ cells were subjected to unsupervised cluster analysis that identified discrete segregation between the two cells. Selected endometrial epithelial cell-specific genes were notably more highly expressed in SSEA-1+ cells, whereas SUSD2+ cells expressed relatively higher endometrial stromal cells-specific genes (Fig. 2c). In line with previous findings, the epithelial stem cell-specific genes were significantly more highly expressed in SSEA-1+ cells, which were related to stem cell maintenance [20]. Gene Ontology (GO) term analysis of the up-regulated genes in the SSEA-1+ cells showed great relationships with keratinocyte differentiation, epidermis development, keratinization, epidermal cell differentiation, etc. (Fig. 2d). Conversely, the down-regulated genes were related to mesenchyme development, extracellular matrix and extracellular matrix/structure organization (Fig. 2d). These gene signatures support the conjecture that the SSEA-1+ cells were associated more with endometrial epithelial development and differentiation than SUSD2+ cells.

Fig. 2figure 2

RNA sequence–based transcriptome profiles of SSEA-1+ and SUSD2+ cells. a Pairwise correlation heatmap of RNA-seq samples based on the Pearson correlation of log gene expression values for all genes. b Volcano plot analysis of the differentially expressed mRNA after comparison between SSEA-1+ and SUSD2+ cells (> twofold changes and P < 0.05). c Left: Heat-map of up- or down-regulated genes in SSEA-1+ cells compared with SUSD2+ cells. Right: Heatmap on the selected epithelial genes, endometrial stem cell-specific genes and stromal genes between SSEA-1+ and SUSD2+ cells (> twofold changes and P < 0.05). d GO analysis of differentially expressed mRNAs. Top 30 GO biological processes enriched among up-regulated and down-regulated genes in SSEA-1+ cells compared with SUSD2+ cells. The color key is shown on the right. The most highly enriched GO categories are indicated in red. The size of the circles reflects the frequency of the GO term

Efficient endometrial epithelial cell-like cells (EEC-like) differentiation of SSEA-1+ cells

In normal endometrium, estrogen (E2) stimulates the proliferation of endometrial epithelial cells. Previous studies reported that differentiation medium containing E2 could induce stem cells to differentiate into EEC-like cells [27,28,29]. This let us to hypothesize that the SSEA-1+ cells might differentiate into EEC-like cells with the E2-based differentiation medium. After incubation in differentiation medium for 20 days, the morphology of SSEA-1+ cells changed from a whorled clone to a fusiform shape and the clone structures were not well-knit anymore (Fig. 3b). With the time extension, apoptosis of SSEA-1+ cells increased greatly. By contrast, no obvious morphologic changes of SUSD2+ cells were found (Fig. 3b).

Fig. 3figure 3

SSEA-1+ cells exhibit endometrial epithelial stem cell phenotype. a Overview of the protocol for SSEA-1+ and SUSD2+ cells differentiation into EEC-like cells. b Morphologic change of SSEA-1+ and SUSD2+ cells after differentiation. Scale bars, 100 µm. SSEA-1+/SUSD2+ Diff, SSEA-1+/SUSD2+ cells differentiation. c Flow cytometric analysis showing the proportion of EpCAM and CD9 positive-cells in SSEA-1+ or SUSD2+ cells at day 0 and day 20 of differentiation. Red, positive-cells; blue, negative controls. d Immunofluorescence analyses demonstrating the expression of Cytokeratin and Vimentin in EEC-like cells differentiated from SSEA-1+ and SUSD2.+ cells. e Gene expression levels for EpCAM, CD9, OCT-4, SOX2 and NANOG. Expression normalized to β-actin (two-tailed unpaired t-test, Error bars represent s.d., n = 3; n.s., non-significant; *P < 0.05, **P < 0.01, ***P < 0.001)

Differentiated EEC-like cells were also detected by flow cytometric analysis for EpCAM and CD9 expressions. The epithelial cell markers were remarkably up-regulated in the EEC-like cells differentiated from SSEA-1+ cells, with over 95% CD9 and over 98% EpCAM expressions (Fig. 3c). A low level of EpCAM (21.9%) expression was found in EEC-like cells differentiated from SUSD2+ cells, however, these cells were negative for CD9 expression (Fig. 3c). As shown in Fig. 3d, the immunofluorescence analysis revealed that the EEC-like cells differentiated from SSEA-1+ cells were positive for Cytokeratin and negative for Vimentin. On the contrary, the EEC-like cells differentiated from SUSD2+ cells were positive for Vimentin and negative for Cytokeratin. The qPCR analysis of EEC-like cells differentiated from SSEA-1+ cells demonstrated the obviously down-regulation of transcripts related to pluripotency (OCT-4, NANOG and SOX2), accompanied by the significantly up-regulation of transcripts of epithelial, lineage-associated transcripts (CD9 and EpCAM) after day 20 differentiation, as compared with SSEA-1+ cells cultured in TEM (Fig. 3e). Nevertheless, the expressions of CD9 and EpCAM increased slightly and the OCT-4, NANOG and SOX2 decreased slightly after differentiation of SUSD2+ cells. As expected, SSEA-1+ cells differentiated into EEC-like cells with a high differentiation efficiency whereas SUSD2+ cells failed to do so. Although MSCs were reported with potential ability that they could cross lineage barriers to differentiate into endometrial epithelial cells [30, 31], SUSD2+ cells seemed to stay in a mesenchymal state with TEM, resulting in a low differentiation efficiency in the present differentiation system. At the late period of differentiation of SSEA-1+ cells, the EEC-like cells displayed less proliferation and increased cell apoptosis (Fig. 3d). This appearance is similar to the epithelial cells in monoculture [32], indicating that the SSEA-1+ cells transformed from a progenitor state to a maturation state after differentiation.

Establishment of SSEA-1+ cells-derived organoid in 3D culture

One common feature to all human organoid is that they are derived from pluripotent stem cells or adult stem cells [33]. Recent efforts to establish endometrial epithelial organoid have been successful with the freshly isolated endometrial cells [34,35,36], suggesting the presence of eESCs. Under the culture system utilizing Matrigel and TEM, the SSEA-1+ cells rapidly self-organized into spheroid-like or organoid-like structures with a hollow center within 7 days that further expanded in size after passages (Fig. 4a). In contrast, the SUSD2+ cells could not form any spheroids but the typical fibroblast-like appearance. While cultured in low-attachment wells, the SUSD2+ cells formed solid and non-opaque spheres with no loss of cell aggregation for up to 10 days, indicating that the SUSD2+ cells-spheres were just the cell aggregation rather than organoids. On the other hand, hollow-center organoids were gained from the SSEA-1+ cells in suspension culture. The spheroid diameter of SSEA-1+ cells in suspension culture was far smaller than that in Matrigel culture. The immunofluorescence staining indicated that no cross-contamination between SSEA-1+ and SUSD2+ cells (Fig. 4a).

Fig. 4figure 4

Organoids develop from SSEA-1+ cells and reproduce the phenotype of the epithelium. a Morphologic change and immunofluorescence analyses of SSEA-1+ and SUSD2+ cells embedded in Matrigel and in suspension culture. Scale bars, 50 µm in light microscopy images, 100 µm in confocal images. b H&E staining for organoids from SSEA-1+ cells. Scale bars, 200 µm in 10X, 100 µm in 20X. c The Cytokeratin, Vimentin and ER immunoexpression in the endometrium of proliferative phase and SSEA-1+-organoids. Scale bars, 100 µm in endometrium, 50 µm in SSEA-1+-organoids. d qPCR analyses for the expressions of OCT-4, SOX2, NANOG, E-cadherin, CD13, EpCAM, CD9 and ER in SSEA-1+ cells and organoids from SSEA-1.+ cells. Expression normalized to β-actin (n = 3 donors; two-tailed unpaired t-test, n.s., non-significant, * P < 0.05, ***P < 0.001)

Organoids could remodel the epithelial compartment of a tissue without the essential presence of mesenchymal compartment of the tissue [37]. The typical HE staining results of SSEA-1+ cells forming organoids were shown in Fig. 4b. The SSEA-1+ cells forming organoids reproduced endometrial epithelium phenotype after 14 days culturing in TEM. The organoids expressed Cytokeratin and ER but Vimentin, mimicking the in vivo epithelium phenotype of the proliferative phase of human endometrial glands (Fig. 4c). In line with this expression, we found that addition of E2, although not strictly required, could slightly promote the expansion of the organoids (data not shown). Moreover, the mature epithelial genes including E-cadherin, EpCAM, CD9 and ER were up-regulated and the pluripotency-maintaining genes (OCT-4, SOX2 and NANOG) were down-regulated in SSEA-1+ cells forming organoids. The mRNA expression of CD13, the marker of the stromal lineage, was not altered after self-organized into organoid (Fig. 4d). Taken together, we established the human endometrial organoids that mimic the epithelial characteristics of the tissue from a purified long-term cultured human endometrial cells, instead of endometrial fragments or other co-culture systems, thereby providing a promising in vitro model for studying human endometrial biology in greater depth.

Obtaining the adult stem cells-derived organoids required essential niche factors to support stem cell activity [38]. Previous studies have demonstrated that epithelial proliferation and stem cell self-renewal are dependent on EGF and WNT activity and the long-term maintenance of human organoids requires inhibition of the TGF-β pathway [39, 40]. Our results demonstrated that the cytokine cocktail in TEM could maintain the adult stem cell biological characteristics of SSEA-1+ cells in both 2D and 3D cultures.

SSEA-1+ cells had higher angiogenic potential than SUSD2+ cells

Angiogenesis plays a critical role in endometrial regeneration. To interrogate the angiogenic capacity of SSEA-1+ and SUSD2+ cells, the tube-CM (conditioned medium) from both the two cells were collected. The HUVECs exhibited more intensive and extended tubular networks after incubation with SSEA-1+-CM (Fig. 5a). The length per tube were significantly higher in SSEA-1+-CM treatment than those in SUSD2+-CM at both 6 and 12 h (Fig. 5b). In accordance, GSEA demonstrated an enrichment of VEGFA-VEGFR2 signaling pathway-related gene sets in SSEA-1+ cells (Fig. 5c). As VEGFA is a vital factor in angiogenesis [41], ELISA was performed to evaluate the VEGFA secretion from SSEA-1+ and SUSD2+ cells. The results revealed that the concentration of VEGFA were obviously higher in SSEA-1+-CM (1016 ± 56.61 pg/ml vs 0.61 ± 0.05 pg/ml, P < 0.001, Fig. 5d). Furthermore, the qPCR analysis validated that the SSEA-1+ cells displayed higher gene expressions of VEGFA and other related transcripts including TNF, IL-1A and IL-1B, than SUSD2+ cells (Fig. 5e). These data suggest that SSEA-1+ cells had higher angiogenic potential than SUSD2+ cells through VEGFA-VEGFR2 signaling pathway.

Fig. 5figure 5

Comparison of functional studies of SSEA-1+ and SUSD2+ cells. a Representative images showing HUVECs tube formation in SSEA-1+-CM and SUSD2+-CM. CM, conditioned medium. b Graph on tube formation for HUVECs incubated with SSEA-1+-CM and SUSD2+-CM (Error bars represent s.d.; n = 3 technical replicates; two-tailed unpaired t-test, ***P < 0.001). c Analysis of gene set enrichment analysis (GSEA) highlighting the upregulation of in SSEA-1+ cells (NES = 2.189 of the VEGFA-VEGFR2 signaling pathway). d VEGFA in supernatant of SSEA-1+ and SUSD2+ cell analyzed by enzyme-linked immunosorbent assay (n = 6,*** P < 0.001). e qPCR analyses for the expression of VEGFA, TNF, IL-1A and IL-1B in SUSD2+ cells and SUSD2+ cells. Expression normalized to β-actin (n = 3 donors, two-tailed unpaired t-test, ***P < 0.001). f Representative images of wound healing assay of endometrial stromal cells carried out in SSEA-1+ and SUSD2+ coculture groups at 0 h, 12 h and 24 h. Scale bars, 100 µm. g Bar charts of relative wound healing from the experiments (n = 3 donors, two-tailed unpaired t-test, **P < 0.01, ***P < 0.001). h Analysis of GESA revealed the upregulation of cell adhesion in SSEA-1+ cells (NES = 1.627 of the cell adhesion molecules cams pathway. The following heatmaps show the subsets of enriched genes involved in their corresponding pathway. (i) qPCR analyses for the expression of ITGB2, ITGB7, E-cadherin and EpCAM in SSEA-1+ cells and SUSD2.+ cells, respectively (n = 3 donors, two-tailed unpaired t-test, **P < 0.05, ***P < 0.001)

Endometrial stromal cells (eSCs) were more migratory when cocultured with SSEA-1+ cells

A rapid repair with eSCs around the wound were observed by scanning electron microscopic, which was vital for wound healing of the endometrium [42]. To investigate the impact of SSEA-1+ and SUSD2+ cells on cell migration, the human eSCs were used for wound healing assay after coculturing with two cells respectively. The eSCs were positive for Vimentin and negative for Cytokeratin by immunocytochemical staining (Additional file 1: Figure S4a). Migration of the eSCs monolayer was monitored over a period of 6 h and 12 h. Higher closure percentages were obtained in eSCs after cocultured with SSEA-1+ cells at both 6 h (62.98 ± 0.47% vs 32.38 ± 0.43%, P < 0.001) and 12 h (91.85 ± 0.73% vs 60.93 ± 1.71%, P < 0.001), than that in SUSD2+ cells coculture group (Fig. 5f, g). Meanwhile, GSEA revealed an enrichment of cell adhesion molecules (CAMS)-related gene sets in SSEA-1+ cells (Fig. 5h). CAMs enable cells to interact with other cells, influencing a wide variety of fundamental processes like tissue remodeling, repair and regeneration [43]. Next, the qPCR analysis validated that the CAMS related genes (ITGB2, ITGB7, E-cadherin and EpCAM) of eSCs were significantly up-regulated in SSEA-1+ cells than in SUSD2+ cells (Fig. 5i). Taken together, these findings demonstrated that the up-regulated CAMS in SSEA-1+ cells might lead to the increased migration of eSCs than SUSD2+ cells did in vitro.

SSEA-1+ cells seeded-chitosan implantation boosted endometrial regeneration of IUA rats

We established the IUA rat models via a scraping treatment to mimic the histopathological changes in IUA patients [44]. The uteri were collected 14 days after endometrial damage for gross, histological, and fibrosis evaluation. The IUA rat model was confirmed to have thinner endometrium, fewer endometrial glands and more fibrosis area than the normal endometrial cavity (Additional file 1: Figure S5a to e).

Chitosan is synthetic high molecular polysaccharide substance that poses great biocompatibility, degradability and biological activity, which can conduct as a physical barrier and inhibit the formation of accumulated scar tissue. All these properties make chitosan increasingly popular for many biomedical applications such as tissue engineering and regenerative therapies [45]. Thus, we further investigated the in vivo therapeutic effects of SSEA-1+ and SUSD2+ cells/chitosan on IUA rats by dividing the 20 rats into four groups: chitosan-saline solution group (chitosan group), SUSD2+ cells-laden chitosan group (SUSD2+ group), SSEA-1+ cells-laden chitosan group (SSEA-1+ group) and combination of SUSD2+ and SSEA-1+ cells-laden chitosan group (SSEA-1+  + SUSD2+ group). The left uteri of each rat received endometrial scraping while the contralateral right uteri were kept as control. The control group received no intervention. Schematic illustration of the experimental procedures was shown in Additional file 1: Figure S5f. As shown in Additional file 1: Figure S6, at day 14 post-surgery, severe hydrometra and structural deformation of the wound uteri were observed in all rats with only chitosan-saline solution treatment. In the SUSD2+ group, although one rat was found with a normal-seeming uterus, the others were still found to be abnormal with a mild to moderate hydrometra. By contrast, normal-seeming uteri were harvested in the SSEA-1+ group and the combination groups.

Restoration of the endometrium following different interventions was assessed by HE staining. The chitosan group and SUSD2+ group exhibited endometrial regeneration with few or no intact luminal structures (Fig. 6c). Endometrial gland numbers and endometrial thickness decreased markedly in the chitosan group (7.4 ± 2.1; 225 ± 13 μm) and SUSD2+ group (9.6 ± 1.5; 341 ± 39 μm), as compared with those in the control group (26.4 ± 2.7, P < 0.05; 441 ± 13 μm, P < 0.05, Fig. 6e,f). By contrast, the structure of the endometrium appeared well organized with more luminal intact glands in the stromal layer and the thickness of endometrium increased significantly in the SSEA-1+ group when compared with the normal uteri (22.6 ± 2.4, P < 0.05; 417 ± 17 μm, P > 0.05). Furthermore, these results were more pronounced in the combination group, as compared to the control group (27.4 ± 2.1, P > 0.05; 440 ± 29 μm, P > 0.05, Fig. 6c, e, f).

Fig. 6figure 6

The regenerative uterine horns after chitosan and cell therapy. a Fluorescent images of rats after transplantation with SSEA-1+ or SUSD2+ cells-laden chitosan. b Representative morphology of uteri following different treatments for 14 days. c HE staining of uteri after different treatments for 14 days. Blue arrow, intact endometrial glands; red arrow, incomplete endometrial gland. Inserted overview pictures are of lower magnification (scale bars, 500 µm); black squares are highly magnified regions (scale bars, 200 µm). d Collagen staining of uteri using Masson trichrome after different treatments for 14 days. Inserted overview pictures are of lower magnification (scale bars, 500 µm); black squares are highly magnified regions (scale bars, 200 µm). e–g Statistical analysis of the number of glands e, the endometrial thickness f and the percentages of collagen positive staining g after different treatments. LSD method was used for pairwise comparison between groups, n = 5, *P < 0.05, compared to the Control group; #P < 0.05, compared to the Chitosan group; ΔP < 0.05, compared to the SUSD2+ group; &P < 0.05, compared to the SSEA1+ group

To evaluate collagen remodeling in the reconstructed endometrium after transplanting SSEA-1+ and SUSD2+ cells/chitosan, Masson’s trichrome staining was performed (Fig. 6d) and fibrotic areas were analyzed quantitatively (Fig. 6g). At day 14 post-transplantation, more collagen deposition was observed in the chitosan group than that in the control group (13.4 ± 0.7% vs. 54.6 ± 3.3%, P < 0.05). A significantly decreased degree of fibrotic area was found in the SUSD2+ group when compared with the normal uteri (44.2 ± 4.2%, P < 0.05). Furthermore, a much greater reduction in fibrotic area was found in the SSEA-1+ group, as compared to the control group (23.4 ± 1.2%, P > 0.05). Remarkably, the collagen deposition of the combination group was similar to that of the control group, suggesting that the SSEA-1+ cells-laden chitosan injection significantly reduced fibrosis after the damage to the endometrium.

The survival of the transplanted cells in vivo is critical to improve the endometrial regeneration. Therefore, we next evaluated the effect of chitosan on the survival of the transplanted cells in vivo. We infected SSEA-1+ and SUSD2+ cells with luciferase by lentivirus for in vivo fluorescent imaging of transplanted cells. The representative morphology of SSEA-1+- and SUSD2+-mNeonGreen-Luc cells was shown in Additional file 1: Figure S4c,e. The fluorescent image of SSEA-1+ cells was presented in Additional file 1: Figure S4d. The activity of luciferase of SUSD2+ cells and SUSD + cells transduced with lentivirus in vitro was also detected by D-Luciferin sodium in vitro (Additional file

留言 (0)

沒有登入
gif