Single cell‐derived clonally expanded mesenchymal progenitor cells from somatic cell nuclear transfer‐derived pluripotent stem cells ameliorate the endometrial function in the uterus of a murine model with Asherman’s syndrome

Abstract Objectives Because primary mesenchymal progenitor cells (adult‐MPCs) have various functions that depend on the tissue origin and donor, de novo MPCs from human pluripotent stem cells (hPSCs) would be required in regenerative medicine. However, the characteristics and function of MPCs derived from reprogrammed hPSCs have not been well studied. Thus, we show that functional MPCs can be successfully established from a single cell‐derived clonal expansion following MPC derivation from somatic cell nuclear transfer‐derived (SCNT)‐hPSCs, and these cells can serve as therapeutic contributors in an animal model of Asherman's syndrome (AS). Materials and methods We developed single cell‐derived clonal expansion following MPC derivation from SCNT‐hPSCs to offer a pure population and a higher biological activity. Additionally, we investigated the therapeutic effects of SCNT‐hPSC‐MPCs in model mice of Asherman's syndrome (AS), which is characterized by synechiae or fibrosis with endometrial injury. Results Their humoral effects in proliferating host cells encouraged angiogenesis and decreased pro‐inflammatory factors via a host‐dependent mechanism, resulting in reduction in AS. We also addressed that cellular activities such as the cell proliferation and population doubling of SCNT‐hPSC‐MPCs resemble those of human embryonic stem cell‐derived MPCs (hESC‐MPCs) and are much higher than those of adult‐MPCs. Conclusions Somatic cell nuclear transfer‐derived‐hPSCs‐MPCs could be an advanced therapeutic strategy for specific diseases in the field of regenerative medicine.


| INTRODUC TI ON
It is now accepted that human pluripotent stem cells (hPSCs), such as embryonic stem cells (ESCs) and reprogrammed stem cells from somatic cells using Yamanaka factors or somatic cell nuclear transfer (SCNT), will be available as cell sources in regenerative medicine. [1][2][3][4][5] Recently, we successfully established SCNT-derived human PSCs (SCNT-hPSCs) using patient fibroblasts 4,6 and have constantly developed the protocol to differentiate several lineage cells, including mesenchymal progenitor cells (MPCs), for application in cell therapy.
MPCs originate from various tissues and are regarded as promising therapeutic cell sources, findings that have already been shown in regenerative medicine and clinical trials. [7][8][9][10] Although the functions of MSCs are well addressed in diverse diseases, some perils of MPCs, such as rapid cell senescence in vitro, individual variations of donors, and non-replenishment, remain to be overcome. To resolve these limitations mentioned above, hPSCs have emerged as a valuable alternative, and their therapeutic effects have been continuously reported in regenerative medicine. Particularly, rare genetic diseases that have maternally inherited mitochondrial DNA mutations, such as sideroblast anaemia and Parkinson's disorders with chromatin decondensation, required novel tools, such as SCNT-hPSCs, to study their mechanism or develop clinical treatments. Because oocytes produce many proteins involved in the modulation of chromatin decondensation and mitochondria, SCNT-hPSCs may provide some clues to overcome the demerits in the concept of therapeutic cloning. [11][12][13][14][15] Additionally, SCNT-hPSCs with a low risk of immune rejection is now the subject of interest, likely to be further applied in clinical trials.
In this study, we developed single cell-derived clonal expansion following MPC derivation from SCNT-hPSCs to offer a pure population and a higher biological activity. Additionally, we investigated the therapeutic effects of SCNT-hPSC-MPCs in Asherman's syndrome (AS), which is characterized by synechiae or fibrosis with endometrial injury, often leading to infertility. A proliferation of endometrial cells should be proceeded to treat AS. However, treatment is difficult due to the intricate machinery of basal layer production or loss of stem cells in the endometrium. 16,17 In this regard, MPCs could be a possible candidate to treat fibrosis of AS because they contain paracrine properties that encompass both angiogenic and anti-inflammatory effects. In the present study, we addressed, for the first time, that single cell-derived clonally expanded SCNT-hPSC-MPCs (SCNT-hPSC-MPC-SCDs) showed increased cell numbers with stable population doublings (PDs) and no teratomas. These cells contributed to F I G U R E 1 Schematic diagram describing the protocol for differentiation of somatic nuclear transfer-derived human pluripotent stem cells into mesenchymal progenitor cells (SCNT-hPSC-MPCs). The formation of embryonic bodies (EBs) from PSCs was performed using SB431542, and then, EBs were reseeded in culture dishes for 28 d; the resultant cells are referred to as MPCs. Using MPCs detached in single-cell suspension, single cell-derived MPCs were clonally expanded in culture plates for at least 23 d. To examine the functions of the generated SCNT-hPSC-MPCs, their in vitro cellular activity was evaluated in confluent MPCs, and cells were applied to the Asherman's syndrome (AS) in vivo model. After cell transplantation in the uterus of the AS model, the assessment of the therapeutic effects was carried out at day 7 reduction in fibrosis in the AS model, resulting in implantation via promoted angiogenesis. Moreover, these cells show functional resemblance to hESC-MPCs. Thus, the successful generation of functional SCNT-hPSC-MPCs can promote therapeutic advances using the novel alternative in regenerative medicine.

| Cell proliferation assay
For the analysis of cell growth during long-term culture, cells were maintained in expansion culture as follows: 5 × 10 4 cells per well were plated in 12-well culture dishes, and the cells were counted in triplicate at least until 75 days after plating. The PDs and doubling time between cell passages were evaluated as previously described. 18 The single cell-derived MPCs were expanded and/or then combined to analyse their doubling time. and were used to induce a traumatized AS model. After anaesthesia by avertin, a vertical incision was made in the abdominal wall, and the uterus was exposed. A small incision was made in each uterine horn at the utero-tubal junction, and the horn was traumatized in a standardized fashion using a 30 gauge needle inserted through the lumen, rotated and withdrawn 10 times. 19 Forty-eight mice with AS were

| Statistical analysis
All results are presented as the mean ± SE. Statistical analyses were performed using the Mann-Whitney U test for comparisons between two groups and the Kruskal-Wallis ANOVA test for >2 groups.
Values of P < 0.05 were considered to denote statistical significance.
GraphPad Prism ver. 4 software (GraphPad Software, La Jolla, CA, USA) was used for statistical analysis.

| Effective differentiation and validation of MPCs from human PSCs
The generated CHA-hES15 and CHA-hNT5 cells from our laboratory were used in the present study ( Figure S1). 4,20 Because the TGF-beta inhibitor SB431542 is a known inducer that can reproduce mesodermal lineage cells, 21 it was mainly used to encourage the commitment of the mesodermal lineage during EB formation. To effectively acquire the MPC population, two phased protocols were explored: The immunophenotype of hPSC-MPCs was strictly addressed using CD29, CD44, CD90 and CD105. All markers for MPCs, according to our data, clearly showed high frequencies over 90%, even CD24, which was rarely expressed in MPCs (data not shown). CD24 negativity is regarded as a marker for MPCs with CD105-positive phenotype. 22,23 The expression of these markers for MPCs was definitively similar to that of BM-MPCs ( Figure 2B). Additionally, the generated hPSC-MPCs were differentiated into multilineage cells involved in adipogenesis, osteogenesis and chondrogenesis ( Figure S2). We

| Clonal expansion of MPCs from single MPCs and their characteristics: clonality, surface markers, euploidy and proliferation
To guarantee high purity and no teratomas of hPSC-MPCs from undifferentiated stem cells, we selected a single MPC from de- and SSEA4 were hardly detected in both hPSC-MPC-SCDs, while SSEA4 was expressed in hBM-MPCs at 46.0%, similar to that reported previously ( Figure 3D). 26 In the case of pluripotency genes, the expression levels of OCT4, SOX2 and NANOG were highly reduced during differentiation ( Figure 3E). The precursor genes for multilineage differentiated cells, precursor markers for each lineage, C/EBPα, RUNX2 and SOX9, were clearly expressed in both hPSC-MPC-SCDs. Maturation-related genes PPARγ for adipocytes, COL1 for osteocytes and COMP for chondrocytes were detected following differentiation ( Figure 3F). Next, we performed confirmative differentiation for lineage cells from SCNT-hPSC-MPC-SCDs, and we found strong adipogenesis, osteogenesis and chondrogenesis in both hPSC-MPC-SCDs ( Figure 3G).
In our first data, we found that cell proliferation and PDs of the SCNT-hPSC-MPC-SCD were remarkably discriminated from BM-MPCs ( Figure 3H and Figure S3). Furthermore, we investigated whether the capacity for clonal propagation and differentiation of hPSC-MSC-SCDs can synergistically increase when MPCs obtained from single cells were combined. To address this, we newly adopted F I G U R E 2 Sequential differentiation of hPSCs into MPCs. (A) The morphology of PSCs in terms of differentiation showed that fibroblastlike MPCs were successfully produced under proper culture conditions at day 16 and at passage 5. MPCs displayed a normal karyotype, suggesting no senescence due to differentiation. Magnification, ×40 (B) FACS analysis showed that the markers for MPCs, CD29, CD44, CD90 and CD105 were abundantly increased in CHA-hES15-MPCs and CHA-hNT5-MPCs, suggesting a phenotypical resemblance between ES cells and SCNT cells. (C) The cellular activities of both hPSC-MPCs revealed that exceeding proliferation by day and passages, and high population doublings were increased compared with those of adult BM-MPCs. Senescence of BM-MPCs occurred at approximately passage 12, while cell senescence in hPSC-MPCs occurred after passage 20 with a marked increase in cell number, suggesting the possibility of overcoming the weakness of adult BM-MPCs. The doubling times of hPSC-MPCs were significantly decreased. The data are presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with those of adult BM-MSCs (**P < 0.05). a fresh paradigm in this test. We formed two groups for analysis: Next, to examine whether these proliferative hPSC-MPCs were safe from teratomas, we carried out the teratoma formation test. hPSC-MPCs were injected into the testis of immunocompromised mice and were observed for 3 months. As shown in Figure 3I, all hPSC-MPC-SCDs did not produce teratomas ( Figure 3I). However, teratomas or teratomas with cysts were clearly formed in parent CHA-hES15 and CHA-hNT5 cells. Based on the present data, we found that our safe and productive protocol for MPC differentiation from SCNT-hPSCs can generate clinically applicable MPCs.

| Establishment of a murine model of AS and inhibition of pro-inflammatory factors in mice treated with SCNT-MPC-SCD
We next sought to elucidate whether transplanted SCNT-hPSC-MPC-SCDs can function as highly efficient therapeutic sources F I G U R E 3 Generation of stable single cell-derived clonally expanded SCNT-hPSC-MPCs (SCNT-hPSC-MPC-SCDs). (A) Schematic diagram of single cell-derived clonally expanded MPCs for the treatment of Asherman's syndrome (AS). (B) The morphology of MPCs derived from both hPSC-MPCs is similar to that of adult tissue-derived MPCs. After 5 passages, cells had a normal karyotype, implying no senescence. Bars, 100 μm. Magnification, ×20. (C) No difference between hESC-and SCNT-hPSC-MPCs was detected in the efficiency of seeding and doubling. When single MPCs were manually seeded into one well of a 96-well plate, the survival rate excluded from seeding damage and capacity for proliferation of survived single MPCs were very similar, suggesting the resemblance of SCNT-hPSC to hESCs. (D) FACS analysis showed that the markers for MPCs, CD29, CD44 and CD105 were highly increased in both hPSC-MPC-SCDs but not in stem cell or hematopoietic markers, suggesting phenotypical maturation. (E) Genes displaying stemness were rarely detected and were gradually decreased upon differentiation. (F) In multiple lineage differentiation, the PPARγ, C/EBPα for adipocytes, COL1, RUNX2 for osteocytes and COMP, SOX9 for chondrocytes in hPSC-MPCs were detected by differentiation. All transcripts were expressed in multilineage cell differentiation. (G) To confirm their maturation, hPSC-MPCs using standard differentiation conditions for adipogenesis, osteogenesis and chondrogenesis were differentiated; however, adipogenesis was not fully differentiated in MPC status, suggesting the not full maturation of hPSC-MPCs into adipocyte, osteocyte and chondrocyte. (H) The cellular activities of both hPSC-MPC-SCDs displayed that proliferation by day and passage and high population doublings were increased. PDs and cumulative cell number were presented in terms of single cell preparation at passage 5. The data are presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with adult BM-MPCs (**P < 0.05). (I) No teratomas in both SCNT-hPSC-MPC-SCDs were observed, implying safety in the clinic applications under pathologic condition. Similar with protocol by Alawadhi et al, 19 we established the AS mouse model and prepared stem cells were directly injected into uterus after incision in anaesthetized mice. No remarkable difference in H&E staining was detected among the wild-type, hBM-MPC-treated, both hPSC-MPCtreated groups, suggesting morphologic recovery into normal endometrium. However, the uteri of AS displayed a low frequency of small-sized glandular epithelium in the stroma and a low density of cells in the stroma, implying fibrosis (Figure 4Ai). In trichrome staining, a blue-coloured region presenting fibrosis was also displayed in the AS uterus. However, the MPC-treated groups seemed to be recovered, shown as a reddish colour in the stroma, implying The data are presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with the AS model (**P < 0.05, *P < 0.01). (C) Consistent with the levels of proteins, the PCR data showed that proinflammatory factors were highly increased in the AS model, but were significantly decreased by MSC therapy. The data are presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with the AS model. Two independent experiments were performed (**P < 0.05, *P < 0.01). cellular fidelity (Figure 4Aii). Additionally, aberrant expression of COL1A1 was detected in the AS uterus (Figure 4Aiii) likely due to abundant collagen fibres and cell proliferative arrest, similar no previous pathogenic events. 27 Additionally, single-layered luminal epithelial cells, which are arrested by AS induction and are located on surface of endometrium, were highly increased by stem cell therapy (Figure 4Aiv). The proliferation of luminal epithelial cells was shown by an enlarged panel with nuclei ( Figure 4Av). Fibrosis is closely involved with the failure of embryo implantation due to a defective endometrium. 28 To investigate whether transplanted F I G U R E 5 Impaired implantation in the AS uterus can be ameliorated in the hPSC-MPC-SCD-treated group by angiogenesis of host cells. (A) Assessment of implantation failure using the AS model. (B) The average number of pups is shown. Statistical analysis of (A), presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with implantation from the AS model. Graphs from 2 independent experiments are shown. N = 6 per group. (all group except AS = 6 horns, AS = 24 horns). (**P < 0.05, *P < 0.01). (C) Host murine cells upregulate the angiogenic factors, including Hgf, Igf, Ang-1 and Vegf-a. To determine whether the high cytokine levels were derived from human or murine cells, real-time PCR was performed using human-and mouse-specific primers. Most cytokines were murine, not human, and were highly sustained at day 14. Individual values were normalized to rPL7. The data for B and C are presented as the means ± SE from at least three experiments. Asterisks depict statistically significant differences compared with implantation from the AS model (**P < 0.05, *P < 0.01). (D) The capillary density was measured by counting Ki-67 + (green) and CD31 + cells (blue). DAPI was used to detect nuclei (red colours). Statistical analyses of the panels and values are presented as the mean ± SE from at least randomly selected 15 fields from four mouse heads per group. Two independent experiments were performed. N = 6 per group. (all group except AS = 6 horns, AS = 24 horns). (**P < 0.05, *P < 0.01) Magnification, ×40.
SCNT-hPSC-MPC-SCDs contributed to fibrosis by the suppression of pro-inflammatory factors in the uterus, Western blotting was carried out using pro-inflammatory-related factors TGFß1 and COL1A1. As shown in Figure  However, signals for X-and Y-chromosome probes in the AS-induced mouse uterus were not detected at day 7 after transplantation of MPCs, and it may suggest that transplanted human cells could be disappeared ( Figure S4). The suboptimal number of IS in the uterus with AS could be caused by insufficient angiogenesis because blood vessels play a role to deliver nutrients and growth litters. Moreover, it is well known that MPCs function to increase angiogenesis through paracrineangiogenic factors. 29,30 To prove this, we determined the capillary density in both hPSC-MPC-SCD-treated groups in the AS condition where the formation of new blood vessels is needed to prevent fibrosis. 31 Intriguingly, we found that angiogenic factors were highly increased in the uterus receiving MPCs, even 7 days after therapy.  Figure 5D, proliferation of endothelial cells was similar between CHA-hES15-MPCs and CHA-hNT5-MPCs but was not observed in the AS group (in sham: 12.9 ± 7.1%; in AS: 0 ± 0%; in hBM-MPCs:

| D ISCUSS I ON
In the present study, we first established single cell-derived clonally expanded MPCs from human pluripotent stem cells using Although the generation of hPSCs using SCNT has technical difficulties, somatic cell reprogramming technology is emerging as a promising tool due to a low risk of immune rejection 32 as well as the presentation of the full genome from a patient, who displays genomic aberrant symptom. Additionally, several reports continuously showed that iPSCs show detailed differences at the molecular levels; gene expression, genomic integrity and DNA methylation have been continuously reported for PSCs compared with those of ESCs. [33][34][35] Because SCNT technology can bring adult cells back to the embryonic stage. It is similar to the classical ESC. This rejuvenate advantage is expected to emphasize that SCNT-hPSC will play an important role in degenerative medicine as well as regenerative medicine. Also, PSC line cells derived from SCNT could be used for the study of their pathologic mechanisms and applied clinically. VEGFR-2 is main protein in blood endothelium, and its inhibition effectively suppresses non-alcoholic steatohepatitis accompanied by fibrosis. 36

CO N FLI C T S O F I NTE R E S T
The authors indicate no potential conflicts of interest.