Different therapeutic effects of cells derived from human amniotic membrane on premature ovarian aging depend on distinct cellular biological characteristics

BACKGROUND: Many reports have shown that various kinds of stem cells have the ability to recover premature ovarian aging (POA) function. Transplantation of human amniotic epithelial cells (hAECs) improves ovarian function damaged by chemotherapy in a mice model. Understanding of how to evaluate the distinct effects of adult stem cells in curing POA and how to choose stem cells in clinical application is lacking. METHODS: To build a different degrees of POA model, mice were administered different doses of cyclophosphamide: light dose (70 mg/kg, 2 weeks), medium dose (70 mg/kg, 1 week; 120 mg/kg, 1 week), and high dose (120 mg/kg, 2 weeks). Enzyme-linked immunosorbent assay detected serum levels of sex hormones, and hematoxylin and eosin staining allowed follicle counting and showed the ovarian tissue structure. DiIC(18)(5)-DS was employed to label human amniotic mesenchymal stem cells (hAMSCs) and hAECs for detecting the cellular retention time in ovaries by a live imaging system. Proliferation of human ovarian granule cells (ki67, AMH, FSHR, FOXL2, and CYP19A1) and immunological rejection of human peripheral blood mononuclear cells (CD4, CD11b, CD19, and CD56) were measured by flow cytometry (fluorescence-activated cell sorting (FACS)). Distinction of cellular biological characteristics between hAECs and hAMSCs was evaluated, such as collagen secretory level (collagen I, II, III, IV, and VI), telomerase activity, pluripotent markers tested by western blot, expression level of immune molecules (HLA-ABC and HLA-DR) analyzed by FACS, and cytokines (growth factors, chemotactic factors, apoptosis factors, and inflammatory factors) measured by a protein antibody array methodology. RESULTS: After hAMSCs and hAECs were transplanted into a different degrees of POA model, hAMSCs exerted better therapeutic activity on mouse ovarian function in the high-dose administration group, promoting the proliferation rate of ovarian granular cells from premature ovarian failure patients, but also provoking immune rejection. Meanwhile, our results showed that the biological characteristics of hAMSCs were superior to hAECs, but not to expression of immune molecules. CONCLUSIONS: These results suggest that hAMSCs are a more effective cell type to improve ovarian function than hAECs. Meanwhile, this distinct effect is attributable to cellular biological characteristics of hAMSCs (telomerase activity, expression level of pluripotent markers, cytokine and collagen secretion) that are superior to hAECs, except for immunological rejection. Sufficient consideration of cell properties is warranted to move forward to more effective clinical therapy. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13287-017-0613-3) contains supplementary material, which is available to authorized users.