Human serum enhances the proliferative capacity and immunomodulatory property of MSCs derived from human placenta and umbilical cord

Abstract

Background: Mesenchymal stromal cells (MSCs) are considered potential candidates that hold great promise in the treatment of immune-related diseases. For therapeutic applications, it is necessary to isolate and expand MSCs with procedures complying with good manufacturing practice (GMP). Recent studies reported the use of human serum (HS) instead of fetal bovine serum (FBS) for the expansion of bone marrow-derived MSCs. Nevertheless, there are only limited data on HS as an alternative to FBS for the isolation and expansion of umbilical (UC-MSCs) and placenta-derived MSCs (PL-MSCs). In this study, we evaluate the effect of HS compared to FBS on the proliferative and immunosuppressive capacities of these MSCs.

Methods: PL-MSCs and UC-MSCs were isolated and cultured in HS- or FBS-supplemented media. The MSC characteristics, including morphology, immunophenotype, and differentiation ability, were verified. The proliferative and immunosuppressive capacities were also examined. In addition, the proliferative-enhancing factors in both sera were explored using proteomic analysis.

Results: PL-MSCs and UC-MSCs proliferated faster in HS-supplemented medium than in equivalent levels of FBS-supplemented medium. Adipogenic and osteogenic differentiations occurred at nearly identical levels in HS- and FBS-supplemented media. Interestingly, MSCs cultured in HS-supplemented medium had a similar immunosuppressive effect as MSCs cultured in FBS-supplemented medium. Proteomic analysis revealed that Con-A binding glycoproteins with a molecular weight > 100 kDa in FBS could significantly enhance MSC proliferation. In contrast, the proliferative enhancing factors in HS were found in the Con-A non-binding fraction and WGA binding fraction with a molecular weight > 100 kDa.

Conclusions: Taken together, our results suggest applications for the use of HS instead of FBS for the isolation and expansion of PL-MSCs and UC-MSCs for cell therapy in the future. Furthermore, this study identifies factors in HS that are responsible for its proliferative and immunosuppressive effects and might thus lead to the establishment of GMPs for the therapeutic use of MSCs.

Keywords: Con-A; Human serum; Immunosuppression; Mesenchymal stromal cell; Placenta; Umbilical cord; WGA.

Fig. 1

The characteristics of MSCs derived from human placenta (PL-MSCs) and umbilical cord (UC-MSCs) cultured in human serum (HS) compared to fetal bovine serum (FBS). a The spindle-shaped morphology of PL-MSCs and UC-MSCs. b Flow cytometry demonstrated the expression of typical MSC markers. The percentage of positive cells was not significantly different from each other whether they were cultured in HS- or FBS-supplemented medium. Data are presented as mean ± SEM. c The adipogenic and osteogenic differentiation potentials of PL-MSCs and UC-MSCs

Fig. 2

Growth kinetics of PL-MSCs (a, c, e, g) and UC-MSCs (b, d, f, h) cultured in MSC growth medium supplemented with 10% HS compared to PL-MSCs and UC-MSCs cultured in MSC growth medium supplemented with 10% FBS. Data from three independent experiments were presented as mean ± standard error of the means (SEM). *p < 0.05 compared with MSCs cultured in FBS

Fig. 3

The immunosuppressive effect of PL-MSCs (a) and UC-MSCs (b) on PHA-activated T cell proliferation. *p < 0.05 compared to PHA-activated T cells. *p < 0.05: significant difference compared to MNCs-PHA

Fig. 4

a–c Mean value of relative gene expression in MSCs co-cultured with PHA-activated T cells or IFN-γ. Data are presented as mean ± standard error of the means. *p < 0.05: significant difference compared to MSCs. ^#p < 0.05: significant difference compared to MSCs+MNCs+PHA. ⁺p < 0.05: significant difference compared to MSCs-FBS+IFN-γ

Fig. 5

a, b The expression level of IDO protein in PL-MSCs and UC-MSCs co-cultured with PHA-activated T cells or IFN-γ. *p < 0.05: significant difference compared to MSCs. #p < 0.05 compared to MSCs cultured in FBS-supplemented medium

Fig. 6

The effect of cytokine inhibition on the immunosuppressive effect of PL-MSCs (a) and UC-MSCs. (b). PHA-activated T cells were co-cultured with either MSCs-FBS or MSCs-HS in the presence or absence of the specific antagonists 1-MT, indomethacin, or L-NAME. Data are presented as mean ± SEM from three independent experiments. *p < 0.05: significant difference compared to MNCs+PHA+MSCs-FBS. ^#p < 0.05: significant difference compared to MNCs+PHA+MSCs-HS

Fig. 7

The effect of fractionated serum on MSC proliferation. PL-MSCs (a, c) and UC-MSCs (b, d) were cultured in DMEM+ 5% serum supplemented with < 3 kDa, 3–10 kDa, 10–30 kDa, 30–100 kDa, or > 100 kDa fractions. The unfractionated serum (10% serum) served as a positive control while 5% serum served as a negative control. Data are presented as mean ± SEM from three independent experiments. *p < 0.05 compared to 10% serum

Fig. 8

The effect of Con-A fractionated FBS (a, b) and HS (c, d) on MSC proliferation. MSCs were cultured in DMEM+ 5% serum supplemented with either Con-A binding proteins or non-Con-A binding proteins. MSCs cultured in DMEM+ 5% serum supplemented by the > 100 kDa fraction serum served as a control. The cells were harvested at days 3, 5, 7, and 10 to determine the cell numbers. Data are presented as mean ± SEM from three independent experiments. *p < 0.05 compared to control

Fig. 9

The effect of WGA fractionated HS on PL-MSC (a) and UC-MSC (b) proliferation. MSCs were cultured in DMEM+ 5% serum supplemented with either WGA binding proteins or non-WGA binding proteins. The > 100 kDa HS served as a control. The cells were harvested at days 3, 5, 7, and 10 to determine the cell numbers. Data are presented as mean ± SEM from three independent experiments. *p < 0.05 compared to control