Cryopreserved mesenchymal stromal cells display impaired immunosuppressive properties as a result of heat-shock response and impaired interferon-γ licensing

Abstract

Human mesenchymal stromal cells (MSC) can suppress T-cell activation in vitro in an indoleamine 2,3-dioxygenase (IDO)-dependent manner. However, their clinical effects on immune ailments have been inconsistent, with a recent phase III study showing no benefit in acute graft-versus-host disease (GvHD). We here tested the hypothesis that the banked, cryopreserved MSC often used in clinical trials display biologic properties distinct from that of MSC in the log phase of growth typically examined in pre-clinical studies. In freshly thawed cryopreserved MSC derived from normal human volunteers, we observed that MSC up-regulate heat-shock proteins, are refractory to interferon (IFN)-γ-induced up-regulation of IDO, and are compromised in suppressing CD3/CD28-driven T cell proliferation. Immune suppressor activity, IFN-γ responsiveness and induction of IDO were fully restored following 24 h of MSC tissue culture post-thaw. These results highlight a possible cause for the inefficacy of MSC-based immunotherapy reported in clinical trials using cryopreserved MSC thawed immediately prior to infusion.

Figure 1

Immunosuppressive potential of freshly thawed human MSC compared with MSC in culture. (A) Cell viability analysis was performed on freshly thawed and cultured MSC using annexin V and PI labeling. (B) T-cell proliferation assays were performed using CFSE-labeled human PBMC activated with 0.4 μg/mL anti-CD3 and -CD28 antibodies and co-cultured for 4 days with or without MSC maintained in culture for 7 days or freshly thawed at a MSC:PBMC ratio of 1:3. Cell proliferation was determined by flow cytometry after gating lymphocytes on the forward and side scatter plot and measuring the percentage of CFSE^low T cells. (C) T-cell proliferation assay performed as in (B) using cultured MSC mixed with PFA-fixed MSC at a ratio of live:fixed of 2:1, 1:1, 1:2 or 100% fixed. A MSC:PBMC ratio of 1:3 was used. (D) T-cell proliferation assays performed as in (B) on two MSC donors freshly thawed and cultured for 1 and 7 days. Figures show representative results with means ± SD.

Figure 2

Protein and mRNA expression level of freshly thawed human MSC compared with MSC in culture. (A) IDO protein expression was analyzed by immunoblot on two human MSC donors freshly thawed and cultured for 1 and 7 days. IDO protein expression was induced by stimulating the MSC with 5 ng/mL recombinant (rh)IFN-γ for 24 h. Protein expression of Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. (B) Phosphorylated STAT-1 protein expression was analyzed by immunoblot on two human MSC donors freshly thawed and cultured for 1 and 7 days. STAT-1 phosphorylation was induced by stimulating the MSC with 5 ng/mL IFN-γ for 20 min. Protein expression of total STAT-1 was used as a loading control. (C) mRNA expression levels of (i) IDO, (ii) CCL2 and (iii) IL6 were measured by real-time qPCR in two MSC donors freshly thawed and cultured for 7 days. MSC were left untreated or stimulated with 5 ng/mL rhIFN-γ for 24 h. (D) mRNA expression level of Hsp27, Hsp47, Hsp56, Hsp70A, Hsp70B and Hsp90 by real-time qPCR performed on three MSC donors immediately after thawing and following a post-thaw recovery period of 4, 8, 16 and 24 h. Ribosomal 18S RNA was used as an internal control. Figures show representative results with means ± SD.