Dose and Timing Limitations in the Use of Bone Marrow-Derived Mesenchymal Stem Cells for the Treatment of Experimental Arthritis.

The recently recognized potential of mesenchymal stem cells (MSCs) to differentiate into a broad spectrum of tissues and to act as immune regulators beyond the barriers of embryonic germ layers and major histocombatibility comlex (MHC) restriction, has emerged intense research interest on their possible use in a broad spectrum of clinical entities. Although the immunoregulatory potential of MSCs has been shown to effectively control GvHD in several preclinical and clinical studies, their role in autoimmune diseases has not been extensively explored in animal models. The goal of this study was to investigate the in vitro effect of rat bone marrow-derived MSCs on cultured fibrobIast-like synoviocytes (FLS) and T-cells from the spleen after induction of adjuvant arthritis (AA) by FCA as well as their in vivo effect in a rat model of AA resembling human rheumatoid arthritis. MSCs were isolated from bone marrow and were characterized by CD45 negativity and CD54, CD29 positivity in FCM analysis. Differentiation assays were performed to confirm their adipogenic, osteogenic and chondrogenic potential. Culture of AA-FLS in the presence of supernatant from syngeneic (syng) or allogeneic (allo) MSCs at passage 2–3, reduced the AA-FLS (p<0.022) and the ConA-stimulated AA-T-cell (p=0.04) proliferation in a dose-dependent manner, as compared to AA-FLS or AA-T-cell proliferation in the absence of supernatant. Cell-to-cell contact by coculture of activated T-cells with syng or allo MSCs produced a stronger inhibition over the supernatant (p<0.0001), in all tested MSCs dilutions and even at the lowest MSCs :T-cell ratio of 0.05:1. The inhibitory effect of allo as compared to syng MSCs in activated AA T-cells, was stronger both by secreted agents (p=0.017) or by cell to cell contact (p=0.0001). In vivo, low doses of syng MSCs (0.5-5x10^⁵cell/recipient) administered iv, intrasplenic or intrabone marrow, at single or multiple infusions, didn’t significantly reduce the disease score of MSC-treated as compared to control rats. In contrast, repeated, higher dose (6x10^⁶cell/recipient), iv infusions of syng or allogeneic MSCs from male donors (Y+MSCs) to female recipients, before the onset of AA (d4 and d9 post AA induction) resulted in significantly lower arthritic scores when compared to control animals. MSC-treated animals preserved a rather normal joint architecture with focal synovial hyperplasia, limited pannus formation and without bone destruction or chondroplasia. In contrast, the joints of arthritic control rats, appeared with a thickened synovial membrane, erosive pannus and dense inflammatory cell infiltration, chondroplasia and osteoplasia. Reduced presence of CD3+, CD11b+, NF-kb+ cells and less intense angiogenesis (FVIII+cells) was demonstrated by immunohistochemistry in the synovium of transplanted rats as compared to the control group. No Y+MSCs were detected in the spleen, bone marrow or in cultured FLS from the synovial membrane at day30 post AA induction, by PCR (sry gene), immunohistochemistry (sry protein) or FISH (Y chromosome), suggesting that the observed benefit was mostly a result of immunomodulation not derived by MSCs homing to target tissues, or migration of MSCs to target tissues may have occured earlier. On the other hand, when the same cell dose was injected after the onset of arthritis (d13 and d20 post AA induction) no clinical benefit could be observed. Our data suggest that MSCs may represent a new therapeutic approach for autoimmune arthritis, however, due to dose and timing limitations in their use, further studies are needed to clinically exploit this potential.