Tolerance induction in experimental autoimmune encephalomyelitis using non-myeloablative hematopoietic gene therapy with autoantigen

Abstract

Experimental autoimmune encephalomyelitis (EAE) constitutes a paradigm of antigen (Ag)-specific T cell driven autoimmune diseases. In this study, we transferred bone marrow cells (BMCs) expressing an autoantigen (autoAg), the peptide 40-55 of the myelin oligodendrocytic glycoprotein (MOG(40-55)), to induce preventive and therapeutic immune tolerance in a murine EAE model. Transfer of BMC expressing MOG(40-55) (IiMOG-BMC) into partially myeloablated mice resulted in molecular chimerism and in robust protection from the experimental disease. In addition, in mice with established EAE, transfer of transduced BMC with or without partial myeloablation reduced the clinical and histopathological severity of the disease. In these experiments, improvement was observed even in the absence of engraftment of the transduced hematopoietic cells, probably rejected due to the previous immunization with the autoAg. Splenocytes from mice transplanted with IiMOG-BMC produced significantly higher amounts of interleukin (IL)-5 and IL-10 upon autoAg challenge than those of control animals, suggesting the participation of regulatory cells. Altogether, these results suggest that different tolerogenic mechanisms may be mediating the preventive and the therapeutic effects. In conclusion, this study demonstrates that a cell therapy using BMC expressing an autoAg can induce Ag-specific tolerance and ameliorate established EAE even in a nonmyeloablative setting.

Figure 1

Retroviral vectors. (a) Bicistronic retroviral vector containing the murine Ii in which the sequence encoding the CLIP region was replaced by that encoding the encephalitogenic peptide MOG_40–55 and the EGFP reporter gene placed after an IRES sequence. (b) The control vector encoding the WT murine Ii and EGFP. (c) A second retroviral control vector which only encodes EGFP. EGFP, enhanced green fluorescent protein; Ii, Invariant chain (CD74); IRES, Internal ribosome entry site; LTR, long terminal repeat; MOG, myelin oligodendrocyte glycoprotein.

Figure 2

Effect of preventive gene therapy on EAE and anti-MOG Ab production. (a) Clinical score. Animals transplanted with IiMOG-BMC before EAE induction were strongly protected from the disease, in comparison with controls. Data from two independent experiments are represented. EAE was evaluated using the following 6-point scale: 0 = no clinical signs; 0.5 = partial loss of tail tonus for three consecutive days; 1 = full tail paralysis; 2 = mild paraparesis of one or both hind limbs; 3 = paraplegia; 4 = tetraparesis; 5 = tetraplegia; 6 = death. (b) Anti-MOG_40–55 IgG reactivity. Serum samples were obtained at the time of killing (day 28 p.i.) and anti-MOG_40–55 Ab were analyzed by ELISA. Anti-MOG_40–55 Ab were more frequently found and their titers were significantly higher in EGFP-treated mice than in their IiMOG-treated counterparts.

Figure 3

Amelioration of EAE clinical course on therapeutic gene therapy. Mice were conditioned with busulfan and transplanted with BMC expressing either Ii or the IiMOG transgene, a median of 5 days after the clinical onset. After transplantation of BMC expressing IiMOG, mice experienced a clinical recovery from EAE, while this was not observed in the Ii-treated controls. Each chart represents an independent experiment. Arrows indicate the day of BMT.

Figure 4

Reduced spinal cord inflammation and axonal injury after liMOG-BMC transfer. The presence of macrophage/microglia (Mac-3⁺ cells), T cells (CD3⁺ cells), B cells (B220⁺ cells), and the extent of demyelination (luxol fast blue) and axonal damage (APP and SMI 32) in the central nervous system were analyzed. The study was performed in one representative experiment, which included PBS-immunized controls (experiment III, n = 9 in each group). The number of infiltrating CD3⁺ T lymphocytes was significantly reduced in IiMOG-treated mice compared with both Ii-treated and NT control mice (P < 0.01). Infiltrating B cells (B220⁺) were significantly reduced in IiMOG mice in comparison with NT controls. In addition, the demyelinating area, the acute axonal damage (measured by the amount of APP deposits/mm²) and functional abnormalities of axons (measured by the amount of SMI 32 deposits/mm²) were significantly reduced in the IiMOG-treated mice. Inflammation was restricted to the white matter, leading to substantial destruction of myelin (luxol fast blue) as well as axonal injury (APP) in control animals. No inflammation was seen in healthy controls receiving the IiMOG-BMC. Bar = 100 µm except for overview, 500 µm.

Figure 5

Splenocytes from IiMOG-treated mice secreted more IL-5 and IL-10 upon autoAg challenge while the humoral response was not affected. (a) Anti-MOG_40–55 IgG reactivity in mice sera. Both the prevalence and the mean levels of anti-MOG Ab were similar in all experimental groups. Dotted lines represent the mean optical density (OD) of the control sera plus 3 SD. (b) Splenocytes from IiMOG-treated mice secreted more IL-5 and IL-10 upon autoAg challenge in comparison with those of Ii controls. Culture supernatants from splenocytes were harvested 72 h after incubation with MOG_40–55 and the concentrations of GM-CSF, IFN-γ, IL-1α, IL-2, IL-4, IL-5, IL-6, IL-10, IL-17, and TNF-α were measured by a Multiplex assay. TGF-β levels were assessed by ELISA. Concentrations of IL-17 and IFN-γ, two important cytokines involved in the pathogenesis of EAE, were similar in both groups.

Figure 6

IiMOG-transduced BMC are cleared in mice with EAE. Donor and molecular chimerism were assessed at different time points after BMT. (a) Levels of donor chimerism (DCh) and molecular chimerism (MCh) in the PB (30 days post-BMT) and in the BM, spleen, and thymus (71 days post-BMT) of a representative experiment. Note the absence of molecular chimerism in the hematopoietic tissues of IiMOG-treated mice. (b) Levels of nontransduced donor chimerism (NTd DCh) and molecular chimerism (MCh) in the PB (30 days post-BMT) of mice pooled from three separate experiments. Note the absence of molecular chimerism in the IiMOG-treated group. (c) Dot plots correspond to analyses of total donor chimerism (CD45.1⁺ cell population) in the PB of a transplanted mouse (left) and analyses of donor and molecular chimerism (CD45.1⁺ EGFP⁺ cell population) in representative samples of Ii-treated (central) and IiMOG-treated mice (right), showing the lack of molecular chimerism in the latter.

Figure 7

Therapeutic benefit after non-myeloablative cell transfer. BMC transduced with the therapeutic vector IiMOG or the control vector Ii were infused into nonmyeloablated mice with EAE. Note the clinical improvement after the BMC infusion in the IiMOG-treated group, similar to that observed in the partially myeloablative experiments.