Autoimmune diseases are incurable. We have hypothesized that these diseases can be cured by the transplantation of bone marrow (BM) stem cells that have been genetically engineered to express self-Ag. Here we have tested this hypothesis in experimental autoimmune encephalomyelitis (EAE) induced by the self-Ag myelin oligodendrocyte glycoprotein (MOG). We show that, in mice, transplantation of BM genetically modified to express MOG prevented the induction and progression of EAE, and combined with antecedent corticosteroid treatment, induced long-term remission of established disease. Mice remained resistant to EAE development upon subsequent rechallenge with MOG. Transfer of BM from these mice rendered recipients resistant to EAE. Splenocytes from these mice failed to proliferate or produce IL-17, IFN-γ, and GM-CSF in response to MOG35–55 peptide stimulation and they failed to produce MOG autoantibody. Mechanistically, we demonstrated in vivo reduction in development of CD4+ MOG35–55-specific thymocytes, indicative of clonal deletion with no evidence for selection of Ag-specific regulatory T cells. These findings validate our hypothesis that transplantation of genetically modified BM expressing disease-causative self-Ag provides a curative approach by clonal deletion of disease-causative self-reactive T cells.

Autoimmune diseases initiated by activation of pathogenic self-reactive T cells are incurable. While autologous and allogeneic bone marrow transplantation (BMT)4 (5) can alleviate these diseases, there are limitations to current approaches (1, 2, 3). Allogeneic BMT requires chronic immunosuppression for graft retention and prevention of graft-vs-host disease, and it is associated with higher mortality than autologous BMT (4, 5, 6). While autologous BMT has no risk of graft-vs-host disease, relapses are frequent (7) and whether it will finally lead to a cure has not been demonstrated (5, 8). We have suggested that a rational approach is to remove disease-causative self-reactive T cells before they leave the thymus by exploiting mechanisms that the immune system uses to maintain self-tolerance throughout life (1).

Thymus progenitors arising from the bone marrow (BM) continuously seed the thymus to develop into T cells and tolerogenic dendritic cells (DCs) (9). Exposure of developing T cells to self-Ags expressed by these bone marrow-derived DCs promotes deletion of T cells with high avidity for self-Ag (10, 11, 12, 13, 14, 15). Using transgenic approaches, we and others have shown that overexpression of self-Ag under the control of an MHC class II promoter prevented development of autoimmune diseases (16, 17), and that transfer of BM from these transgenic mice induced T cell tolerance and prevented disease (18, 19). These studies support our suggestion that exposure of developing T cells to self-Ag delivered through the BM compartment could be exploited as a strategy to induce immunological tolerance and permanently remit disease (1). The ability to genetically manipulate the BM compartment through retroviral transduction has been demonstrated in a range of models of autoimmunity and transplantation tolerance (20, 21, 22, 23, 24, 25). While transfer of transduced BM stem cells expressing self-Ag promotes Ag-specific tolerance, the mechanisms associated with tolerance induction remain poorly defined.

Here, we have tested our hypothesis using myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE) as a model of an autoimmune disease. We report that ex vivo genetic manipulation of BM to encode MOG prevented and reversed established disease in BM recipients. Mixed chimera experiments with MOG-specific T cells suggest that this is achieved by MOG-specific deletional tolerance without the generation of Ag-specific CD4+Foxp3+ regulatory T cells (Tregs).

Female C57BL/6 mice were from Monash Animal Services (Monash, Victoria, Australia). Generation of the 2D2 TCR transgenic (Tg) mice expressing MOG35–55-specific Vα3.2/Vβ11 TCR has previously been described (26, 27). Animals were housed in specific pathogen-fee environment (Monash Medical Centre Animal Facilities, Monash, Victoria, Australia). Syngeneic bone marrow donors were 5–6-wk-old female mice. All studies were performed with approval of the local animal ethics committee.

Mouse MOG cDNA was PCR cloned from brain mRNA using primer sets: sense, 5′-ATGGGCCTGTTTGTGGAGCTTCTCTTGGCCCAGC-3′; anti-sense, 5′-ACAACCATCACTCAAAAGGGG-3′. Mouse proinsulin II (ProII) cDNA was cloned as previously described (22). The MOG sequence was confirmed with published sequence and subcloned into retrovirus vector pMYs-IG (Fig. 1,A) to create the vector pMYs-MOG-IRES-EGFP (pMOG). Retrovirus vector pProII encoding the cDNA of ProII was generated as control. PCR were used to confirm the presence of cDNA in each of the retroviral vectors (Fig. 1 B). Recombinant retroviruses were generated and viral titers determined as previously described (22).

FIGURE 1.

A, Structure of bicistronic retroviral vector. Modified pMYs-IG retroviral vector, flanked by long terminal repeats (LTR) carrying MOG or ProII cDNA. MOG (∼800 bp) or ProII (∼900 bp) was subcloned upstream of the internal ribosomal entry site (IRES) and eGFP gene. B, PCR amplification was used to confirm the presence of MOG and ProII insert in pMYs-IG using corresponding primer sets with lane i (pMOG) and lane ii (pProII). C, pMOG-transduced NIH-3T3 (MOG-3T3) cells stained for MOG expression using anti-MOG mAb. MOG-3T3 cells showing (i) eGFP expression, (ii) positive staining for MOG protein (red), and (iii) Overlay of i and ii colocalization of eGFP and MOG expression. D, eGFP-sorted pProII-transduced NIH-3T3 cells were cultured for 7 days, after which supernatants (s/n) and cell lysates were collected and subjected to ELISA for proinsulin. E, Summary of experimental protocols.

FIGURE 1.

A, Structure of bicistronic retroviral vector. Modified pMYs-IG retroviral vector, flanked by long terminal repeats (LTR) carrying MOG or ProII cDNA. MOG (∼800 bp) or ProII (∼900 bp) was subcloned upstream of the internal ribosomal entry site (IRES) and eGFP gene. B, PCR amplification was used to confirm the presence of MOG and ProII insert in pMYs-IG using corresponding primer sets with lane i (pMOG) and lane ii (pProII). C, pMOG-transduced NIH-3T3 (MOG-3T3) cells stained for MOG expression using anti-MOG mAb. MOG-3T3 cells showing (i) eGFP expression, (ii) positive staining for MOG protein (red), and (iii) Overlay of i and ii colocalization of eGFP and MOG expression. D, eGFP-sorted pProII-transduced NIH-3T3 cells were cultured for 7 days, after which supernatants (s/n) and cell lysates were collected and subjected to ELISA for proinsulin. E, Summary of experimental protocols.

Close modal

NIH-3T3 cells were transduced with pMOG- or pProII-encoding retrovirus. Enhanced GFP (eGFP)-positive cells were sorted and cultured for 7 days. pMOG-infected NIH-3T3 cells (MOG-3T3) were fixed in 4% paraformaldehyde for 5 min at 4°C, followed by 10 min at room temperature (RT). Cells were washed with PBS, incubated with chilled methanol for 6 min at −20°C, washed and incubated with 1% swine serum for 15 min at RT, followed by monoclonal mouse anti-MOG Ab (clone 8-18C5) for 1 h at RT. After three further washes, secondary Ab (Alexa 594 nm goat anti-mouse IgG, Molecular Probes) was added and incubated for 1 h. Cells were washed and viewed under a fluorescence microscope (Olympus IX71 inverted research microscope). All steps were conducted in the dark.

For pProII-infected NIH-3T3 cells (ProII-3T3), 7-day cell culture supernatant and cell lysate of eGFP-sorted ProII-3T3 were analyzed for ProII production using an ELISA kit (rat/mouse insulin ELISA, Linco Research).

BM cells collected from donor mice were transduced with retrovirus as previously described (22). Donor mice were treated with 5-fluorouracil (140 mg/kg body weight) 3.5 days before BM harvest. BM cells were cultured in DMEM/10% FCS supplemented with recombinant cytokines: recombinant murine IL-6 (10 ng/ml, R&D Systems) and recombinant murine stem cell factor (50 ng/ml, R&D Systems). After 24 h, cells were transduced with retrovirus and cultured for a further 3–4 days before transferring sorted eGFP+ BM cells into recipient mice that had been subjected to 2 × 550 cGy total body irradiation. Between 40,000 and 50,000 eGFP+ cells were transferred into each recipient. On days 0, 7, 14, and 21, 0.25 mg of purified anti-CD4 (clone GK1.5) Ab was injected i.p. into BM recipients and untreated control groups to deplete residual CD4 T cells. BMT was conducted as follows: 1) For the prevention study, recipients were transferred with BM, allowed to reconstitute for 8–9 wk, followed by disease induction. 2) For the disease onset preclinical study, mice were first immunized with MOG35–55 followed by BMT on day 14. After 8–9 wk of reconstitution, recipients were reimmunized with MOG. 3) For the curative study, EAE mice with a clinical score of 2 were remitted by methylprednisolone treatment (45). Mice in remission (score 0) for 3 consecutive days were removed from the methylprednisolone treatment for 2 days before BMT. Following 8–9 wk of reconstitution, recipients were reimmunized with MOG35–55. In some experiments, mice received mixed BM consisting of retroviral-transduced BM and BM from 2D2 TCR Tg mice (30,000 cells at 70:30 transduced BM/2D2 ratio).

Methylprednisolone (Sigma-Aldrich) was administered at a dose of 10 mg/kg/day in the drinking water as previously described (28).

MOG-specific Ab was measured by ELISA as described (29, 30). Sera were collected at the end point of each experiment and tested at dilutions of 1/100, 1/300, and 1/900 in 96-well plates coated with 5 μg/ml of MOG35–55. Mean absorbance of triplicates was calculated with deduction of absorbance from noncoated wells.

A total of 200 μg of the encephalitogenic peptide MOG35–55 (GL Biochem) emulsified in CFA (Sigma-Aldrich) supplemented with 4 mg/ml Mycobacterium tuberculosis was injected s.c. into both femoral regions. Mice were immediately injected i.v. with 350 ng pertussis toxin (Sigma-Aldrich) and again 48 h later. Animals were monitored daily and neurological impairment was scored on an arbitrary clinical score: 0, no clinical sign; 1, limp tail; 2, limp tail and hind limb weakness; 3, severe hind limb paresis; 4, complete hind limb paresis; 5, moribund or death (29). As required by animal ethics, mice were euthanized upon reaching a clinical score of 3.

Brain and spinal cord were removed and fixed in 10% formalin. The cervical, thoracic, and lumbar segments of each spinal cord were embedded in paraffin. Between 8 and 16 sections, each at least 20 μm apart, were examined. The extent of inflammation, demyelination, and axonal loss was evaluated on tissue sections (5 μm) with H&E to assess inflammation, Luxol fast blue for demyelination, and Bielschowsky silver staining for axonal damage. Histological sections were scored blind. For inflammation, evaluation was performed using H&E-stained sections and scored as follows: 0, no inflammation; 1, cellular infiltrate only in the perivascular areas and meninges; 2, mild cellular infiltrate in parenchyma; 3, moderate cellular infiltrate in parenchyma; 4, severe cellular infiltrate in parenchyma. Myelin breakdown was assessed as pale staining with Luxol fast blue and scored as follows: 0, no demyelination; 1, mild demyelination; 2, moderate demyelination; 3, severe demyelination. Axonal loss was assessed as pale staining with Bielschowsky silver and scored as follows: 0, no axonal loss; 1, mild axonal loss; 2, moderate axonal loss; 3, severe axonal loss (29, 31).

Splenocytes were cultured at 500,000 cells per well in complete RPMI 1640 supplemented with 10% FCS, l-glutamine, penicillin, and streptomycin. Cells were stimulated with 1–100 μg/ml MOG35–55 peptide. Con A (10 μg/ml final concentration) was used as positive control and cells without peptide stimulation as background. After 72 h incubation at 37°C with 5% CO2, 20 μl of [3H]thymidine (2 μCi/well; Amersham) was added to each well and incubated for a further 24 h. Incorporated [3H]thymidine was determined with a Packard Tri-Carb 1900TR liquid scintillation analyzer. All cultures were in triplicates. Proliferation data are presented as a stimulation index. In some experiments, 72-h culture supernatants were collected for cytokine detection (Bender MedSystems).

Data are presented as means ± SEM. One-way ANOVA was used to analyze regulatory T cell population and histological scores. A Mann-Whitney test was used for cytokine and autoantibody analysis and Vα3.2 and Vβ11 expression levels. Fisher’s exact test was used for T cell proliferation assays. A value of p < 0.05 was considered significant.

Construction of retroviral vectors encoding MOG (pMOG) and control self-Ag ProII (pProII) in a single transcript with the reporter protein eGFP and confirmation of protein expression are shown in Fig. 1,A–D. MOG protein expression was detected in pMOG-transduced NIH-3T3 cells but not in ProII-transduced NIH-3T3 or nontransduced NIH-3T3 cells (data not shown). pProII-transduced NIH-3T3 cells produced and secreted ProII, which was readily detected in the supernatant. Therefore, subsequent cellular analysis was based on eGFP expression reflecting MOG/ProII expression. Three experimental scenarios were designed to test our hypothesis that transfer of MOG-transduced BM promotes Ag-specific tolerance and EAE resistance. The first was a prevention model in which BMT was performed before immunization with MOG to induce EAE. The second was a preclinical model of EAE induction before BMT, followed by rechallenge at 8–9 wk after BMT. The third was a curative model where mice with established EAE were treated with methylprednisolone to induce remission before BMT, followed by rechallenge with MOG at 8–9 wk after BMT (Fig. 1 E).

BM cells were transduced with either pMOG or pProII, sorted for eGFP expression to enhance chimerism levels, and transferred into lethally irradiated recipients (22). Similar chimerism levels were detected in primary and secondary lymphoid organs of pMOG and pProII groups as well as within individual cell populations from these organs (Fig. 2 and data not shown). Since BM-derived DCs are associated with induction of thymic self-tolerance, we confirmed that chimeric MHCII+CD11c+ DCs were present in the thymus (Fig. 2).

FIGURE 2.

Chimerism levels of pMOG and pProII mice determined by eGFP+ cells in thymus, spleen, and inguinal lymph nodes (left panels) and within lymphoid cell populations (right panels) from the prevention study. All data are means ± SEM.

FIGURE 2.

Chimerism levels of pMOG and pProII mice determined by eGFP+ cells in thymus, spleen, and inguinal lymph nodes (left panels) and within lymphoid cell populations (right panels) from the prevention study. All data are means ± SEM.

Close modal

BMT was conducted and mice were rested for 8 wk to allow for BM reconstitution before immunization with MOG peptide. All the mice in control groups (pProII, anti-CD4, and wild-type EAE (mice with EAE that had not received BMT)) and 67% (four of six) of mice in the normal BM (nBM) group developed EAE (Fig. 3, A and B, and Table I). The nBM group (Fig. 3,A) had a delayed response, but their clinical scores were not significantly different from those of the other control groups. The mean day of onset and mean maximum score were also not significantly different across all control groups (Table II). In striking contrast, none of the pMOG mice developed EAE (Fig. 3, A and B, and Table I).

FIGURE 3.

pMOG BMT prevented and reversed preclinical EAE. A, Prevention study. Clinical scores (means ± SEM) of MOG-immunized C57BL/6 mice that received nBM, BM transduced with retrovirus encoding MOG (pMOG), pProII, wild-type EAE mice, and mice with a course of anti-CD4 alone. B, Percentage of mice from A remaining disease-free over time following EAE induction. C, Preclinical study. Clinical scores (means ± SEM) of C57BL/6 mice that were immunized with MOG peptide and on day 14 postimmunization given nBM, pMOG, or pProII-transduced BM. Remission in cohorts is indicated. Mice were observed for relapse over 8–9 wk after BMT. On day 76 after the first immunization, disease-free mice were reimmunized with MOG peptide. D, Percentage of mice from C remaining disease-free over time. Point at which disease-free mice were reimmunized is indicated.

FIGURE 3.

pMOG BMT prevented and reversed preclinical EAE. A, Prevention study. Clinical scores (means ± SEM) of MOG-immunized C57BL/6 mice that received nBM, BM transduced with retrovirus encoding MOG (pMOG), pProII, wild-type EAE mice, and mice with a course of anti-CD4 alone. B, Percentage of mice from A remaining disease-free over time following EAE induction. C, Preclinical study. Clinical scores (means ± SEM) of C57BL/6 mice that were immunized with MOG peptide and on day 14 postimmunization given nBM, pMOG, or pProII-transduced BM. Remission in cohorts is indicated. Mice were observed for relapse over 8–9 wk after BMT. On day 76 after the first immunization, disease-free mice were reimmunized with MOG peptide. D, Percentage of mice from C remaining disease-free over time. Point at which disease-free mice were reimmunized is indicated.

Close modal
Table I.

Overall EAE incidence and histological features of spinal cord from different experimental protocolsa

ProtocolsExperimental GroupsOverall IncidenceHistological Score
H&ELuxol Fast BlueBielschowsky Silver Stain
Prevention pMOG BM 0/8 (0%) 
 pProII BM 9/9 (100%) 2.6 ± 0.2B 1.8 ± 0.3B 2.4 ± 0.2B 
 nBM 4/6 (67%) 2.3 ± 0.8A 2 ± 0.6B 1.7 ± 0.6 
 Anti-CD4 8/8 (100%) 2.3 ± 0.3B 1.5 ± 0.2A 2.3 ± 0.3B 
 Wild-type EAE 7/7 (100%) 3 ± 0.3C 2 ± 0.3B 2.4 ± 0.2B 
Preclinical pMOG BM 0/10 (0%) 
 pProII BM 8/9 (89%) 2.3 ± 0.4B 1.4 ± 0.3B 1.9 ± 0.3B 
 nBM 7/8 (88%) 1.9 ± 0.5A 1.6 ± 0.4B 2 ± 0.4B 
Curative pMOG BM 0/9 (0%) 0.6 ± 0.2D 0.7 ± 0.2D 
 pProII BM 5/5 (100%) 3.6 ± 0.4C 2.8 ± 0.2C 2.6 ± 0.2B 
 nBM 4/5 (80%) 2 ± 0.7 2 ± 0.3 2 ± 0.3 
Methylprenisolone Remitb NA 0.5 ± 0.2E 1.5 ± 0.2 1.6 ± 0.2 
 treatment Relapsec NA 3.1 ± 0.3C 2.3 ± 0.2B 2.6 ± 0.2C 
ProtocolsExperimental GroupsOverall IncidenceHistological Score
H&ELuxol Fast BlueBielschowsky Silver Stain
Prevention pMOG BM 0/8 (0%) 
 pProII BM 9/9 (100%) 2.6 ± 0.2B 1.8 ± 0.3B 2.4 ± 0.2B 
 nBM 4/6 (67%) 2.3 ± 0.8A 2 ± 0.6B 1.7 ± 0.6 
 Anti-CD4 8/8 (100%) 2.3 ± 0.3B 1.5 ± 0.2A 2.3 ± 0.3B 
 Wild-type EAE 7/7 (100%) 3 ± 0.3C 2 ± 0.3B 2.4 ± 0.2B 
Preclinical pMOG BM 0/10 (0%) 
 pProII BM 8/9 (89%) 2.3 ± 0.4B 1.4 ± 0.3B 1.9 ± 0.3B 
 nBM 7/8 (88%) 1.9 ± 0.5A 1.6 ± 0.4B 2 ± 0.4B 
Curative pMOG BM 0/9 (0%) 0.6 ± 0.2D 0.7 ± 0.2D 
 pProII BM 5/5 (100%) 3.6 ± 0.4C 2.8 ± 0.2C 2.6 ± 0.2B 
 nBM 4/5 (80%) 2 ± 0.7 2 ± 0.3 2 ± 0.3 
Methylprenisolone Remitb NA 0.5 ± 0.2E 1.5 ± 0.2 1.6 ± 0.2 
 treatment Relapsec NA 3.1 ± 0.3C 2.3 ± 0.2B 2.6 ± 0.2C 
a

Data are reported either as fraction of total number examined or mean ± SEM. Statistical difference compared to pMOG group at Ap < 0.05, Bp < 0.01, or Cp < 0.001. Significant difference compared to wild-type EAE at Dp < 0.01 or Ep < 0.001. NA indicates not applicable.

b

Mice (n = 8) were immunized with MOG peptide to induce EAE and treated with methylprednisolone when they reached a clinical score of 2 or above to induce remission. Mice were killed when remitted (3 consecutive days of clinical score of 0).

c

Mice (n = 12) were treated as for remit. Methylprednisolone treatment was withdrawn when mice remitted (3 consecutive days of clinical score 0) and allowed to relapse to a score of 3 before killing.

Table II.

Mean day of EAE onset and mean maximal score from different experimental protocols

ProtocolsExperimental GroupsMean Day of Onset: First ImmunizationaMean Maximal Score: First ImmunizationMean Day of Onset: Second ImmunizationbMean Maximal Score: Second Immunization
Prevention pMOG BM NA NA NA 
 pProII BM 23.1 ± 1.6 2.7 ± 0.2 NA NA 
 nBM 24.3 ± 3.0 1.9 ± 0.6 NA NA 
 Anti-CD4 19.6 ± 1.8 2.6 ± 0.2 NA NA 
 Wild-type EAE 22.3 ± 1.8 2.8 ± 0.2 NA NA 
Preclinical pMOG BM 14 ± 0 (n = 2) 0.2 ± 0.1 
 pProII BM 13.5 ± 0.2 (n = 2) 0.6 ± 0.3 16 ± 2.9 2.7 ± 0.5 
 nBM 12 ± 0 (n = 1) 0.3 ± 0.3 18.3 ± 2.3 2.5 ± 0.4 
Curative pMOG BM 21.4 ± 1.0 1.7 ± 0.4 
 pProII BM 25 ± 2.4 1.8 ± 0.7 22.5 ± 0.3 2.3 ± 0.1 
 nBM 20.6 ± 1.1 1.8 ± 0.5 24 ± 2.6 2 ± 0.5 
ProtocolsExperimental GroupsMean Day of Onset: First ImmunizationaMean Maximal Score: First ImmunizationMean Day of Onset: Second ImmunizationbMean Maximal Score: Second Immunization
Prevention pMOG BM NA NA NA 
 pProII BM 23.1 ± 1.6 2.7 ± 0.2 NA NA 
 nBM 24.3 ± 3.0 1.9 ± 0.6 NA NA 
 Anti-CD4 19.6 ± 1.8 2.6 ± 0.2 NA NA 
 Wild-type EAE 22.3 ± 1.8 2.8 ± 0.2 NA NA 
Preclinical pMOG BM 14 ± 0 (n = 2) 0.2 ± 0.1 
 pProII BM 13.5 ± 0.2 (n = 2) 0.6 ± 0.3 16 ± 2.9 2.7 ± 0.5 
 nBM 12 ± 0 (n = 1) 0.3 ± 0.3 18.3 ± 2.3 2.5 ± 0.4 
Curative pMOG BM 21.4 ± 1.0 1.7 ± 0.4 
 pProII BM 25 ± 2.4 1.8 ± 0.7 22.5 ± 0.3 2.3 ± 0.1 
 nBM 20.6 ± 1.1 1.8 ± 0.5 24 ± 2.6 2 ± 0.5 
a

With regard to the preclinical protocol for the first immunization, data are only included up to day 14 after the first immunization.

b

Data are only included from the time after the second immunization. NA indicates not applicable.

To determine whether BMT given at disease onset (preclinical study) can influence EAE development, BMT was performed 14 days after MOG immunization. The mean day of EAE onset was similar between pMOG, pProII, and nBM groups (Table II). At the time of BMT (day 14), 5 of 27 mice showed clinical scores of between 1 and 2.5, with 1 in the nBM group and 2 in the pMOG and pProII groups (Fig. 3, C and D). BMT preconditioning by irradiation remitted clinical signs in these five mice (Fig. 3, C and D).

During the 8–9 wk after BMT, three of eight (∼38%) mice from the nBM and six of nine (∼67%) mice from the pProII groups relapsed with EAE. By day 62 after BMT, EAE was observed in the three of eight mice in the nBM group and three of nine for the pProII group (three of six mice in the pProII group had remitted by this day) (Fig. 3,D). In striking contrast, none of the pMOG-mice developed EAE throughout this entire observation period (Fig. 3, C and D).

On day 62 after BMT, disease-free mice in all groups were reimmunized with MOG to determine whether they remained resistant or susceptible to further development of EAE. While none (0 of 10) of the pMOG group developed EAE (Fig. 3, C and D), EAE was induced in controls with a total of 8 of 9 pProII and 7 of 8 nBM mice developed disease at the completion of the experiment (Fig. 3,D and Table I).

The CNS of diseased mice in wild-type EAE, pProII, and nBM recipients revealed extensive inflammatory infiltrates of mononuclear cells and severe edema in the spinal cord (Fig. 4,A) and cerebellum. Luxol fast blue and Bielschowsky silver staining showed marked myelin loss and axonal injury around lesion sites (Fig. 4,A). In contrast, the spinal cord of pMOG mice remained free of inflammatory infiltrate, myelin loss, and axonal injury in the prevention and preclinical studies (Fig. 4, C and D, and Table I). The pMOG group was significantly different from the control groups across all histological parameters (Table I).

FIGURE 4.

Histology of spinal cord. Spinal cord sections were stained with H & E for lymphocyte infiltration, Luxol fast blue for myelin, and Bielschowsky silver stain for axon damage. A, Representative sections from pProII mice with EAE showing extensive inflammatory infiltrates, demyelination, and axonal loss. B, Normal mice showing no cellular infiltrates with intact myelin and axons. C, pMOG mice from prevention protocol and (D) pMOG mice from preclinical protocol showing features resembling normal spinal cord with no signs of cellular infiltrate, myelin, or axon loss. E, pMOG mice from curative protocol showing no signs of cellular infiltrate, as well as reduced demyelination and axon loss. Original magnification, ×20. Arrows indicate regions of cellular infiltrates, demyelination, or axon damage.

FIGURE 4.

Histology of spinal cord. Spinal cord sections were stained with H & E for lymphocyte infiltration, Luxol fast blue for myelin, and Bielschowsky silver stain for axon damage. A, Representative sections from pProII mice with EAE showing extensive inflammatory infiltrates, demyelination, and axonal loss. B, Normal mice showing no cellular infiltrates with intact myelin and axons. C, pMOG mice from prevention protocol and (D) pMOG mice from preclinical protocol showing features resembling normal spinal cord with no signs of cellular infiltrate, myelin, or axon loss. E, pMOG mice from curative protocol showing no signs of cellular infiltrate, as well as reduced demyelination and axon loss. Original magnification, ×20. Arrows indicate regions of cellular infiltrates, demyelination, or axon damage.

Close modal

Splenocytes of diseased mice from pProII and nBM control groups of the prevention (Fig. 5,A) and preclinical (Fig. 5,B) studies displayed dose-dependent proliferative responses to MOG peptide similar to those observed for splenocytes from wild-type EAE mice. In contrast, proliferation of pMOG mice was similar to that of naive mice. Cells from all groups proliferated in response to Con A (Fig. 5, A and B). Splenocytes from pMOG mice produced little or no IL-17, IFN-γ, or GM-CSF, in contrast to cytokine levels produced by pProII mice and wild-type EAE mice (Fig. 5,D). We also assessed Th2-associated cytokines IL-4 and IL-10 that might suggest a suppressive environment associated with the tolerance observed with pMOG mice and found that none of the groups produced significant amounts of Th2 cytokines (Fig. 5,D). These mice also produced significantly less anti-MOG Ab than did control pProII and nBM mice (Fig. 5, F and G).

FIGURE 5.

Splenocyte proliferation, cytokine profiles, and Ab response. Stimulation index (means ± SEM) of splenocytes from (A) prevention, (B) preclinical, and (C) curative protocols stimulated with MOG peptide. pMOG BMT resulted in significantly less splenocyte proliferation to MOG peptide. Proliferation to Con A is shown for comparison. D and E, Splenocyte cytokine production in response to MOG peptide challenge from (D) prevention and (E) curative protocols. Splenocytes were cultured for 72 h with indicated MOG peptide concentrations and supernatants collected for cytokine detection. Samples were assayed in triplicate. F–H, MOG-specific Ab response from sera samples demonstrated reduced anti-MOG reactivity in mice treated with pMOG BMT from (F) prevention, (G) preclinical, and (H) curative protocols. Each bar represents mean ± SEM from triplicates of specific absorbance corrected against background staining. ∗, p < 0.05 and ∗∗, p < 0.01 compared with pMOG group.

FIGURE 5.

Splenocyte proliferation, cytokine profiles, and Ab response. Stimulation index (means ± SEM) of splenocytes from (A) prevention, (B) preclinical, and (C) curative protocols stimulated with MOG peptide. pMOG BMT resulted in significantly less splenocyte proliferation to MOG peptide. Proliferation to Con A is shown for comparison. D and E, Splenocyte cytokine production in response to MOG peptide challenge from (D) prevention and (E) curative protocols. Splenocytes were cultured for 72 h with indicated MOG peptide concentrations and supernatants collected for cytokine detection. Samples were assayed in triplicate. F–H, MOG-specific Ab response from sera samples demonstrated reduced anti-MOG reactivity in mice treated with pMOG BMT from (F) prevention, (G) preclinical, and (H) curative protocols. Each bar represents mean ± SEM from triplicates of specific absorbance corrected against background staining. ∗, p < 0.05 and ∗∗, p < 0.01 compared with pMOG group.

Close modal

To investigate mechanisms of tolerance in pMOG mice, we used the 2D2 TCR Tg mouse with skewed CD4 T cells encoding TCR (Vα3.2 and Vβ11 chains) specific for MOG35–55 peptide in the context of H-2b (26, 27). We performed a “mixed” BMT by transferring pMOG BM with 2D2 Tg BM (70:30 ratio) into irradiated recipients. Eight weeks after BMT, thymocytes from mixed chimeras were isolated and stained for CD4, CD8, and Vα3.2 or Vβ11 to identify 2D2 T cells. Naive 2D2 Tg mice and chimeras generated by mixing 2D2 BM and pProII BM were used as controls. We found a significant decrease in transition of Vα3.2 and Vβ11 single-positive CD4+ T cells from the CD4+CD8+ double-positive population in pMOG + 2D2 BM chimeras compared with pProII + 2D2 BM chimeras or naive 2D2 TCR Tg mice (Fig. 6, A and B). The reduction in CD4+ MOG-specific 2D2 T cells was accompanied by a significant increase in apoptosis of 2D2 thymocytes in pMOG + 2D2 BM chimeras (Fig. 6,C), which is consistent with deletion of MOG-specific T cells (Fig. 6, A and B).

FIGURE 6.

Thymic deletion of MOG35–55-specific T cells in mice transferred with pMOG BM. Mixed BM chimeras generated by mixing 2D2 TCR BM with BM transduced with pMOG or pProII. Eight weeks after BMT, thymocytes were stained with CD4, CD8, and 2D2 TCR-specific Vα3.2 or Vβ11 Abs. A, Representative flow cytometry plots showing expression profiles of CD4 and CD8 in Vα3.2- and Vβ11-positive cells in pMOG and pProII transplanted and 2D2 wild-type mice. B, Summary of data from A showing reduction in Vα3.2- and Vβ11-positive thymocytes transitioning from CD4+CD8+ double-positive (DP) to CD4+ single-positive T cells in the pMOG BM group compared with the pProII and wild-type 2D2 profile. Data are means ± SEM. C, Percentages of Vα3.2- and Vβ11-expressing thymocytes undergoing apoptosis as determined by annexin V and 7-aminoactinomycin D staining. Data are means ± SEM. D, Percentages of CD4 and Foxp3 double-positive cells in total CD4+ splenocytes from prevention, preclinical, and curative protocols. Data are means ± SEM. E, Thymocytes and splenocytes of mixed chimeras from A showing proportion of CD4, Vα3.2, and Foxp3 triple-positive cells in pMOG (n = 12), pProII (n = 6), and wild-type C57BL/6 (n = 5) groups. Data are means ± SEM.

FIGURE 6.

Thymic deletion of MOG35–55-specific T cells in mice transferred with pMOG BM. Mixed BM chimeras generated by mixing 2D2 TCR BM with BM transduced with pMOG or pProII. Eight weeks after BMT, thymocytes were stained with CD4, CD8, and 2D2 TCR-specific Vα3.2 or Vβ11 Abs. A, Representative flow cytometry plots showing expression profiles of CD4 and CD8 in Vα3.2- and Vβ11-positive cells in pMOG and pProII transplanted and 2D2 wild-type mice. B, Summary of data from A showing reduction in Vα3.2- and Vβ11-positive thymocytes transitioning from CD4+CD8+ double-positive (DP) to CD4+ single-positive T cells in the pMOG BM group compared with the pProII and wild-type 2D2 profile. Data are means ± SEM. C, Percentages of Vα3.2- and Vβ11-expressing thymocytes undergoing apoptosis as determined by annexin V and 7-aminoactinomycin D staining. Data are means ± SEM. D, Percentages of CD4 and Foxp3 double-positive cells in total CD4+ splenocytes from prevention, preclinical, and curative protocols. Data are means ± SEM. E, Thymocytes and splenocytes of mixed chimeras from A showing proportion of CD4, Vα3.2, and Foxp3 triple-positive cells in pMOG (n = 12), pProII (n = 6), and wild-type C57BL/6 (n = 5) groups. Data are means ± SEM.

Close modal

We did not observe any significant changes in the proportion of CD4+Foxp3+ Tregs in the splenocyte population across the different animal groups from the prevention and preclinical (Fig. 6,D) protocols. In the mixed chimera study with MOG35–55-specific 2D2 T cells, there was no significant increase in CD4 single-positive thymocytes or splenocytes expressing Vα3.2+, Foxp3+, indicating that 2D2 thymocytes were not selected to become Foxp3+ Tregs (Fig. 6 E).

We conducted two experiments to address whether our strategy targeted hemopoietic stem cells. First, mice that had received pMOG or pProII BM were rested for 7 mo and then immunized for EAE. Mice that received pMOG BM were resistant to EAE (none of four) while 100% (six of six) of pProII mice developed EAE (Fig. 7,A). Second, sorted eGFP+ BM cells were isolated from mice at the end of the prevention study, transferred into naive, irradiated recipients, and allowed 8 wk of reconstitution before immunization with MOG peptide. All mice that received pMOG BM were resistant to EAE (none of three) while 100% of pProII-mice (three of three) developed EAE (Fig. 7 B), indicating that the ability to promote Ag-specific tolerance was transferable through the transduced BM cells.

FIGURE 7.

Transfer of tolerance and disease resistance from pMOG-treated mice. A, Mice received pMOG or pProII BM were rested for 7 mo before MOG immunization and EAE development was monitored. B, eGFP+ BM cells were sorted from pMOG and pProII chimeras and transferred into naive, irradiated recipients. After 8 wk, mice were immunized with MOG peptide and EAE development was monitored. A group of wild-type mice with EAE was included for comparison. All data are means ± SEM.

FIGURE 7.

Transfer of tolerance and disease resistance from pMOG-treated mice. A, Mice received pMOG or pProII BM were rested for 7 mo before MOG immunization and EAE development was monitored. B, eGFP+ BM cells were sorted from pMOG and pProII chimeras and transferred into naive, irradiated recipients. After 8 wk, mice were immunized with MOG peptide and EAE development was monitored. A group of wild-type mice with EAE was included for comparison. All data are means ± SEM.

Close modal

We next explored whether this strategy could be used to reverse established disease. We asked whether remission induced with a course of methylprednisolone (45) followed by pMOG BMT could cure established EAE. A cohort of mice was immunized, allowed to develop EAE to a clinical score of 2, and treated with methylprednisolone to promote remission. Mice free of signs of EAE (score 0) for 3 consecutive days were withdrawn from methylprednisolone treatment and 2 days later subjected to BMT. After the 8–9-wk reconstitution period, none of the nine pMOG mice developed signs of EAE, while one mouse from each of the nBM (one of five) and pProII (one of five) groups relapsed (Fig. 8). The diseased mouse from the nBM group developed a score of 3 and was killed due to ethical requirements, while the diseased mouse from the pProII group remitted by day 93 and was included in the reimmunization process. At day 93, disease-free mice from pMOG, pProII, and nBM groups (pMOG, n = 9; pProII, n = 5; nBM, n = 4) were reimmunized with MOG peptide and observed for a further 40 days. Overall, none of the pMOG group (none of nine) developed EAE, while 100% (five of five) of pProII and 80% (four of five) of nBM groups developed EAE (Fig. 8,B and Table I). The mean day of onset after the second immunization was similar between pProII and nBM groups (Table II).

FIGURE 8.

pMOG BMT together with methylprednisolone promotes long-term remission of established EAE. A, C57BL/6 mice with disease score of 2 or more were treated with methylprednisolone until they remitted (see Materials and Methods) and then received pProII, nBM (upper panel), or pMOG (lower panel) BMT (day 35; BMT, arrows). Mice were monitored for relapse over 58 days. At day 93, disease-free mice were reimmunized with MOG peptide (arrows). Data are presented as mean clinical scores ± SEM. B, Percentage of mice from A remaining disease-free after BMT and immunizations.

FIGURE 8.

pMOG BMT together with methylprednisolone promotes long-term remission of established EAE. A, C57BL/6 mice with disease score of 2 or more were treated with methylprednisolone until they remitted (see Materials and Methods) and then received pProII, nBM (upper panel), or pMOG (lower panel) BMT (day 35; BMT, arrows). Mice were monitored for relapse over 58 days. At day 93, disease-free mice were reimmunized with MOG peptide (arrows). Data are presented as mean clinical scores ± SEM. B, Percentage of mice from A remaining disease-free after BMT and immunizations.

Close modal

Histological analysis revealed methylprednisolone-induced remission was associated with marked reduction in inflammatory infiltrate in the spinal cord that was reversed upon methylprednisolone withdrawal and subsequent relapse (Table I). The effect on demyelination and axonal damage improved marginally compared with wild-type EAE but was not as profound as the effect on the inflammatory infiltrate (Table I). Following sequential methylprednisolone treatment and BMT, pMOG mice were devoid of any CNS inflammatory infiltrate, while the nBM and pProII groups displayed significant inflammatory infiltrate (Table I and Fig. 4,E). pMOG-mice showed reduced demyelination and axonal damage compared with the pProII group (Table I).

Splenocytes from all groups proliferated similarly to Con A, while splenocytes of the pMOG group proliferated significantly less than the nBM control group in response to MOG peptide (Fig. 5,C). pMOG mice produced little or no IL-17, IFN-γ, or GM-CSF (Fig. 5,E) in response to MOG peptide compared with the pProII group and produced significantly less anti-MOG Ab than did the pProII and nBM mice (Fig. 5,H). None of the groups produced significant amount of Th2 cytokines (Fig. 5,E). The levels of CD4+Foxp3+ splenocytes were similar across the pMOG, pProII, and nBM groups (Fig. 6 D).

Since autoimmunity is due to the failure of self-tolerance, we hypothesized that harnessing the natural process of central tolerance by using genetically manipulated autologous BMT may promote Ag-specific tolerance (1). Our studies using the MOG-induced EAE mouse model of autoimmune disease support the notion that transplantation with normal syngeneic (autologous-equivalent) BM is not curative in the three different experimental settings we examined. While remission was induced in 60–80% of these mice, reimmunization with MOG promptly precipitated disease. However, 100% of mice that received pMOG BM were rendered tolerant and completely resistant to MOG35–55-induced EAE, even after rechallenge with MOG. Thus, simple “rebooting” of the immune system with nonmanipulated BM may have limited benefit given that there is no inherent process to promote immune tolerance to the target Ag. For instance, recent clinical trials of autologous hematopoietic stem cell transplantation in multiple sclerosis showed <50% of patients with disease-free survival (18, 32, 33), with one study reporting only 7.1% of patients remaining disease-free after 6 years follow-up (18).

Our mixed chimera experiments using 2D2 Tg mice suggest that the complete resistance of these mice to MOG-induced EAE is due to deletion of CD4+ MOG-reactive T cells in the thymus. We found a significant decrease in transition of MOG-specific single-positive CD4+ transgenic T cells from the double-positive CD4+CD8+ stage, suggesting that these cells were deleted during this developmental stage. Selection of MOG-specific Foxp3+ natural Tregs did not appear to be a mechanism of tolerance and disease resistance because we did not find an increase in this population. Our observations are consistent with a recent study showing that Ag presentation by medullary thymic epithelial cells is responsible for the generation of natural Treg generation, whereas DCs mediate thymocyte deletion (34). The findings relating to DC-mediated deletion are consistent with an earlier report that these DCs mediate negative but not positive selection (10). Indeed, in all three experimental protocols, we found detectable molecular chimerism in MHCII+CD11c+ DCs. Furthermore, while MOG expression in the thymus is primarily in medullary thymic epithelial cells driven at least in part by the autoimmune regulator AIRE (35), it is clearly insufficient to establish solid T cell tolerance to MOG, as susceptible mice such as the C57BL/6 strain develop EAE upon immunization with MOG. This is highlighted by persistence of MOG-specific Vα and Vβ T cell repertoires in MOG-expressing mice as well as in MOG-deficient mice (36). The findings we report herein suggest that the additional expression of self-Ag through the BM compartment can establish deletion of self-reactive T cells and thereby promote tolerance.

High levels of molecular chimerism do not appear to be required for tolerance induction. Studies in transplantation tolerance reported that 10–15% chimerism (37) and ∼5% chimerism in the thymic CD11c population are sufficient to induce tolerance (38). Even chimerism levels of <1% (microchimerism) can maintain cytotoxic T cell tolerance (39). These findings suggest that if chimerism can be induced in recipients, there is a good chance of promoting Ag-specific tolerance. An underlying feature of our strategy is that Ag is constantly expressed to maintain the tolerogenic environment. Notably, we observed tolerance in chimeric mice 7 mo after BMT, and we could also transfer chimerism and disease resistance to naive mice through the BM compartment. Persistent Ag expression appears paramount for success of this strategy and is in line with a recent finding for transplantation tolerance that long-term Ag expression is required to maintain tolerance (40).

In sharp contrast to what we observed in all of the different control groups, splenocytes from pMOG mice failed to proliferate in response to stimulation by MOG peptide in a dose-dependent manner. These mice also failed to produce anti-MOG autoantibody and IL-17, IFN-γ, and GM-CSF, which have been identified as major inflammatory cytokines in EAE (31, 41, 42, 43). These findings are consistent with our observations in the mixed chimera experiments that tolerance in pMOG mice is associated with deletion of MOG35–55-reactive T cells. Failure to produce anti-MOG Ab may be a consequence of the lack of cognate T cell help (44), although direct deletion of MOG-reactive B cells cannot be excluded.

It was apparent from both the preclinical and curative EAE studies that BMT with either nBM or pProII has some beneficial effect in most mice. However, when the rebooted immune system was challenged with the same disease-inducing antigenic trigger, most mice responded with disease development. In extrapolating to the human example, a similar situation may apply to patients receiving autologous BMT. While remission and benefit may be observed in the initial period, our findings suggest that the immune system maintains the capacity to recognize and respond to self-Ags associated with disease that ultimately may result in relapse. From our findings the simple modification of this process to include induced expression of self-Ag through the BM compartment had a profound effect on tolerance induction and disease development. We suggest that this approach offers a logical and feasible strategy with potential for clinical application in settings where the causative self-Ags are known.

We thank J. Forbes and F. Yap with the ProII protein detection, Z. Nasa, D. Oh, and E. Tim for technical support, K. Murphy for cloning the mouse MOG cDNA, and V. Kuchroo for providing the 2D2 Tg mice.

The authors have no financial conflicts of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by program grants from the National Health and Medical Research Council of Australia, Cure MS (Australia), the Baker Foundation, and the National Multiple Sclerosis Society of New York.

4

Abbreviations used in this paper: BMT, bone marrow transplantation; BM, bone marrow; DC, dendritic cell; MOG, myelin oligodendrocyte glycoprotein; EAE, experimental autoimmune encephalomyelitis; Treg, regulatory T cell; Tg, transgenic; ProII, proinsulin II; pMOG, retrovirus encoding MOG; pProII, retrovirus encoding ProII; eGFP, enhanced GFP; RT, room temperature; nBM, normal bone marrow.

1
Alderuccio, F., K. Murphy, B. H. Toh.
2003
. Stem cells engineered to express self-antigen to treat autoimmunity.
Trends Immunol.
24
:
176
-180.
2
Chan, J., C. Siatskas, J. Field, B.-H. Toh, F. Alderuccio.
2006
. Mechanisms and applications of stem cell gene therapy in autoimmunity.
Drug Discov. Today: Dis. Mech.
3
:
219
-223.
3
Bagley, J., J. Iacomini.
2003
. Gene therapy progress and prospects: gene therapy in organ transplantation.
Gene Ther.
10
:
605
-611.
4
Armitage, J. O..
1994
. Bone marrow transplantation.
N. Engl. J. Med.
330
:
827
-838.
5
Marmont, A. M..
2008
. Will hematopoietic stem cell transplantation cure human autoimmune diseases?.
J. Autoimmun.
30
:
145
-150.
6
Burt, R. K., A. Testor, R. Craig, B. Cohen, R. Suffit, W. Barr.
2008
. Hematopoietic stem cell transplantation for autoimmune diseases: what have we learned?.
J. Autoimmun.
30
:
116
-120.
7
Gualandi, F., B. Bruno, M. T. Van Lint, S. Luchetti, A. Uccelli, E. Capello, G. L. Mancardi, A. Bacigalupo, A. Marmont.
2007
. Autologous stem cell transplantation for severe autoimmune diseases: a 10-year experience.
Ann. NY Acad. Sci.
1110
:
455
-464.
8
van Gelder, M., D. W. van Bekkum.
1995
. Treatment of relapsing experimental autoimmune encephalomyelitis in rats with allogeneic bone marrow transplantation from a resistant strain.
Bone Marrow Transplant.
16
:
343
-351.
9
Lai, A. Y., M. Kondo.
2007
. Identification of a bone marrow precursor of the earliest thymocytes in adult mouse.
Proc. Natl. Acad. Sci. USA
104
:
6311
-6316.
10
Brocker, T., M. Riedinger, K. Karjalainen.
1997
. Targeted expression of major histocompatibility complex (MHC) class II molecules demonstrates that dendritic cells can induce negative but not positive selection of thymocytes in vivo.
J. Exp. Med.
185
:
541
-550.
11
Goodnow, C. C., J. Sprent, B. F. de St. Groth, C. G. Vinuesa.
2005
. Cellular and genetic mechanisms of self tolerance and autoimmunity.
Nature
435
:
590
-597.
12
Kubach, J., C. Becker, E. Schmitt, K. Steinbrink, E. Huter, A. Tuettenberg, H. Jonuleit.
2005
. Dendritic cells: sentinels of immunity and tolerance.
Int. J. Hematol.
81
:
197
-203.
13
Kyewski, B., J. Derbinski.
2004
. Self-representation in the thymus: an extended view.
Nat. Rev. Immunol.
4
:
688
-698.
14
Sprent, J., S. R. Webb.
1995
. Intrathymic and extrathymic clonal deletion of T cells.
Curr. Opin. Immunol.
7
:
196
-205.
15
Gallegos, A. M., M. J. Bevan.
2004
. Central tolerance to tissue-specific antigens mediated by direct and indirect antigen presentation.
J. Exp. Med.
200
:
1039
-1049.
16
Alderuccio, F., B. H. Toh, S. S. Tan, P. A. Gleeson, I. R. van Driel.
1993
. An autoimmune disease with multiple molecular targets abrogated by the transgenic expression of a single autoantigen in the thymus.
J. Exp. Med.
178
:
419
-426.
17
French, M. B., J. Allison, D. S. Cram, H. E. Thomas, M. Dempsey-Collier, A. Silva, H. M. Georgiou, T. W. Kay, L. C. Harrison, A. M. Lew.
1997
. Transgenic expression of mouse proinsulin II prevents diabetes in nonobese diabetic mice.
Diabetes
46
:
34
-39.
18
Saiz, A., Y. Blanco, J. Berenguer, M. Gomez, E. Carreras, T. Arbizu, F. Graus.
2008
. Clinical outcome 6 years after autologous hematopoietic stem cell transplantation in multiple sclerosis (In Spanish).
Neurologia
23
:
405
-407.
19
Steptoe, R. J., J. M. Ritchie, L. C. Harrison.
2003
. Transfer of hematopoietic stem cells encoding autoantigen prevents autoimmune diabetes.
J. Clin. Invest.
111
:
1357
-1363.
20
Ally, B. A., T. S. Hawley, K. J. McKall-Faienza, T. M. Kundig, S. U. Oehen, H. Pircher, R. G. Hawley, P. S. Ohashi.
1995
. Prevention of autoimmune disease by retroviral-mediated gene therapy.
J. Immunol.
155
:
5404
-5408.
21
Bagley, J., C. Tian, D. H. Sachs, J. Iacomini.
2002
. Induction of T-cell tolerance to an MHC class I alloantigen by gene therapy.
Blood
99
:
4394
-4399.
22
Chan, J., W. Clements, J. Field, Z. Nasa, P. Lock, F. Yap, B. H. Toh, F. Alderuccio.
2006
. Transplantation of bone marrow genetically engineered to express proinsulin II protects against autoimmune insulitis in NOD mice.
J. Gene Med.
8
:
1281
-1290.
23
Sykes, M., D. H. Sachs, A. W. Nienhuis, D. A. Pearson, A. D. Moulton, D. M. Bodine.
1993
. Specific prolongation of skin graft survival following retroviral transduction of bone marrow with an allogeneic major histocompatibility complex gene.
Transplantation
55
:
197
-202.
24
Xu, B., P. Haviernik, L. A. Wolfraim, K. D. Bunting, D. W. Scott.
2006
. Bone marrow transplantation combined with gene therapy to induce antigen-specific tolerance and ameliorate EAE.
Mol. Ther.
13
:
42
-48.
25
Zambidis, E. T., A. Kurup, D. W. Scott.
1997
. Genetically transferred central and peripheral immune tolerance via retroviral-mediated expression of immunogenic epitopes in hematopoietic progenitors or peripheral B lymphocytes.
Mol. Med.
3
:
212
-224.
26
Bettelli, E., M. Pagany, H. L. Weiner, C. Linington, R. A. Sobel, V. K. Kuchroo.
2003
. Myelin oligodendrocyte glycoprotein-specific T cell receptor transgenic mice develop spontaneous autoimmune optic neuritis.
J. Exp. Med.
197
:
1073
-1081.
27
Petersen, T. R., S. Gulland, E. Bettelli, V. Kuchroo, E. Palmer, B. T. Backstrom.
2004
. A chimeric T cell receptor with super-signaling properties.
Int. Immunol.
16
:
889
-894.
28
Biondo, M., J. Field, B. H. Toh, F. Alderuccio.
2006
. Prednisolone promotes remission and gastric mucosal regeneration in experimental autoimmune gastritis.
J. Pathol.
209
:
384
-391.
29
Karnezis, T., W. Mandemakers, J. L. McQualter, B. Zheng, P. P. Ho, K. A. Jordan, B. M. Murray, B. Barres, M. Tessier-Lavigne, C. C. Bernard.
2004
. The neurite outgrowth inhibitor Nogo A is involved in autoimmune-mediated demyelination.
Nat. Neurosci.
7
:
736
-744.
30
Slavin, A., C. Ewing, J. Liu, M. Ichikawa, J. Slavin, C. C. Bernard.
1998
. Induction of a multiple sclerosis-like disease in mice with an immunodominant epitope of myelin oligodendrocyte glycoprotein.
Autoimmunity
28
:
109
-120.
31
McQualter, J. L., R. Darwiche, C. Ewing, M. Onuki, T. W. Kay, J. A. Hamilton, H. H. Reid, C. C. Bernard.
2001
. Granulocyte macrophage colony-stimulating factor: a new putative therapeutic target in multiple sclerosis.
J. Exp. Med.
194
:
873
-882.
32
Fassas, A., J. R. Passweg, A. Anagnostopoulos, A. Kazis, T. Kozak, E. Havrdova, E. Carreras, F. Graus, A. Kashyap, H. Openshaw, et al
2002
. Hematopoietic stem cell transplantation for multiple sclerosis: a retrospective multicenter study.
J. Neurol.
249
:
1088
-1097.
33
Fassas, A., A. Anagnostopoulos, A. Kazis, K. Kapinas, I. Sakellari, V. Kimiskidis, C. Smias, N. Eleftheriadis, V. Tsimourtou.
2000
. Autologous stem cell transplantation in progressive multiple sclerosis: an interim analysis of efficacy.
J. Clin. Immunol.
20
:
24
-30.
34
Aschenbrenner, K., L. M. D'Cruz, E. H. Vollmann, M. Hinterberger, J. Emmerich, L. K. Swee, A. Rolink, L. Klein.
2007
. Selection of Foxp3+ regulatory T cells specific for self antigen expressed and presented by Aire+ medullary thymic epithelial cells.
Nat. Immunol.
8
:
351
-358.
35
Derbinski, J., J. Gabler, B. Brors, S. Tierling, S. Jonnakuty, M. Hergenhahn, L. Peltonen, J. Walter, B. Kyewski.
2005
. Promiscuous gene expression in thymic epithelial cells is regulated at multiple levels.
J. Exp. Med.
202
:
33
-45.
36
Fazilleau, N., C. Delarasse, C. H. Sweenie, S. M. Anderton, S. Fillatreau, F. A. Lemonnier, D. Pham-Dinh, J. M. Kanellopoulos.
2006
. Persistence of autoreactive myelin oligodendrocyte glycoprotein (MOG)-specific T cell repertoires in MOG-expressing mice.
Eur. J. Immunol.
36
:
533
-543.
37
Wekerle, T., M. Sykes.
2001
. Mixed chimerism and transplantation tolerance.
Annu. Rev. Med.
52
:
353
-370.
38
Tian, C., J. Bagley, N. Cretin, N. Seth, K. W. Wucherpfennig, J. Iacomini.
2004
. Prevention of type 1 diabetes by gene therapy.
J. Clin. Invest.
114
:
969
-978.
39
Bonilla, W. V., M. B. Geuking, P. Aichele, B. Ludewig, H. Hengartner, R. M. Zinkernagel.
2006
. Microchimerism maintains deletion of the donor cell-specific CD8+ T cell repertoire.
J. Clin. Invest.
116
:
156
-162.
40
Tian, C., J. Bagley, J. Iacomini.
2006
. Persistence of antigen is required to maintain transplantation tolerance induced by genetic modification of bone marrow stem cells.
Am. J. Transplant.
6
:
2202
-2207.
41
Weaver, C. T., L. E. Harrington, P. R. Mangan, M. Gavrieli, K. M. Murphy.
2006
. Th17: an effector CD4 T cell lineage with regulatory T cell ties.
Immunity
24
:
677
-688.
42
Langrish, C. L., Y. Chen, W. M. Blumenschein, J. Mattson, B. Basham, J. D. Sedgwick, T. McClanahan, R. A. Kastelein, D. J. Cua.
2005
. IL-23 drives a pathogenic T cell population that induces autoimmune inflammation.
J. Exp. Med.
201
:
233
-240.
43
Stromnes, I. M., L. M. Cerretti, D. Liggitt, R. A. Harris, J. M. Goverman.
2008
. Differential regulation of central nervous system autoimmunity by TH1 and TH17 cells.
Nat. Med.
14
:
337
-342.
44
Keech, C. L., A. D. Farris, D. Beroukas, T. P. Gordon, J. McCluskey.
2001
. Cognate T cell help is sufficient to trigger anti-nuclear autoantibodies in naive mice.
J. Immunol.
166
:
5826
-5834.
45
Chan, J., E. J. Ban, K. H. Chun, S. Wang, J. McQualter, C. Bernard, B-H. Toh, F. Alderuccio.
2008
. Methylprednisolone induces reversible clinical and pathological remission and loss of lymphocyte reactivity to myelin oligodendrocyte glycoprotein in experimental autoimmune encephalomyelitis.
Autoimmunity
41
:
405
-413.