Search

To examine the effects of HLA-DRB1 alleles and amino acid sequences that carry the shared epitope (SE) upon rheumatoid arthritis (RA) susceptibility and disease severity in patients from Eastern France.HLA-DRB1 alleles were determined in 120 patients and 104 healthy controls by polymerase chain reaction/sequence specific oligonucleotide probes. Subtyping of DRB1*01 and *04 were performed using sequence specific primers. Patients were retrospectively evaluated for disease duration, age at disease onset, presence of rheumatoid factors, subcutaneous nodules, vasculitis and other extraarticular diseases, for the need for arthroplasty and immunosuppressive/immunoregulatory agents, and for radiographic damage.The prevalence of HLA-DRB1*04 was significantly higher in patients (46.6%) than in controls (17.3%) (Pcorr = 0.000003). HLA-DRB1*0101 and *0401 were the most prominently associated subtypes in patients with RA (33.3%, Pcorr = 0.011, and 28.3%, Pcorr = 0.00008, respectively). A significant fraction of patients (72.5%) expressed one or 2 copies of the SE (p0.0000001; OR 4.77, CI 2.61-8.78). The presence of double SE was associated with a higher risk of developing RA (OR 4.83, CI 1.91-12.71; p = 0.0001). No significant differences in the clinical records among patients expressing no RA linked alleles, one and 2 copies of the SE, were observed. However, analyzing the specific effect of each amino acid sequence, we observed a significant association of the QKRAA motif with vasculitis (p = 0.03) and history of joint replacement surgery (p = 0.05), suggesting a role for lysine in position 71 of the shared sequence.These findings differ from those of previous HLA-DRB1 allele studies in patients with RA from other regions of France. Thus, the heterogeneity in both the expression of DRB1 alleles and the association of these alleles with disease severity could be relevant within a country such as France.

Clinical and experimental data suggest a role for the immune response in preventing leukemic relapses after allogeneic bone marrow transplantation (BMT): the graft-versus-leukemia (GVL) effect. In this report, we have evaluated the response of normal donor lymphocytes against allogeneic leukemic cells as an in vitro model of the GVL effect. We used a limiting dilution technique in order to determine the frequency of cytotoxic T-lymphocyte precursors (pre-CTL) against allogeneic leukemic blasts among normal donor lymphocytes. We demonstrate a considerable variability of CTL precursor frequency. This variability depended on leukemic populations since, for a given leukemia, the pre-CTL frequency was comparable among our tested normal allogeneic donors. Moreover, when HLA-DR negative leukemias were used as allostimulators, the pre-CTL frequencies were extremely low. In order to verify the impact of leukemic DR expression on the stimulatory capacity of leukemic cells, we selected and analyzed in mixed lymphocyte tumor cell culture (MLTC), a panel of myelogenous and lymphoblastic leukemias with variable levels of DR expression, each against different allogeneic responders. Our results demonstrated a close correlation (r = 0.953, p0.0001) between the proliferative response of alloactivated lymphocytes and the percentage of stimulatory leukemic cells expressing HLA-DR molecules. Anti-MHC class II monoclonal antibodies inhibited the lymphocyte proliferation in the MLTC, confirming the preponderant role of DR in the generation of this response. Overall, our results demonstrate the extreme variability of leukemic cells in their allostimulatory capacity and the central role of DR expression in determining leukemic allo-recognition. In the setting of a clinical protocol, our data suggest that the infusion of allogeneic T lymphocytes in a DR negative leukemia will not lead to an alloreactive T-cell anti-tumor effect.