Claude de Préval

Antiinflammatory and immunoregulatory action of methotrexate in the treatment of rheumatoid arthritis: Evidence of increased interleukin‐4 and interleukin‐10 gene expression demonstrated in vitro by competitive reverse transcriptase‐polymerase chain reaction

OBJECTIVE To look for in vitro modulation of the main immunoregulatory and antiinflammatory cytokines by methotrexate (MTX) during the course of rheumatoid arthritis (RA). METHODS We quantified interleukin-2 (IL-2), IL-4, IL-10, and interferon-gamma (IFNgamma) gene expression by peripheral blood mononuclear cells ex vivo under basal conditions and in vitro after stimulation with phytohemagglutinin (PHA) or PHA plus MTX, by competitive reverse transcriptase-polymerase chain reaction (RT-PCR), in 12 patients with untreated active RA (group 1), 10 patients with MTX-treated disease in partial remission (group 2), and 11 healthy control subjects. Simultaneously, under the same experimental conditions, we quantified cytokine production by specific enzyme-linked immunosorbent assays (ELISAs). RESULTS Under basal conditions, we found no differences in IL-2, IL-10, and IFNgamma gene expression in the 3 groups, while IL-4 gene expression was significantly decreased in RA patient group 1 compared with the control group. In vitro, under the action of MTX, IL-10 gene expression was significantly increased in the 3 groups, IL-4 gene expression was significantly increased in RA group 1 and in the control group, and IL-2 and IFNgamma gene expression was significantly decreased in RA group 1. Cytokine gene expression assessed by RT-PCR and cytokine production assessed by specific ELISAs were highly correlated. CONCLUSION In vitro modulation of the cytokine network by MTX, increasing Th2 cytokines and decreasing Th1 cytokines, could explain its antiinflammatory and immunoregulatory actions in vivo during the treatment of RA.

The susceptibility to insulin-dependent diabetes mellitus is associated with C4 allotypes independently of the association with HLA-DQ alleles in HLA-DR3,4 heterozygotes.

In the genetically homogeneous Danish population, 27 HLA-DR3,4 heterozygous patients with insulin-dependent diabetes mellitus (IDDM) and 19 DR3,4 heterozygous controls without family history of IDDM were investigated for HLA-region markers and Gm and Km immunoglobulin allotypes. The aim was to define susceptibility factors for IDDM development other than HLA-DR using a number of techniques: lymphocytotoxicity (HLA-DR and DQ antigens), cellular methods (Dw and DP typing), restriction fragment length polymorphism (DQ alleles), electrophoresis and immunofixation (BF and C4 allotypes), and passive hemagglutination inhibition (Gm and Km immunoglobulin allotypes). The complement allotype C4A3 and the HLA-DQw8 (DQw3.2) antigen were found in all of the patients, whereas this was the case for only 8 of the 19 controls (P=6 x 10−6): five lacked C4A3, five others lacked DQw8, and one of the controls lacked both of these factors. Fourteen of the patients had the complement allotype C4B3 versus three of the controls (P=0.01). Previously reported family studies suggest that these alleles are part of the following haplotype: B15, BFS, C4A3, C4B3, DR4, Dw4, DQw8, and these factors were found together in ten of the patients versus one of the controls (P=0.01). The markers usually associated with DR3 did not show significant differences between IDDM patients and controls, and the non-HLA markers studied showed no significant deviation from what was expected. In addition to the susceptibility factor DQw8, the study suggests the existence of susceptibility genes for IDDM near the complement C4 genes on DR4-carrying haplotypes. Since recent works have shown that the structural gene for the monokine tumor necrosis factor alpha (TNF-α) is located between the HLA-B and C4 loci and that TNF-α might be of importance in IDDM pathogenesis, the hypothesis is put forward that the C4-associated IDDM susceptibility reflects linkage dis-equilibrium between the C4 gene and a gene controlling TNF-α production. The high relative risk for IDDM in HLA-DR3,4 heterozygotes might be explained by the combined action of IDDM-specific susceptibility genes on DR4 haplotypes and DR3-linked susceptibility genes associated with predisposition to autoimmunity.

Human collagen II peptide 256-271 preferentially binds to HLA-DR molecules associated with susceptibility to rheumatoid arthritis.

Abstract The binding ability of 23 overlapping peptides, all derived from the CB11 fragment of CII, was tested on several HLA-DR molecules associated or not with disease susceptibility. These experiments were performed on a variety of cells expressing different HLA-DR molecules, using both indirect and direct binding assays. The CII (256–271) fragment was shown to bind to a restricted population among which the HLA-DR molecules associated with susceptibility to rheumatoid arthritis. The results also clearly indicate that the binding specificity of CII (256–271), among the DR4 molecules, is controlled by the nature of the HLA-DR molecule β-chain residues 71 and 74, residues previously shown by X-ray crystallography to be involved in the HLA-DR/peptide interaction. The human CII (256–271) peptide is thus likely to play a role in the disease process.

T-cell receptor variable region of the β-chain gene use in peripheral blood and multiple synovial membranes during rheumatoid arthritis

In order to look for a site-specific T-cell response in RA SM, PCR analyses using oligonucleotide primers specific for 24 TCRBV (V beta) families were performed to compare the respective usage of each TCRBV gene by T cells present in PB and SM of 13 patients with RA. In four patients, SM cells from two or three sites of inflammation were subjected to analysis. In one patient, synovial tissue was studied at two different phases of the disease, resulting in a total number of 19 samples of SM cells, which were compared with paired samples of PB cells. The results showed that whereas all 24 TCRBV gene families could be detected in both PB and SM cells, there was some skewing of increased or decreased usage frequencies of particular TCR V beta genes among SM cells. Three TCRBV families were often overexpressed in SM: V beta 3, V beta 17, and V beta 22. Moreover, V beta 4 was often decreased in SM (7 out of 13). This decrease was statistically significant in the RA population studied. SM from different joints of a given patient showed similar variations of T-cell repertoire compared to PB, even 6 months later in the course of the disease. These results demonstrate a biased TCRBV gene utilization in RA SM. This bias appears to be similar in different joints and at different times in the course of the disease. No correlation was found between the bias of TCR repertoire in SM and the HLA typing of these patients.

A very conservative region of ApoB-100 in the putative binding region to the LDL receptor in the Toulouse population

The nucleotide sequence of the putative binding site of ApoB-100 was studied in Hypercholesterolemic IIa patients and controls from the Toulouse area. Only one patient possesses the 3,500 mutation, which is responsible for defective familial hypercholesterolemia. The other individuals (including 39 patients and 14 controls) display the same nucleotide sequence although four nucleotide substitutions have been described in this region. This homogeneity of the Toulouse population in the putative binding site of ApoB-100 is discussed.

Biased T-cell receptor usage is associated with allelic variation in the MHC class II peptide binding groove.

Abstract A comprehensive analysis was carried out of the tri-molecular complex of peptide, major histocompatibility class II molecule, and T-cell receptor (TcR) involved in the recognition of the promiscuous HA (306–318) peptide, restricted by one of two closely related HLA-DR alleles, HLA-DRB1*0101 and HLA-DRB1*0103. These two DR molecules differ by only three amino acids at positions 67, 70, and 71, in the third variable region of the DRB1 chain. None of the HA (306–318)-specific T-cell clones restricted by these two DR molecules tolerated amino acid substitution at the peptide-binding position 71, despite the fact that the substitution did not interfere with peptide binding. The majority of the DRB1*0103-restricted clones tolerated substitution of the amino acid at the TcR-contacting position 70, while the DRB1*0101-restricted T cells did not. Biased usage of TRVA and TRVB segments was observed for the DRB1*0103-restricted clones; in contrast, apparently random usage was seen in the DRB1*0101-restricted T cells. Finally, limiting dilution analysis revealed a lower frequency of T cells reactive with the HA peptide in a DRB1*0103 compared with a DRB1*0101 individual. Taken together these data suggest that biased TcR gene usage may reflect a relatively low precursor frequency of T cells, and the need for clonal expansion of a limited set of high avidity T cells.

EXTENDED HLA‐DQw2 HAPLOTYPES: MOLECULAR ANALYSIS

The HLA‐DQw2 specificity, homogeneous in serology, is strongly associated to two HLA‐DR specificities: DR3 and DR7. These alleles are found mainly on DQw2 bearing extended haplotypes with strong linkage disequilibrium. We describe, with BamHI, HindIII and RsaI, two restriction fragments length polymorphisms (RFLP) for the A gene of DQw2. These two subtypes correlated with the DR3 and DR7 specificities. Interestingly, by non‐equilibrium pH gradient electrophoresis (NEPHGE), two DQα chains were also found, respectively correlated with the same DR specificities. In addition, Hindi polymorphism allowed us to distinguish several patterns of B genes for (DR7) DQw2 haplotypes but without any detectable association with another HLA marker. However, only one DQ(3 chain was found by NEPHGE in the (DR7) DQw2 haplotype. Furthermore, MncII discriminated the B genes of the two extended haplotypes: (B8, DR3) DQw2 and (B18, DR3) DQw2. The same result was found by NEPHGE: two DQβ chains were described, corresponding to the same extended haplotypes. The use of exon‐specific DQB probes showed that the genomic polymorphism in DOw2 haplotypes is located, at least, at the 3’end of the gene. These data add new characteristics to the different DQw2 extended haplotypes.

Influence of histidine β81 of HLA-DR101 on peptide binding and presentation to T-cell receptor

Abstract HA(306-318) is an immunodominant peptide of the hemagglutinin of influenza virus that binds to most human leukocyte antigen (HLA-DR) alleles, while p18(73-85) is a HIV peptide characterized as a DR101 binding peptide. Our results demonstrate that crystal relaxation leads to the loss of a hydrogen bond between the β81 histidine and the HA(306-318) peptide. This histidine is also involved in the binding of superantigens like SEA via a coordination of a zinc atom. To monitor the interaction of these peptides with this histidine of HLA-DR molecules, chemical modification, peptide binding on HLA-DR101 wild type and mutated molecules, and proliferation experiments were conducted, together with molecular simulation of HLA-DR/peptide molecular complexes. Our data suggest a different binding peptide pattern, depending of whether the peptide is HLA-DR101 allele specific or a shared one. Furthermore, tyrosine substitution at position β81 does not affect either peptide binding or HA(306-318) clone-specific T-cell proliferation. On the contrary, the alanine substitution at position HLA-DR101 β81 abrogated both peptide binding and T-cell proliferation. These results suggest that the histidine 81 on the DRβ chain plays an important role in the HLA-DR peptide binding, more likely by polar interactions of the amino acid side chain ring with the peptide.