HLA and Graves' disease: an association with HLA-DRw3.

HLA-A, -B, and -C antigens were tested by a standard lymphocyte microcytotoxicity technique in 86 Caucasians patients from western France with Graves' disease, and the data were compared with findings in 356 healthy controls. For HLA-DR antigen typing performed by lymphocyte microcytotoxicity testing using a long incubation time, the data were compared to findings in 100 healthy controls. An increase was found in the frequency of HLA-DRw3 [51.16% of patients vs. 20% of controls, corrected P (Pc) < 0.0003; relative risk (rr), 4.19) associated with an increased frequency of HLA-B8 (44.19% of patients vs. 22.47% of controls; Pc < 0.001; rr, 2.73) and HLA-A1 (40.7% of patients vs. 28.93% of controls; Pc < 0.03; rr, 1.71). In contrast, a diminished frequency was found for HLA-B12 (12.79% vs. 31.74%; Pc < 0.01). The antigen combination B8-DRw3 was noted in 37 of the 86 Graves' disease patients compared with 13 of 100 controls (Pc < 0.00003). No association was observed between HLA antigens and the different manifestations of the disease, such as the presence of goiter and/or exophthalmos, or the severity of clinical or biochemical signs. The present findings confirm the reported increase in the frequency of HLA-B8 in patients with Graves' disease. The most striking finding was the prevalence of HLA-DRw3, which, together with recent reports on lymphocyte-defined D locus determinants pointing to an increase frequency of HLA-Dw3, suggests that the gene or genes conferring susceptibility to Graves' disease may be located close to the HLA-D (DR) region of the sixth chromosome.

No major monoclonal lymphocyte population in the thyroid of patients with Graves' disease: study of gene rearrangement by restriction fragment length polymorphism and polymerase chain reaction.

The present study aimed at determining the mono-, oligo-, or polyclonal nature of intrathyroid lymphocytes at the DNA level in patients with Graves' disease. Two techniques were used to seek monoclonal rearrangement in DNA derived from intrathyroidal lymphocytes obtained from six patients. The first was restriction fragment length polymorphism using two specific probes from the B-chain of T-cell receptor and the other from the heavy chain immunoglobulin gene; the second was polymerase chain reaction using a couple of specific primers from the variable and joining regions of heavy chain immunoglobulins. The results for the patients with Graves' disease were compared with those obtained for circulating T-and B-lymphocytes, granulocytes (negative controls), and T- and B-leukemic cells (positive controls). The results with restriction fragment length polymorphism favored a polyclonal origin for the lymphocytes in all cases, since no rearrangement was visualized. The results with polymerase chain reaction were analogous, and the technique was 10 times more sensitive in the detection of rearrangement.

A prospective study of the relationship between relapse of hyperthyroid Graves' disease after antithyroid drugs and HLA haplotype.

One hundred and eleven unselected patients with hyperthyroidism due to Graves' disease received decreasing doses of carbimazole for 18 months. Clinical examination and hormonal assays (serum T3, T4, free T4 index) were done at 4, 9, and 18 months of treatment. Patients were typed for 35 HLA antigens and were followed for 2 yr after withdrawal of treatment; 39 patients were excluded for various reasons and 72 were retained for study. Of the 72 patients, 37 relapsed and 35 remained in remission; 40 patients were DR3+ (20 relapsed) and 32 were DR3- (17 relapsed). HLA frequency was not significantly different in patients who relapsed and those who remained in remission. Thus, under the conditions of this study, HLA frequency could not be used to predict relapse of hyperthyroidism due to Graves' disease.

Antithyroid drugs and Graves' disease: a prospective randomized evaluation of the efficacy of treatment duration.

A prospective randomized study was performed in patients with hyperthyroid Graves' disease (GD) in order to compare long (18 months) and short term (6 months) antithyroid drug treatment on the remission rate. A therapeutic protocol was offered to all GD patients who had not been treated for this disease previously. All patients studied who followed the protocol were rechecked 2 yr after treatment was withdrawn, or earlier in the case of relapse. Of the patients having undergone long term treatment, 61.8% still were in remission 2 yr after treatment withdrawal, whereas only 41.7% of the patients treated for 6 months were in remission (P less than 0.05). Such findings clearly establish that treatment duration has a direct beneficial incidence on the remission rate. These results were confirmed by the fact that treatment for 18 months resulted in remission in 7 of 15 patients who had previously relapsed after a 6-month course of therapy. This improvement in relation to treatment duration might be due to the immunosuppressive action of carbimazole. No significant difference was observed between relapse and remission groups, regardless of treatment duration, for HLA ABDr, serum T3 and T4, and T3/T4 ratio determined before treatment. Only the thyroid-stimulating antibody levels determined at the time of diagnosis and at the end of treatment were higher in the relapse group, a difference that was relevant only globally, due to value scattering. Furthermore, thyroid-stimulating antibody levels at the end of treatment may indicate remission or, conversely, continuance of the pathological process.(ABSTRACT TRUNCATED AT 250 WORDS)

A new immunoreagent for cell labeling. CD3 monoclonal antibody covalently coupled to fluorescent polymethacrylic nanoparticles.

The preparation and application of immunonanospheres are described. CD3 monoclonal antibodies were covalently coupled to fluorescent polymethacrylic nanoparticles by the glutaraldehyde reaction and the resultant conjugate purified by gel filtration on a Sepharose 4B column. Peripheral blood mononuclear cells labeled with this immunoreagent were observed by both fluorescence microscopy and scanning electron microscopy in order to evaluate the technique.

Flow cytometric quantitative evaluation of phagocytosis by human mononuclear and polymorphonuclear cells using fluorescent nanoparticles.

The use of fluorescent polymethacrylic nanoparticles (0.3 micron) as a flow cytometric reagent in the quantitative evaluation of phagocytosis by human mononuclear and polymorphonuclear cells is described. The preparation of the nanoparticles, by emulsion copolymerization of methacrylic monomers, and their physicochemical properties are briefly summarized. Nanoparticles coupled with a fluorescent agent (ethidium bromide) were used in a flow cytometric assay to study opsonin-independent phagocytosis by human polymorphonuclear cells and by human monocytes. The phagocytosis of nanospheres by monocytes was determined by flow cytometry from the fluorescence distribution and ingestion was visualized by scanning and transmission electron microscopy. One possible application of the fluorescent nanoparticles is the simultaneous analysis of cell surface antigens and cell phagocytic activity.

Occurrence and specificity of anti-B lymphocyte antibodies in renal allograft recipients.

Anti-HLA-A,B and anti-B lymphocyte antibodies were screened as part of a prospective alloimmunity monitoring study in 29 renal allograft recipients using a standard microlymphocytotoxicity test. Warm-reactive and/or cold-reactive lymphocytotoxins were directed against a panel of B lymphocytes, the donor's B lymphocytes, and the recipient's own B lymphocytes. A small proportion of patients had pretransplant antibodies, whereas about one-half of the patients had post-transplant antibodies. One-year allograft survival rates were lower among the patients with warm- and cold-reactive sera than among those with nonreactive sera or pure B cold-reactive sera. The sera of 20 patients were tested against donor B lymphocytes. The presence of donor-specific antibodies correlated closely enough with graft loss to be of predictive value. Autoantibodies appeared to have an enhancing effect in this study.

The role of human spleen interferon on natural killer activity of peripheral blood lymphocytes enriched in large granular lymphocytes.

The elimination of monocytes as well as B- and T-lymphocytes by forming rosettes with high affinity for sheep red blood cells yielded an enriched population of both natural killer (NK) activity (cytotoxicity: 65.4 +/- 9.9% with an E/T ratio of 12:1, P less than 0.005) and large granular lymphocytes (LGL: 76 +/- 13%) compared to the untreated lymphocyte population where NK activity is 35.7 +/- 17.3% (E/T 12:1) and the percentage of LGL of 26 +/- 6%. We studied the action of type I interferon (IFN) obtained from human spleens, on NK activity of 9 peripheral blood lymphocyte populations and 9 enriched in LGL. NK activity of the total lymphocyte population is significantly increased (P less than or equal to 0.05) in 6 out of 9 cases after treatment by interferon. Cell populations enriched in LGL showed increased NK activity in only one case after treatment by interferon, but no increased activity was found in the other cases. These results are compatible with the notion of cellular cooperation in increased NK activity by interferon.

Spleen cell populations in normal and tumor-bearing hamsters.

Spleen cell subpopulations from normal and tumor-bearing hamsters (TBH) were quantified using Petri dishes coated with specific antibodies, and by flow cytometric immunofluorescence analysis. The relative numbers of T cells (Thy + cells) decreased, by both methods, as a function of tumor growth, while the number of B cells (bearing surface Ig) increased. Cells without T or B markers (null cells) were more numerous in the spleen of TBH and a large number of them expressed the receptor for the Fc fragment of IgG. Splenic cells were also sorted according to their light-scattering properties, and electron microscopic analysis was performed on the sorted fractions. It showed the presence of secreting plasmocytes and activated macrophages in the spleen of TBH.

A monoclonal antibody to the HLA-A3 alloantigen.

Monoclonal antibody production recognizing the HLA-A3 antigen is described. The XI-23 antibody reacted with all of the 89 cell suspensions carrying the HLA-A3 antigen (100% cytotoxicity) among a total of 191 suspensions tested. No extra-reactivity or cross-reactivity was observed, particularly with that of HLA-A11. This antibody can thus be considered as a good HLA-typing reagent.

Diminution of circulating polymorphonuclear leucocytes by administration of a monoclonal antibody reacting with granulocytes and monocytes.

The monoclonal antibody 3A35 which binds mouse polymorphonuclear leucocytes (PMN) and monocytes was injected intravenously (i.v.) into normal mice. A great diminution of circulating PMN was observed. The percentage of PMN passed through a minimum (2.5%) 20 min after antibody injection and returned to normal value (18.3%) within 24 h. After repeated daily injections, the ability of the antibody to induce granulopenia attenuated. Moreover, mice bearing the 3A35-producing hybridoma as an ascitic tumor had a normal percentage of blood PMN and a normal granulopoiesis as judged from bone marrow cytological examination. Thus, the monoclonal antibody produced a transitory diminution of PMN but could not induce a lasting granulopenia.

New class I in man: serological and molecular characterization.

New class I antigens in linkage disequilibrium with HLA-A antigen are demonstrated in PHA T and EBV preferential target cells using human alloantisera. These new antigens are defined as class I antigens by immunoprecipitation of a 41-12 k dimer. The molecule is shown to be distinct from the HLA-A, -B, -C molecule and in particular from the A3 molecule as in sequential immunoprecipitations, the depletion of the HLA-A, -B, -C molecule or A3 molecule (44-12 k) has no effect on the new molecule (41-12 k). Being present on the PHA T cells and lymphoblast lines, these antigens are considered as new epitopes involved in the the cell activation process.

Evidence for a polymorphism of HLA-G gene.

HLA-G gene polymorphism was analyzed by RFLP using seven restriction enzymes and an HLA-G locus-specific probe. Hybridization of 55 DNAs digested with three enzymes (Taq I, Pst I, and Bgl II) revealed two polymorphic bands in each case. RFLP patterns obtained with Taq I and Pst I corresponded to the same allelic polymorphism and differed from the Bgl II polymorphism. Combining both polymorphisms enabled determination of four alleles. Allelic frequencies were calculated: 40% of the subjects tested had allele 1, 36% had allele 2, 22% had allele 3, and 2% had allele 4. Analyzing the complete HLA class I phenotype revealed strong linkage disequilibrium with the HLA-A locus. The polymorphism described is located in the 3' flanking region of the gene. Moreover, extended HLA-A haplotypes were constructed by combining the HLA-G polymorphism with other class-I-sequence polymorphisms.

Implication of the HLA-DRB3 gene in Graves' disease: predominance of allele Dw24.

Using RFLP, the present study sets off to determine the MHC class II gene polymorphism in Graves' disease, in order to define the HLA-related genetic susceptibility. Considering the preferential link between Graves' disease and the HLA-DR3 antigen, 42 HLA-DR3 Graves' disease patients were studied and compared with 42 HLA-DR-matched controls. Hybridization with a DQ alpha probe of DNAs digested by Taq I revealed a polymorphism of the DR3 haplotype with an overrepresentation of a 2.1 kb(U) fragment in patients, but this was merely a sign of the linkage disequilibrium between U and B8DR3. Hybridization with the DR beta probe of DNAs digested by Taq I yielded more facts. It revealed the overrepresentation of the Dw24 specificity (Taq I:9.8 kb) in DR3 Graves' disease patients. This study thus enabled us to determine precisely the susceptibility linked to the DR3 haplotype, implicating the DRB3 gene and its Dw24 allele, which appear to be the most reliable markers of the disease, providing a higher relative risk than B8DR3.

Genetic susceptibility and anti-human platelet antigen 5b alloimmunization role of HLA class II and TAP genes.

Platelet alloimmunization may result in post-transfusion purpura, and during pregnancy may cause neonatal alloimmune thrombocytopenia (NAIT), with a frequency estimated at 1.3 per 1000 live births. The risk of morbidity is significant: 20% of affected infants have neurologic sequelae and the death rate is about 10%. A better understanding of the immune response to platelet alloantigens would allow for a better definition, and thus better management of pregnant women at high risk. Limited data are available on the immune response against HPA-5b, the second most frequent antigen, after HPA-1a, implicated in NAIT. We studied HLA class II and TAP gene polymorphism in 50 women immunized against HPA-5 system antigens. Our results suggest a strong association of alloimmunization with a cluster of HLA DR molecules sharing a particular polymorphic amino acid sequence at position 69-70 (Glu-Asp encoded by GAA-GAC nucleotide sequence) of the DR beta 1 chain (RR = 2.95, RR = 5.70 when patients were homozygous for this sequence), and a negative association with the DRB1*0301 allele (2.1% vs. 28%; RR = 0.08). Furthermore, increased frequency of a TAP2 dimorphism at position 379 was observed in immunized women against the HPA-5 antigens (RR = 4.7).

The role of HLA-DR-DR and HLA-DR-DP interactions in genetic susceptibility to rheumatoid arthritis.

In order to analyze the relationships between the DR and DP loci in the genetic susceptibility to RA, HLA-DRB1 and -DPB1 polymorphism was studied in 155 RA patients compared to 150 controls, using a reverse dot-blot analysis. Our data were consistent with the involvement of the amino acid in position 71 of the third hypervariable region of the DR beta 1 chain in susceptibility to the disease. The higher risk for RA was observed in patients who carried the association of a lysine (K), characterizing the DRB1* 0401 susceptibility allele, with an arginine (R), observed in all the other DRB1* susceptibility alleles (21.9% vs 0.6%, p(c) < 10(-6), OR = 42) In the absence of arginine, the presence of lysine was still associated with the disease (33% vs 19%, p(c) < 0.03, OR = 2). In contrast, in the absence of lysine, the frequency of arginine in position 71 was similar in patients and controls (30% vs 26%, p = NS). On another hand, the analysis of the HLA-DPB1 locus showed that the DPB1 *0401 allele frequency was significantly increased in the RA patient group (n = 47) who expressed only arginine at the position 71 of the beta 1 chain (82% vs 56% in controls, p < 0.008), with role of HLA-DR--DR and -DR-DP interactions in the genetic susceptibility to RA.

Interaction between certain major histocompatibility complex class II and T-cell receptor V beta alleles promotes the antibody production to extractable nuclear antigen-related peptides.

Our objective was to study the interaction between major histocompatibility complex (MHC) class II and T-cell receptor (TCR) alleles in the recognition of extractable nuclear antigen-derived peptides in 32 patients with systemic lupus erythematosus and 173 of their family members. MHC genes were analyzed using sequence specific oligonucleotides, and TCR beta-chain gene polymorphism using restriction fragment-length polymorphism. One dominant peptide (as defined by enzyme-linked immunosorbent assay autoantibody reactivity) was identified in each antigen studied: peptide 1-20 in Sm-D, peptide 35-58 in U1-RNP-A, and peptide 304-324 in the Ro/SSA 60 Kd protein. None of the MHC class II and TCR beta haplotypes was directly associated with any of the autoantibodies. Twenty-six subjects had antibodies to the peptide Sm-D1-20; nine of them were DRB1*0101/DQB1*0501. Among subjects with this haplotype, the number of responders was higher (p < 0.028, p corrected, pc = 0.336) in those with the 2-25-9 TCR beta haplotype than in the remainder. Conversely, the number of DRB1*04/DQB1*0302 responders was lower (p < 0.030, pc = 0.360) among subjects with the 23-20-9 TCR beta haplotype than in those without. The odds ratios (OR) were 4.23 and 0.21, respectively. Of the 54 subjects positive for anti-U1-RNP-A 35-38, 13 were DRB1*0101/DQB1*0501 and eight DRB1*04/DQB1*0302. The percentage of responders was higher (p < 0.041, pc = 0.492, OR = 3.48) in the former group of subjects with the 2-25-9 TCR beta haplotype, and lower (p < 0.02, pc = 0.024, OR = 0.09) in the latter with the 23-20-9 TCR beta haplotype. Three of the 12 anti Ro/SSA 60Kd 304-324-positive subjects were DRB1*0101/DQB1*0501. All had the 2-25-9 TCR beta haplotype (p < 0.046, pc = 0.552, OR = 6.29) and none the 23-20-9 (p < 0.031, pc = 0.372, OR = 0.10). The same combinations of genes were associated with high/low response toward the three peptides. These data provide evidence for an interplay of the MHC class II and TCR beta alleles in the control of specific autoantibody response to well-defined nuclear Ag peptides.

TAP2 gene polymorphism contributes to genetic susceptibility to multiple sclerosis.

MS is an autoimmune demyelinating disease that has been known to be associated with the HLA-DRB1*1501-DQA1*0102-DQB1*0602 haplotype. TAP1 and TAP2, two genes encoded within the MHC class II region between HLA-DP and -DQ loci, display genetic variability and are involved in the transport of antigenic peptides from the cytoplasm to the endoplasmic reticulum. Comparison of 116 MS patients with Caucasoid controls did not reveal any significant correlation between the previously described alleles of the TAP1 and TAP2 genes and MS. We report here an additional TAP2 dimorphism at codon 386, called I and J, corresponding to a silent mutation. An increased frequency of the J variant was observed in the patient population. The J mutation was not found in linkage disequilibrium with the HLA-DRB1*1501 allele and can be considered an additional genetic susceptibility marker of the disease.

Longitudinal study of HIV-specific cytotoxic lymphocytes in HIV type 1-infected patients: relative balance between host immune response and the spread of HIV type 1 infection.

To evaluate the contribution of a specific cytotoxic response in the control of HIV infection in relation to clinical status, we performed serial analysis of anti-Env and anti-Gag cytotoxic activity in 13 infected individuals over a 6- to 10-year period, using cryopreserved peripheral blood mononuclear cells (PBMCs). Autologous EBV-transformed B cell lines infected in vitro with recombinant vaccinia viruses expressing HIV-1 env and gag genes were used as targets. Without any stimulation of the effector cells, we were able to show an anti-HIV cytotoxic activity in the PBMCs of 12 of 13 HIV-1-infected patients, consistent with chronic immune activation in HIV infection. Different patterns of HIV-specific cytotoxic activity were observed, and the extent of this cytotoxic response varied between the clinically defined groups of individuals. No direct relationship was observed with the number of CD4 and CD8 lymphocytes during the observation period. However, patients who remained asymptomatic had a more vigorous cytotoxic response than patients with clinical deterioration during the observation period, and a significant difference was observed for HIV Gag-specific CTL activity. From these data, we suggest that the HIV-specific cytotoxic response has a protective role in the course of HIV infection.

Anti-pancreatic autoimmunity and Graves' disease: study of a cohort of 600 Caucasian patients.

The aim of this study was to investigate the frequencies of clinical diabetes and humoral markers of anti-pancreatic autoimmunity in a homogeneous population of 600 Caucasian patients with recently diagnosed Graves' disease (GD), in order to characterize the specific features of this group of endocrine patients among subjects at risk of diabetes. Ten were already diabetic at GD diagnosis. Among the 590 non-diabetic patients, 29 had islet cell antibodies (ICA), including 15 with low titre ICA and only 1 ICA-positive subject with a familial history of diabetes. Twenty-four patients had insulin autoantibodies, including three in association with ICA. Glutamic acid decarboxylase (GAD)/64 kDa antibodies were found in 16 of the 150 tested sera, including 13 of the 29 ICA-positive sera. Four ICA-positive patients displayed 37/40 kDa antibodies, including three in association with GAD/64 kDa antibodies. During follow-up, one of the ICA-positive patients developed insulin-dependent diabetes, 14 years after the GD diagnosis. To summarize, this anti-pancreatic autoimmunity study was focused on a large but specific and homogeneous group of subjects at risk for diabetes: recently diagnosed GD patients. This population was characterized by a high prevalence of GAD/64 kDa antibodies but also by a low frequency of evolution towards diabetes and the slowness of the process which could be due to the fact that only a minority of subjects possessed a sufficient combination of anti-pancreatic markers at the same time.