Search

A sample of 100 individuals from 50 French families of known pedigrees were typed for 14 loci of the HLA region (DPB1, DQB1, DQA1, DRB1, DRB3, 4, 5, C4B, C4A, Bf, C2, TNFa, TNFb, B, Cw, A). Linkage disequilibrium in each pair of loci was investigated by an exact test using a Markov chain algorithm. The results indicate no disequilibrium between DPB1 and the other loci, whereas the other class II genes are all significantly linked to each other. Linkage disequilibrium is also detected between some pairs of class I and class II-class I loci despite the long physical distance separating the loci (e.g. A-B, Cw-DRB1). On the other hand, some contiguous loci of the class III region are found to be in equilibrium with each other. Several hypotheses including selection, but also unequal allelic diversity at different MHC loci are discussed to explain this complex pattern of linkage disequilibrium.

HLA-G, a nonclassical HLA class I antigen, presents tissue-restricted expression on human trophoblasts and may play an important role in immune tolerance of mother-versus-fetus. In this work we have demonstrated extensive HLA-G genomic polymorphism within three CEPH reference families, by PCR-SSCP analysis and direct sequencing. Among six unrelated parents we assigned eight HLA-G alleles, seven of which are new. We observed the segregation of HLA-G alleles of heterozygous parents among their offspring that matched the segregation of the HLA class I haplotypes. Only one of the mutations observed was found to be nonsynonymous indicating low polymorphism of the HLA-G molecule.

We propose to use the extreme diversity of HLA in order to detect, by anonymous check, double registration on computer files, of patients waiting for organ transplantation. A simulation on a panel of 69461 unrelated individuals, caucasoid, confirms the relevance of the method.

A likelihood approach to HLA serology has been developed in which the aim is not to define a recognition set for a serum but to describe the serum's ability to react with each and every antigen in the test cells, this ability being quantified in terms of the probability of a positive reaction. For a given set of probabilities, one for each antigen, it is possible to derive the probability of the observed set of reactions (the likelihood of the set of probabilities). The maximum possible value of the likelihood for any possible combination of the probability set can then be sought, but this requires a maximization of likelihood with respect to 60-100 independent parameters. Theoretical considerations of the shape of the likelihood surface prove that, in this particular case, this is a feasible proposition. This approach allows the recognition of three groups of antigens: those for which there is considerable evidence of a specificity, those for which there is either no specificity or a very weak specificity, and those for which there is insufficient evidence on which to base a conclusion. The existence of a specificity can be tested using a log likelihood ratio as a statistic, but the usual assumption of a chi 2 distribution of this statistic cannot automatically be made in this situation. Therefore, the distribution is estimated by simulation. A serologist using this approach would receive considerably more information as to the serum's reaction patterns and valid statistics for the existence, or not, of a specificity.

To determine whether genetic markers can improve the predictive value of islet cell antibodies for development of Type 1 (insulin-dependent) diabetes mellitus, 536 siblings aged 2-29 years were consecutively enrolled in a 8-year prospective survey. The risk of developing diabetes was estimated, using life-table methods, by years of follow-up and age, according to genetic factors (shared HLA-haplotypes, DR antigens, C4 allotypes) and islet cell antibody status. Fifteen siblings (2.8%) developed Type 1 diabetes during the study period (risk 4.4% after 8 years, 4% by age 22 years). DR3,4 heterozygosity identified higher risk (16% after 8 years, 12% by age 22 years, p less than 10(-5)) than HLA-identity (10% and 7%, respectively, p less than 0.01); risks for DR3 or DR4 positive and for haplo-identical siblings were low (4%, 3% and 4.4%, respectively, NS). C4BQO also conferred significant risk (11% vs 3% in non-C4BQO siblings, p less than 0.01). The predictive value of genetic markers alone was poor (12% for DR3,4, 7% for HLA-identity, 9% for C4BQO) compared with that of islet cell antibody levels greater than 4 Juvenile Diabetes Foundation units (41%, risk 56% after 8 years, p less than 10(-7)). HLA markers significantly contributed to risk prediction in combination with islet cell antibodies: islet cell antibody-positive DR3,4+ subjects had the highest risk (70% after 8 years, predictive value 58%, p less than 10(-7)) compared with islet cell antibody-positive DR3,4- (37% and 20%, respectively) and islet cell antibody-negative DR3,4+ (5% and 3.6%, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)

The marker association segregation chi-square (MASC) method was applied to a sample of 416 Caucasians affected with insulin-dependent diabetes mellitus (IDDM), for which information on the parental and sibship status was available, as well as HLA typing. We show that the model which best explains all the observations assumes a cis or trans complementation of two tightly linked genes within the HLA region, an additional maternal effect, and other familial factors. The HLA molecule corresponding to the complementation of Arg52(+) and Asp57(-) has been recently proposed as explaining susceptibility to IDDM. However, this hypothesis does not account for the overall observations made on the HLA marker in IDDM patients and their relatives. The MASC method may also be applied to evaluate the risk for relatives of an affected individual (the "index"). For example, the risk for a sib depends not only on the parental status and on the number of HLA haplotypes he shares with the index, but also on which haplotype the index himself inherited from his mother and father.

HLA genotypes were ascertained in 53 French Caucasian families, comprising 68 juvenile onset insulin-dependent diabetic siblings. Among the 49 alleles detected at different loci in the HLA complex (A, C, B, Bf, DR) 4 appeared to occur at a significantly higher frequency among the 53 index cases than in a control series of 116 healthy individuals: HLA-B18 (p < 10(-3)), DRw3, DRw4 and BfF1 (p < 10(-6)). The excess of HLA identical affected siblings confirms genotype disequilibrium and supports the hypothesis of an HLA-linked gene(s) conferring susceptibility. There was no increase of homozygosity for HLA DRw3 and DRw4 whereas there was a marked excess heterozygosity for HLA DRw3/DRw4 in diabetic patients (32% versus 0% in the control series, p < 0.001). These data provide evidence for the existence of two cooperating genes, linked to each of the HLA DR alleles.

R4 heterozygotes was observed. By contrast, the observed frequency of patients homozygous for DR3 or DR4 was not increased, but even slightly decreased. The data support a model of inheritance comprising at least two closely linked specifically "diabetic" loci (most of the time marked by B18, BfF1, DR3 and B15, BfS, DR4) and a non-specifically "diabetic" haplotype favouring auto-immunisation (most of the time marked by B8, BfS, DR3). This model is discussed in the light of the presented data and of those of the literature.

135 patients, suffering from aplastic anaemia (AA) and their families were genotyped for HLA. The antigen and haplotype frequencies were compared to an HLA genotyped control panel composed of 209 normal couples and their healthy offsprings, and to another series of 2286 normal individuals. An excess of HLA-A2 was observed in the patients: 61% versus 42% (pc less than 0.001) (relative risk: 2) and versus 48.5% (p less than 0.01) in two control series, respectively. When considering the HLA-A, B antigens shared in common by the parents of the AA patients, an excess of HLA-A2 was observed: 32% as compared to 17% shared by normal couples (p less than 0.001). An excess of homozygous HLA-A2 was noted in the AA patients (14%) in comparison to the normal controls (4%) (p less than 0.001). The mechanism of this association is discussed as well as the hypothesis of a gene involved in haematopoiesis which might interact within the HLA-A region.

HLA associations with 21-OH deficiency were studied on respectively 109 and 60 congenital and late onset French index cases. Significant negative associations were found with antigens B8: congenital forms; B5, DR3: late onset. Significant positive associations were observed with A3, Bw47 (A3 Cw6 Bw47 DR7): congenital forms; B40: salt-wasting form; B5: simple virilizing form; Aw33, B14, DR1, DR2, DRw6 (Aw33 B14 DR1): late onset form. Among late onset patients not bearing B14 antigens significant positive associations were observed with B12 and B35.

A retrospective study of HLA immunization profile before a first graft of unrelated kidney in 212 immunized recipients (with an initial peak greater than or equal to 25%) shows that one can define at least 3 groups of responder with a different graft outcome. Recipients with either weak or strong variance of HLA antibody (VP) before graft (group I: VP less than 230 and group III: VP greater than 830) have a significantly lower graft survival: 54% (p = 0.006) and 45% (p = 0.002) respectively at 4 years, than recipients of group II (230 less than VP less than 830) for whom the prognosis is of 77% after the same time period. If similar results are observed on independent series, VP would be of a great interest for the definition of a new classification of responders, and on a practical level for the choice of renal transplant.

Eighty-two healthy individuals have been typed for HLA-A, B, C, antigens, and 49 of them also for HLA DR alleles. They were a sample representative of 11 of the 14 Malian ethnic groups living in the area of Bamako (Mali). Phenotypic frequencies have been compared to those of other Negroid and Caucasoid reference populations. As expected, in Negroids the increased frequency of HLA A23, A28, A30, and ATh was confirmed in the present series. Additionally, a significantly increased frequency of HLA B5 (B51 and Bw52) was noted--already observed in some but not all Negroid populations. Conversely, a decreased frequency--compared with that usually found in Negroid population--was observed for the alleles Bw42, Bw58, Cw6, and DRw6. Our results underline the originality of the Malian population among Western Africans.

Thirty-one unrelated patients, 15-52 years old, were typed by microlymphocytotoxicity for 27 alleles of the HLA system. In addition, 12 families including 1 or more patient were also analysed. This criteria for diagnosis were those of Sohar et all. (Am. Intern. Med., 1967, 43, 227-253). All patients were of Israelite-Sephardin origin except two (Armenian and French); they were from North-Africa (Tunisia, Morocco and Algeria) and Israël. The results were compared to the antigen frequencies of 3 reference normal populations. The frequencies of the studied alleles do not differ from those of controls, except for HL-A28 and B14 slightly increased when compared to the normal frequencies. The study of 7 families with at least two sibs suffering from FMF shows a random distribution of the genotypes : 2 HLA identical, 6 different and 10 haploidentical diseased sibs. This distribution differs significantly (p less than 0.01) from that expected in the case of a recessive inheritance. These data do not support the hypothesis of a linkage between genes controlling FMF and HLA genes.

The transmission of HLA genes was studied in an isolated population of French origin on the lesser Antilles islands in the West Indies. The study of 74 unrelated individuals, 44 of whom were genotyped, was carried out for the alleles of HLA loci: A, B, C and Bf (proactivator factor of properdin). As a result of the founder effect and the inbreeding process, the HLA haplotypes were noted to be less polymorphic than in a French continental population. Two haplotypes: A2, Cw5, B12, BfS and A3, C-, B14, BfF represent 24% of the observed haplotypes, and only 2% of the reference haplotypes in France. No significant excess or deficit of homozygotes was observed at the A and B loci.

Interaction between HLA and Gm for susceptibility to insulin-dependent diabetes (IDD) has been studied in 135 IDD families comparing HLA and Gm phenotype distributions in patients and sibling controls. The association analysis comparing the 135 index cases to 124 sibling controls did not show any significant interaction, but only a tendency for Gm 4,5 phenotypes to be more frequent in IDD patients, particularly in those who carried DR3/X. The sib pair analysis of 16 pairs of affected sibs did not show any distortion of segregation of Gm. On the other hand, among 98 sib pairs of one affected and one nonaffected sib, there was a significant distortion in favour of sib pairs with non-identical Gm phenotypes. However, no interaction between HLA and Gm was found. These results provide suggestive evidence that Gm or a locus very close to Gm could be involved in susceptibility to IDD.

One hundred fifty-five healthy nonagenarians, 45 men and 100 women, all French Caucasians, were phenotyped for alleles of the A, B, C, DR loci of the HLA complex. The observed HLA antigen frequencies were compared to those of a control series of 133 males and 179 females whose ages ranged from 10 to 50 years. When comparing the total young and elderly series, no significant differences were observed with respect to HLA antigen distribution or heterozygosity at any of the loci. When taking sex difference into account, however, an excess of the Cw1 antigen was found in the group of elderly females (p less than 0.001) and an excess of the Cw7 antigen in the group of elderly males (p less than 0.001). Of particular significance was the fact that Cw7 belonged in this instance to a phenotypic combination (and most probably to the corresponding haplotype) A1/Cw7/B8/DR3 which was found significantly increased in male nonagenarians (p less than 0.001). These results support the hypothesis that certain HLA haplotypes are associated with survival advantage.

In order to identify factors other than the HLA-linked susceptibility involved in the risk to the siblings of diabetic children, we compared the age at onset, birth order and HLA genotypes in 23 IDD multiplex (Mx) families and 140 simplex (Sx) families. The results showed a relationship between age and recurrence risk in sibs. Probands (= 1st affected sibs) of Mx families were significantly younger at onset (5.8 +/- 0.8 yr) than probands of Sx families (9.0 +/- 0.4 yr, p less than 0.01). The 2nd affected sibs of Mx families, although affected at an older age (12.4 +/- 1.6 yr), had been significantly younger at the time of the proband's onset, than the siblings who have remained unaffected (6.2 +/- 1.2 vs 11.5 +/- 0.5 yr, p less than 0.01). Probands of Mx families were more often the first born and probands of Sx families, more often came later in the birth order (p less than 0.02). Prevalence of IDD was higher among siblings born after than among those born before the proband (12% vs 4%, p less than 0.002). The HLA haplotype distribution in unaffected siblings deviated from the random assortment in relation to birth order, showing an excess of HLA-identical sibs born before the proband and, inversely, an excess of non-identical sibs born after the proband. The results suggest that age-dependent factors are likely to increase the penetrance of HLA-linked IDD susceptibility. These factors could be related to the environmental insult. However, a genetically determined form of type I diabetes characterized by higher familial penetrance and earlier onset cannot be ruled out.

Etiological and epidemiological studies of triploid and hydatidiform molar conceptuses were done using HLA polymorphism. The segregation of HLA markers allowed to know the etiology of 25 triploidies and 19 hydatidiform moles. Five other moles and a post molar choriocarcinoma were also studied by molecular hybridization. This confirms that triploidies in about 3/4 of the cases involved two sets of paternal chromosomes mainly by di-sperm. Hydatidiform moles from Algeria, France and Senegal were all of androgenic origin excepted for one case. DNA analysis of the choriocarcinoma demonstrated the presence of a paternal marker suggesting for this case a direct cellular lineage from the mole. Positive associations with HLA A 28 and B 7 were found which could be related to gametogenesis-fecundation dysfunction. A slight excess of antigens shared by parents of triploidies was shown. This was not observed for parents of hydatidiform moles but when they shared HLA antigens a preferential inheritance in the mole of the shared specificities was observed. This relative compatibility of the molar conceptus with the mother may be an element of the process that prevent its early rejection.

HLA typing of 100 blood samples was carried out in an unrelated Tunisian population, chosen according to its aptitude in giving blood and therefore, in undergoing a complete medical examination. The results obtained show a strong similarity with those already recorded by H. Betuel and coworkers. Two haplotypes are in linkage disequilibrium; which are found in the Turks and the Sardinians.

HLA markers (A, B, C, DR loci) were determined for the members of 52 unrelated families with at least one child suffering from congenital adrenal hyperplasia due to 21 hydroxylase deficiency, permitting genotyping. The gene frequencies of the 52 index cases were compared with those obtained from the patients' normal haplotypes and with those of a control reference panel. No significant differences were observed, except a clear decrease in the frequency of HLA-B8 among the haplotypes that carry the gene for congenital adrenal hyperplasia.

The prevalence of ICA and CFICA in relation to HLA-DR genotypes was analyzed in 107 insulin-dependent diabetic (IDD) patients with a duration of IDD ranging from onset to 30 years, and 247 nondiabetic first-degree relatives. In 46 patients tested at onset of IDD, the prevalence of ICA and of CFICA was 61 and 50%, respectively; with the longer duration of IDD, the prevalence of CFICA decreased more rapidly than that of ICA. No significant association was observed between ICA and any HLA allele in patients tested from onset till 2 years after onset. However, a higher prevalence of ICA was found in DR3 positive patients with a duration of IDD of more than 2 years (p less than 0.02). Among the healthy relatives, the prevalence of ICA was 7% in parents and 13% in siblings; one out of 101 siblings had CFICA. No association was found between ICA and any HLA marker. The presence of ICA in sibs was independent of the number of haplotypes they shared with the IDD proband: among the 13 ICA-positive healthy sibs, one shared 2 haplotypes, nine shared 1 haplotype and three shared no haplotype with the proband, which is not different from the random distribution.

A simulation of the matching for HLA-DR, B, A loci was achieved by taking into account the practical situation of kidney transplants for 89 donors and 322 potential recipients. A complete HLA-DR identical graft was theoretically possible in 52% of the cases. Conversely, if the identity for B or particularly the identities for both B and A were sought, the necessary pool of recipients would have to be much larger, requiring intense international cooperation.

Seventy IDDM patients (insulin-dependent diabetics), 48 females and 22 males, most of them adults at the onset of diabetes, and suffering from at least one other associated autoimmune manifestation (AAM) were studied for HLA A,B,C, DR markers and Bf, C4 complement components. Comparisons were made with 108 normal controls and a series of 287 IDDM patients with juvenile onset (under 25 years) and no patent other autoimmune disease. The increase in frequency of HLA-B8 among IDDM patients with AAM was confirmed (36% versus 20% in controls) (p less than 0.04). The frequency of DR4 among diabetics with AAM (33%) was not significantly different from the normal frequency (27%), and the allelic combination DR3/4 was found in only 13% of IDDM with AAM. Corresponding frequencies in patients with IDDM alone were 66% for DR4 and 34% for DR3/4 (p less than 10(-6) and 10(-3) respectively). These results confirm the heterogeneity of IDDM and support, by genetic arguments, the concept of overlapping entities. The hypothesis of a common background of autoimmunity associated with B8 DR3 can be postulated, while the organ specific target process should be associated with various DR alleles.

The sera of 74 diabetic patients and of their first degree relatives were studied for 4 different types of islet cell antibodies (ICA, CFICA, ICSA, ICACT) and the results were compared according to the HLA genotype. Seventy two percent of the patients' sera were positive for at least one type of auto-antibody. Two types of auto-antibodies were observed: anticytoplasmic antibodies (ICA and CFICA), and surface antibodies (ICSA and ICACT). The different types of antibodies were not associated with any HLA-DR. These antibodies were also present in 15% of the healthy relatives. The fact that they were detected in the sera of siblings sharing no haplotype with the proband, some of them bearing neither DR3 nor DR4 antigens, suggests that non HLA linked genes and/or environmental factors partly control their secretion.

One hundred and thirty-three insulin-dependent diabetic (IDD) patients were genotyped for HLA-A, B, C, DR and 123 of them for Bf. The study shows a relationship between these genes and the age of onset of diabetes and emphasizes the possible heterogeneity of the disease. When patients under the age of 0 years at the onset of the disease were compared with those aged 11 to 29 years, a significant excess of HLA-B18 (41% versus 15%, p less than 0.001) and BfF1 (40% versus 21%, P less than 0.05) was observed. The haplotype B18, BfF1 was also more frequent in the first group of patients (haplotype frequency 18% versus 6%, p less than 0.02). The frequency of the whole haplotype Cw5, B18, BfF1, DR3 and of its segment BfF1, DR3, and the strength of the gametic associations between these alleles were much higher in IDD patients than in non-diabetic controls, irrespective of the age of onset of their diabetes. The association between early age of onset of IDD and the B18, BfF1 haplotype independently of DR3 (no association between age and DR3) suggests that a factor influencing the onset of the disease in young children could be under the control of (1) gene (s) in linkage disequilibrium with B18 and/or BfF1.

The survival rate among 458 cadaver kidney grafts with "full-house" four HLA-A,B antigens detected in the donor, three being identical to those of the recipient and only one mismatched, was studied specifically in relation to antigen incompatibility. At the A locus, the A3 incompatibility was associated with a lower transplant survival (44% at 2 years) than all of the others, and particularly more than the A11 (80% at 2 years) (P less than 0.003 but without significance after correction multiplying by the number of tested alleles). At the B locus, there was no significant difference in survival rate among the alleles. These results are only preliminary and need confirmation based on longer series permitting possible cross-reactions which exit between donor and recipient antigens to be taken into consideration.

HLA A and B typing was performed in 86 unrelated obese subjects and in 10 families including at least one parent and one obese sibling. The results in the series of unrelated subjects show no significant difference in antigen frequencies as compared to the control series. However in the 10 families studied, a group of 5 families is characterized by a high penetrance of obesity. In this group, inheritance of obesity seems to be transmitted through a dominant mode, and antigen B 18 appeared 4 times out of 5. The possible existence of a genetic form of obesity is considered in the discussion.

In 94 children suffering from cystic fibrosis, no abnormal frequencies of HLA markers of the A and B locus were observed in comparison with the distribution of these antigens in control series. Furthermore, the HLA genotypes of seven pairs of diseased sibs are incompatible with the hypothesis of a closed linkage between CF--an autosomal recessive transmitted disease--and HLA.

Of more than 500 diseases or syndromes studied for HL-A markers, more than 40 are known to be associated with an allele of class I, II, or III. Seven are linked to the HL-A region: six are recessive (idiopathic hemochromatosis, C2, C4A, and C4B deficiencies, congenital and late-onset deficiencies) and one is dominant (spinocerebellar ataxia). In addition, insulin-dependent diabetes mellitus is also linked to HL-A with more than one single locus. HL-A typing is of practical interest for diagnosis of ankylosing spondylitis by B27 antigen determination and for prevention of idiopathic hemochromatosis by genotyping of siblings of the index case. Prenatal diagnosis of 21-OH deficiency by genotyping fetal cells permits genetic counseling. Indeed, the discovery of the relationship between HL-A and disease can be considered a new approach to medical genetics. Extensive use of HL-A technology will probably allow better prediction of risk and may elucidate the mechanisms of certain diseases. For the first time the study of one single immunogenetic system may have a significant effect on public health through the possibility of wide-scale prevention.

A detailed gene marker study was performed on a partial 6p trisomic child resulting from a balanced maternal translocation t (2;6) (p 2505; p2105). HLA typing and mixed lymphocyte reaction showed that the breakpoint on chromosome 6 was located within the HLA gene cluster, allowing an accurate location of the D determinants. Localization of the P blood group locus within the region 6 p 2105 to 6 p ter was excluded.

We propose a new method to analyse data on HLA associated diseases. The method uses the simultaneous information on the marker associations and segregation with the disease. It may also take into account the differential risk of being affected for specific relatives of a patient as well as the differential HLA haplotype sharing according to the marker genotype of the patient. It is based on the principle of minimization of a sum of independent chi-squares. It allows us to test the goodness-of-fit of various models in an easy and economical way. The method is applied to a sample of 269 French IDDM patients and their relatives leading to the rejection of models with one locus closely linked to HLA with two and three alleles.

HLA-A and B phenotypes were determined in 38 patients with Cushing's syndrome (29 Cushing's diseases, 9 others etiologies) and compared with those of 591 normal subjects. No significant differences was found in the HLA antigens distribution in the two groups. However the series is too small to definitely support the hypothesis of a lack of correlation with HLA. Indeed, the increased prevalence of B8 antigen in Cushing's disease (28%) as compared to that in the control group (17%), although not statistically significant, emphasizes the need for further typing of HLA-DR, particularly RDw3 which is known to be linked to D8 through desequilibrium and associated with several autoimmune diseases.