[Ethical aspects of biological sample banks]

Numerous activities in the domain of epidemiology require the constitution or the use of biological sample banks. Such biobanks raise ethical issues. A number of recommendations are applicable to this field, in France and elsewhere. Major principles applicable to biobanks include the respect of person's autonomy, the respect of human body, the respect of confidentiality. These principles are translated into practices through the following procedures: relevant information to the persons regarding their sample management prior to informed consent, opinion of an independent ethics committee, actual implementation of conditions for protecting samples and data. However, although those principles may appear quite simple and obvious, in the context of a largely international practice of research and given the large variety of biobanks, it is not always obvious for researchers to find their way. The attitudes vary between countries, there are numerous texts for various types of biobanks, the same texts raise different interpretations in different institutions, there are new ethical opinions expressed, and mainly the novelty of questions raised by the uses of samples that are possible today, especially in genetics, and were not foreseeable at the time of sampling make the field difficult in practice. This article reviews the types of biobanks, the relevant ethical issues. It also underlines the still unclear or ambiguous situations using some examples of practical situations.

Evidence of at least two type 1 diabetes susceptibility genes in the HLA complex distinct from HLA-DQB1, -DQA1 and -DRB1

Susceptibility to, and protection against development of type 1 diabetes (T1D) are primarily associated with the highly polymorphic exon 2 sequences of the HLA class II genes: DQB1, DQA1 and DRB1. However, several studies have also suggested that additional genes in the HLA complex influence T1D risk, albeit to a lesser degree than the class II genes. We have previously shown that allele 3 of microsatellite marker D6S2223, 4.9 Mb telomeric of DQ in the extended class I region, is associated with a reduction in risk conferred by the DQ2-DR3 haplotype. Here we replicate this finding in two populations from Sweden and France. We also show that markers in the HLA class II, III and centromeric class I regions contribute to the DQ2-DR3 associated risk of T1D, independently of linkage disequilibrium (LD) with both the DQ/DR genes and the D6S2223 associated gene. The associated marker alleles are carried on the DQ2-DR3-B18 haplotype in a region of strong LD. By haplotype mapping, we have located the most likely location for this second DQ2-DR3 haplotype-modifying locus to the 2.35 Mb region between HLA-DOB and marker D6S2702, located 970 kb telomeric of HLA-B.

Cost-effectiveness analysis of two strategies for typing unrelated donors for bone marrow transplantation in France

Large Registries of HLA-typed potential volunteer donors have been set up in numerous countries to find HLA-matched unrelated bone marrow donors. This study compared two strategies medicoeconomically for HLA typing in the context of a bone marrow volunteer donor Registry. It investigated the cost-effectiveness of strategies before and after the French Registry was modified in January 1999: HLA- A, B typing only at registration and secondary HLA-DR typing on part of the Registry (AB strategy) vs. typing at once for HLA- A, B, and DR (ABDR strategy). The point of view considered was that of payers, French typing tariffs with a 5% discount rate were used, effectiveness was defined as identification of at least one donor with no HLA-A, B, DR incompatibilities for a given recipient (compatible potential donor), the observation period was 9 months, and the judgement criterion was the differential cost-effectiveness ratio. The ABDR strategy identified 94.7% (142/150) of compatible potential donors. The differential cost-effectiveness ratio between the two strategies was 387,005 francs (Euro 58,995) for one supplementary compatible potential donor. Compared with a "do nothing" policy, the ratio was 3,744,087 francs (Euro 570,745) for the AB strategy vs. 576,136 francs (Euro 87,826) for the ABDR strategy. The ABDR strategy is thus more effective but also more costly than the AB strategy. Nevertheless, because of its lower effectiveness and the size of available ABDR Registries, the AB strategy will become obsolete.

Interferon gamma gene polymorphism and susceptibility to, and severity of, rheumatoid arthritis

A strong association between an interferon gamma (IFN-gamma) gene polymorphism and rheumatoid arthritis susceptibility and severity has been reported in a case-control study. We investigated this polymorphism in 103 patients with early rheumatoid arthritis and 130 controls. Severity of rheumatoid arthritis was measured after 4-year follow-up with a validated radiographic score. The median radiographic score in patients increased from 1 (IQR 0-4) to 11.5 (2-35) over the 4-year follow up. The distribution of IFN-gamma alleles did not differ between patients and controls, and the distribution of radiographic scores did not differ among patients carrying the different IFN-gamma alleles. We have failed to confirm the association between the IFN-gamma gene polymorphism and rheumatoid arthritis susceptibility or severity.

West Mediterranean islands (Corsica, Balearic islands, Sardinia) and the Basque population: contribution of HLA class I molecular markers to their evolutionary history

The genetic structure of Balearic islands (Corsica and Sardinia), situated on the same trans-Mediterranean maritime routes and having very similar histories, were compared and their position among the neighbouring Caucasian populations was inferred. For this purpose, three HLA loci (HLA-A, -B and -Cw) were typed at the DNA level in these populations and the allelic and haplotypic frequencies were estimated. Because previous studies have shown common genetic features in the Sardinians and Basques, HLA-Cw molecular typing was also performed in a sample of French Basques in order to establish the haplotypic structure of this population for a more accurate comparison with the three others. By its allelic composition, the Corsican population has an intermediate position between the two other islander populations. Its close relationship with the Sardinian population, however, was clearly revealed by the phylogenetic analysis which also suggests a proximity with eastern Mediterranean peoples, whereas the Balearic islands are more narrowly related to Spain and western Europe. Peculiarities were observed in the distributions of some common haplotypes in the populations of the islands that confirm the results of the phylogenetic analysis and could be related to their history. Noteworthy is the presence of the HLA-A30-Cw*0501-B18 haplotype at frequencies approximately 2% in Corsica and the Balearic islands, yet the estimated frequencies of this haplotype are much lower than in the Sardinian and Basque populations.

Microsatellites in the HLA region: 1999 update

In this third update of a series of reviews on microsatellites in the HLA region or close to it we report 155 microsatellites, corresponding to 51 newly described markers, in addition to the 103 reported in the 1997 and 1998 reviews. This work is based both on a literature review and on data publicly available in molecular databases on the internet (http://www.gdb.org; http://bioinfo.weizmann.ac.il/cards/; http://cedar.genetics.soton.ac.uk/) up to September 1999. Thanks to numerous studies involving major histocompatibility complex (MHC) microsatellites, documentation on HLA region is proposed, including information on microsatellites described through MHC sequence projects and presenting documented location, polymorphism and amplification condition, together with additional information on previously described microsatellites when available and information on data in the literature regarding gametic associations between HLA region loci and alleles and microsatellite alleles. As basic information are presented various documents: i) a table showing the following characteristics of the 155 microsatellites: name, localisation, polymorphism, primer sequences, reference; ii) an integrated map of some HLA region genes and the 155 microsatellites considered; and iii) a summary table on HLA and microsatellites association patterns. In addition, an overview on HLA microsatellite analysis application is presented, with a special focus on disease genetics studies in the form of recent references where the use of microsatellites of the HLA region was a key tool. This review aims at providing the human immunogenetics community with a tool for helping optimal choice of microsatellites to be used in various studies.

Genetic structure of north-west Africa revealed by STR analysis

We have analysed a large set of autosomal short tandem repeat (STR) loci in several Arabic and Berber-speaking groups from north-west Africa (ie Moroccan Arabs, northern-central and southern Moroccan Berbers, Saharawis, and Mozabites). Two levels of analysis have been devised using two sets of 12STR loci, (D3S1358, vWA, FGA, THO1, TPOX, CSF1PO, D8S1179, D21S11, D18S51, D5S818, D13S317 and D7S820) and 21 (the former set plus D9S926, D11S2010, D13S767, D14S306, D18S848, D2S1328, D4S243, F13A1, and FES/FPS). For each set, data for a number of external reference populations were gathered from the literature. Several methods of analysis based on genetic distances (neighbour-joining trees, principal coordinate analysis, boundary detection), as well as AMOVA, showed that genetic differentiation among NW African populations was very low and devoid of any spatial pattern. When the NW African populations were grouped according to cultural or linguistic differences, the partition was not associated with genetic differentiation. Thus, it is likely that Arabisation was mainly a cultural process. A clear genetic difference was found between NW African populations and Iberians, which underscores the Gilbraltar Straits as a strong barrier to genetic exchange; nonetheless, some degree of gene flow into Southern Iberia may have existed. NW Africans were genetically closer to Iberians and to other Europeans than to African Americans.

Microsatellites in the HLA region: 1998 update

In order to update the review published in Tissue Antigens in 1997, we present here a new overview on microsatellites in the HLA region, including additional information, with focus on the following points: * Description of 103 microsatellite characteristics in the HLA region, 50 markers having been newly described since 1996. * An integrated map of the HLA region, including microsatellites and some HLA genes, revealing an important microsatellite density in the MHC (Class I, Class II and Class III regions). * A synthesis of microsatellite analysis in disease studies, summarizing results of microsatellite approaches in 24 pathologies, including autoimmune diseases, HLA-associated or HLA-linked diseases and cancers. * Other applications of HLA region microsatellites in population or transplantation studies.

HLA class II allele and haplotype frequencies in Ethiopian Amhara and Oromo populations

HLA class II alleles were identified in 181 healthy unrelated Ethiopian children of both sexes and in 350 European controls from the South of France. The Ethiopian individuals belonged to the two major ethnic groups of the country: Oromo (N=83) and Amhara (N=98). In both panels, genetic polymorphism of HLA class II alleles was analysed for the first time by molecular typing of DRB1, DQA1 and DQB1 loci. Allelic and phenotypic frequencies were compared with those of European controls and other African populations. Construction of HLA class II three-locus haplotypes was also performed. The study revealed some differences between the two groups. Characteristic features of Central and North African populations appeared on the Ethiopian HLA genotypes. Surprisingly, DRB1*11 presented one of the lowest gene frequencies in both Ethiopian ethnic groups in contrast to Europeans and West Africans. Furthermore, this decrease was more marked than those observed using serological techniques in other geographically close East African countries. Oromo and Amhara only showed minor differences in spite of their different origins and histories. One significant difference consisted of a lower DRB1*01 gene frequency in Oromo as reported in most West African people. Some new or rare haplotypes were also observed in the Oromo group. Our results underline the distinctive features of the Ethiopian populations among the few HLA genotyping data available for East African groups and emphasise the major interest of such investigations in this region of Africa.

Insulin VNTR allele-specific effect in type 1 diabetes depends on identity of untransmitted paternal allele. The IMDIAB Group

The IDDM2 type 1 diabetes susceptibility locus was mapped to and identified as allelic variation at the insulin gene (INS) VNTR regulatory polymorphism. In Caucasians, INS VNTR alleles divide into two discrete size classes. Class I alleles (26 to 63 repeats) predispose in a recessive way to type 1 diabetes, while class III alleles (140 to more than 200 repeats) are dominantly protective. The protective effect may be explained by higher levels of class III VNTR-associated INS mRNA in thymus such that elevated levels of preproinsulin protein enhance immune tolerance to preproinsulin, a key autoantigen in type 1 diabetes pathogenesis. The mode of action of IDDM2 is complicated, however, by parent-of-origin effects and possible allelic heterogeneity within the two defined allele classes. We have now analysed transmission of specific VNTR alleles in 1,316 families and demonstrate that a particular class I allele does not predispose to disease when paternally inherited, suggestive of polymorphic imprinting. But this paternal effect is observed only when the father's untransmitted allele is a class III. This allelic interaction is reminiscent of epigenetic phenomena observed in plants (for example, paramutation; ref. 17) and in yeast (for example, trans-inactivation; ref. 18). If untransmitted chromosomes can have functional effects on the biological properties of transmitted chromosomes, the implications for human genetics and disease are potentially considerable.

Oligotyping for HLA-DQA, -DQB, and -DPB in idiopathic nephrotic syndrome

Associations of human leukocyte antigens (HLA) with the idiopathic nephrotic syndrome (NS) have mainly been described for alleles of the HLA-DR locus. In the present study the polymorphism of HLA-DQ and -DP at the molecular level was investigated in 167 children with NS (129 steroid-sensitive) using the polymerase chain reaction and sequence-specific oligonucleotides in a French and a German cohort. HLA-DR typing was also performed by classical serology. In steroid-sensitive patients we observed an increased frequency of the alleles HLA-DQA1*0201 and -DQB1*0201 in both populations with relative risks ranging from 3.8 to 8.5 (Pb < 0.01 to Pb < 0.00001 after Bonferoni's correction). In contrast, the frequency of HLA-DQA1*0102 and DQB1*0602 was significantly decreased. In children with frequent relapses the HLA associations were generally more pronounced than in those with infrequent or no relapses. Applying logistic regression analysis, a nephrotic child bearing DQA1*0201 or DR7 was five times more likely to be in the steroid-sensitive group of patients than in the steroid-resistant group compared with nephrotic children not bearing one of these alleles. These HLA alleles therefore seem to be useful indicators of a steroid-sensitive frequently relapsing course of NS. No associations with DPB alleles were observed, which narrows the region genetically involved in the disease susceptibility to the DR-DQ region. Steroid-resistant NS was not associated with HLA.

Microsatellites in the HLA region: on overview

Microsatellites are repeats of a DNA base motif (1-6 bp, mostly CA repeats) up to 100 times; they are distributed regularly all over the genome. Many of them are polymorphic and their high polymorphism is based upon a variable number of repeats. They are widely used for genetic mapping, linkage analysis, population genetics, evolutionary studies and in forensic medicine. Such markers have also been described in the HLA region since 1991, and a growing interest in their potential applications is being expressed. The aims of this review are: 1) to outline the presently available information from literature and molecular databases concerning 53 microsatellites in the HLA region (localization, type of repeat, number of alleles, heterozygosity, primers used for amplification); 2) to address the question of technical pitfalls when using such markers; 3) to discuss specific features such as their mutation rate (10 (-3) to 10 (-6), which is higher than that reported for HLA genes, and their linkage disequilibrium with HLA alleles; 4) to present an integrated map of microsatellites and genes of this region; and 5) to provide a synopsis of their different applications in HLA-related fields (disease studies, population genetics, recombination point studies, HLA region mapping, transplantation) along with perspectives for future use. Although some HLA region microsatellites have already been applied to the analysis of more than 10 diseases, it is now evident that their use in population genetics and the determination of genomic compatibility in bone marrow transplantation represent growing areas of application.

Evidence of a non-MHC susceptibility locus in type I diabetes linked to HLA on chromosome 6

Linkage studies have led to the identification of several chromosome regions that may contain susceptibility loci to type I diabetes (IDDM), in addition to the HLA and INS loci. These include two on chromosome 6q, denoted IDDM5 and IDDM8, that are not linked to HLA. In a previous study, we noticed that the evidence for linkage to IDDM susceptibility around the HLA locus extended over a total distance of 100 cM, which suggested to us that another susceptibility locus could reside near HLA. We developed a statistical method to test this hypothesis in a panel of 523 multiplex families from France, the United States, and Denmark (a total of 667 affected sib pairs, 536 with both parents genotyped), and here present evidence (P = .00003) of a susceptibility locus for IDDM located 32 cM from HLA in males but not linked to HLA in females and distinct from IDDM5 and IDDM8. A new statistical method to test for the presence of a second susceptibility locus linked to a known first susceptibility locus (here HLA) is presented. In addition, we analyzed our current family panel with markers for IDDM5 and IDDM8 on chromosome 6 and found suggestions of linkage for both of these loci (P = .002 and .004, respectively, on the complete family panel). When cumulated with previously published results, with overlapping families removed, the affected-sib-pair tests had a significance of P = .0001 for IDDM5 and P = .00004 for IDDM8.

New highly polymorphic microsatellite marker in linkage disequilibrium with HLA-B

The difficulty of molecular typing of the HLA class I genes and the relevance of the genes of this region to disease susceptibility and transplantation have provided an impetus to develop useful typing markers. We have characterized by polymerase chain reaction analysis a new highly informative CA repeat localized approximately 25-kb centromeric to the gene HLA-B and 10-kb telomeric to the gene MICA. Twelve alleles defined by length were found in a sample of French Basques, with the PIC being 0.82. A detailed haplotype analysis was performed to investigate the association between this microsatellite and two others markers of the region (HLA-B gene and TNF region microsatellite). The 10 haplotypes with the highest estimated frequencies show evidence of a gametic association or linkage disequilibrium. A very strong association between the expressed HLA-B polymorphism and microsatellite alleles was also revealed in this sample and confirmed in the workshop cells lines of the Fourth Asia-Oceania Histocompatibility Workshop. This marker can be used in the fine mapping of this region and the association with some alleles of HLA-B may allow the replacement of HLA-B typing at least in a preliminary study. Moreover, these studies support the hypothesis of a high mutability for large alleles in microsatellite loci.

No linkage or association of telomeric and centromeric T-cell receptor beta-chain markers with susceptibility to type 1 insulin-dependent diabetes in HLA-DR4 multiplex families

The T-cell receptor beta locus (TCRB) on chromosome 7q35 was studied as a candidate region for genetic susceptibility to type 1 insulin-dependent diabetes (IDDM). A highly polymorphic microsatellite marker mapping to the TCRBV6.7 gene and a TCRB C-region RFLP were used to genotype the members of a total of 21 multiplex IDDM families from two different geographical areas. There was no evidence to support linkage to either of these markers with IDDM, and conventional two-point analysis excluded linkage to the telomeric end of the TCRB complex, in the region of the highly informative TCRBV6.7 marker. There was significant linkage of IDDM to the class II HLA-D locus with significant lod scores > 3.0 obtained for the HLA-DRB1 and HLA-DQB1 genes. Affected sib-pair (ASP) and transmission disequilibrium (TDT) association tests confirmed these findings.

Genetic diversity in the Iberian Peninsula determined from mitochondrial sequence analysis

We have analysed 302 bp of the first hypervariable region of the mitochondrial D-loop in 271 individuals from different regions of the Iberian Peninsula and 85 individuals from Algeria. The Basque population is significantly different from neighbouring populations in terms of overall levels of diversity. This is because the majority of sequences in the Basques are restricted to the lineage group defined by the CRS (Cambridge Reference Sequence) and its derivatives although, like other Iberian populations, they showed a unimodal distribution of pairwise sequence differences. The timing of divergence of populations within Iberia points to a shared ancestry of all populations in the Upper Palaeolithic. Further genetic subdivision is apparent in Catalonia and Andalusia, with increased genetic diversity in the latter. Lineage diversity comparisons of Iberian populations with European (Tuscan) and North African (Algerian) populations shows the Iberian Peninsula to be more similar to other European populations, although a small number of Iberian lineages can be traced to North Africa.

Localization of the recombination points in a family with two DR/DP recombinations

In a family with a maternal DR/GLO recombination, cellular DP typing showed it to be located between DR and DP. RFLP studies done during the 9th international histocompatibility workshop gave anomalous segregation patterns of DPA and DPB bands that could be interpreted as being due to a second, paternal DR/DP recombination. This assumption was confirmed later by PCR-SSO typing. A more precise mapping has been done by new markers showing the maternal recombination to be within the TAP2 locus and the paternal recombination to be between DQB1 and DQB3. This supports earlier suggestions of a hot spot of recombination in the TAP region. The recombinations involve parental haplotypes that presently show DR/DP linkage disequilibrium in the French population and it is proposed that DR/DP recombinations occur randomly while B/DR recombinations preferentially occur on haplotypes without strong linkage disequilibrium. Existing DR/DP linkage disequilibria in a given population will thus be broken down with time. The mixed lymphocyte culture response towards an isolated DP difference was tested in this and another DR/DP recombinant family. It showed that an alloresponse towards DP may be highly variable and this suggests that it might be important to define the rules for the strength of this reaction and the possible implications for allotransplantation.

Strong association between microsatellites and an HLA-B, DR haplotype (B18-DR3): implication for microsatellite evolution

The HLA haplotype B18-DR3 has a widespread geographical distribution, but has its greatest frequencies in Southern Europe, probably vestigial of the earliest populations of this region, particularly in the Pays Basque and Sardinia. This haplotype is of medical significance, being that most implicated as a factor of risk in insulin-dependent diabetes mellitus. In this study, the closely linked microsatellite markers (TNFa,b,c) in the region of the tumor necrosis factor (TNF) genes have been used in an attempt to subtype this haplotype in the two populations and/or in healthy and diabetic populations. A total of 79 HLA-B18-DR3 haplotypes were analyzed: 54 in Basques (12 from healthy individuals and 42 from diabetics or their first-degree relatives) and 25 in Sardinians (13 from healthy and 12 from diabetic individuals). The TNF haplotype a1-b5-c2 is completely associated with B18-DR3 in both populations. The homogeneity of the B18-DR3 haplotype in two ethnically pure populations implies stability in evolution, which suggests that the mutation rate of these microsatellite markers must be less than is usually assumed (i. e., approximately 5 x 10(-4) per site per generation). Such markers should be powerful tools for studying genetic drift and admixture of populations, but it remains to be established whether this stability is a rule for all microsatellites in HLA haplotypes or whether it is restricted to some microsatellites and/or some HLA haplotypes. The population genetics of those microsatellites associated with HLA B18-DR3 was also studied in a random sample of the Basque population.

Clinical implications from studies of HLA antigens in idiopathic nephrotic syndrome in children

HLA class I and II antigen frequencies were determined in two large cohorts of children with idiopathic nephrotic syndrome (NS) from Southwest France (n = 199) and Southwest Germany (n = 152) and compared with unrelated healthy individuals from the same geographical areas. Strength of HLA association was expressed by the relative risk (RR) estimated by Odd's ratio. We examined 105 steroid-resistant and 242 steroid-sensitive NS patients who were subdivided in non-relapsers, infrequent relapsers and frequent relapsers or steroid-dependent patients. In steroid-sensitive patients significant associations were found with HLA-DR7 (RR 5.1 in French, 3.2 in Germans), -DQ2 (RR 4.7/2.3) and with the phenotypic combination HLA-DR3/DR7 (RR 5.6/7.7). Significant negative associations were encountered with HLA-DR2, -DR6 and -DQ1. The associations were stronger in frequent relapsers/steroid-dependent patients than in infrequent relapsers and were not significant in non-relapsers. In steroid-resistant patients the only significant association found was with the combined occurrence of HLA-DR3/DR7. We propose that in childhood NS tissue typing for selected HLA class II antigens is helpful in prediciting the clinical course.