The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kralovicova, J.
Right arrow Articles by Vorechovsky, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kralovicova, J.
Right arrow Articles by Vorechovsky, I.
The Journal of Immunology, 2003, 170: 2765-2775.
Copyright © 2003 by The American Association of Immunologists

Fine-Scale Mapping at IGAD1 and Genome-Wide Genetic Linkage Analysis Implicate HLA-DQ/DR as a Major Susceptibility Locus in Selective IgA Deficiency and Common Variable Immunodeficiency 1

Jana Kralovicova*, Lennart Hammarström{dagger}, Alessandro Plebani{ddagger}, A. David B. Webster§ and Igor Vorechovsky2,*

* Division of Human Genetics, School of Medicine, Southampton University Hospital, University of Southampton, Southampton, United Kingdom; {dagger} Department of Biosciences at Novum, Karolinska Institute, Huddinge, Sweden; {ddagger} Istituto di Medicina Molecolare Angelo Nocivelli, University of Brescia, Brescia, Italy; § Royal Free Campus, University College London, London, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Selective IgA deficiency (IgAD) and common variable immunodeficiency (CVID) are the most common primary immunodeficiencies in humans. A high degree of familial clustering, marked differences in the population prevalence among ethnic groups, association of IgAD and CVID in families, and a predominant inheritance pattern in multiple-case pedigrees have suggested a strong, shared genetic predisposition. Previous genetic linkage, case-control, and family-based association studies mapped an IgAD/CVID susceptibility locus, designated IGAD1, to the MHC, but its precise location within the MHC has been controversial. We have analyzed a sample of 101 multiple- and 110 single-case families using 36 markers at the IGAD1 candidate region and mapped homozygous stretches across the MHC shared by affected family members. Haplotype analysis, linkage disequilibrium, and homozygosity mapping indicated that HLA-DQ/DR is the major IGAD1 locus, strongly suggesting the autoimmune pathogenesis of IgAD/CVID. This is supported by the highest excess of allelic sharing at 6p in the genome-wide linkage analysis of 101 IgAD/CVID families using 383 marker loci, by previously reported restrictions of the T cell repertoires in CVID, the presence of autoantibodies, impaired T cell activation, and a dysregulation of a number of genes in the targeted immune system. IgAD/CVID may thus provide a useful model for the study of pathogenesis and novel therapeutic strategies in autoimmune diseases.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Although a number of disease-causing mutations have been identified in monogenic primary immunodeficiencies (PIDs)3 in the last 15 years (1), the molecular mechanisms underlying the development of selective IgA deficiency (IgAD) and common variable immunodeficiency (CVID), the most common PIDs in Caucasoid populations, have not been elucidated. Both PIDs have been characterized by a defect in the terminal stage of lymphocyte differentiation, leading to an impaired production of one or more Ig isotypes, resulting in susceptibility to infections (1, 2). Clinically overt CVID necessitates long-term Ig replacement and antimicrobial therapy with considerable health-care expenditures, but treated patients still show a higher mortality than the general population (3).

Although IgAD/CVID do not show Mendelian segregation in families, they exhibit a high degree of familial clustering (4, 5) and marked differences in the population prevalence among ethnic groups, ranging from ~1/500 in Caucasians to 1/18500 in Japanese (for review see Ref.2), supporting a strong involvement of hereditary factors. The association of both PIDs in families (6, 7), individuals (8, 9, 10, 11, 12), and with identical HLA Ags (6, 13, 14) strongly suggested that the two clinically discernible disorders share genetic predisposition. Screening of family members of index cases for serum Ig levels showed that about a third of multiply affected (multiplex) families had both CVID and IgAD in close relatives, typically CVID in theparental generation followed by IgAD in the offspring (7). These observations indicated that a large subset of cases diagnosed as CVID, which is much less prevalent than IgAD (~1/25,000 in the U.K.; A. D. B. Webster, unpublished data), represents a more severe manifestation of a defect in common.

Functional lymphocyte studies in CVID have revealed a number of abnormalities (for review see Ref.15), but the elucidation of a primary defect has been hampered by the complexities of interactions between immune effector cells. This led us to initiate a genetic linkage study to identify chromosome susceptibility loci. We previously found a significant increase of allele sharing on chromosome 6p (7), consistent with MHC associations in case-control (for review see 2) and family-based (7) association studies, providing the proof for the presence of a susceptibility locus designated IGAD1. This evidence was further supported by a parental allele segregation distortion at IGAD1 accompanied by parent-of-origin penetrance differences of IgAD, and by the observation of a higher frequency of anti-IgA Abs (Ab) in females transmitting the disease to the offspring than in female nontransmitters, implicating anti-IgA Ab in familial clustering (7). The evidence for genetic linkage to IGAD1 was stronger in families with anti-IgA-positive female transmitters than in pedigrees with anti-IgA-negative individuals, suggesting a major role of IGAD1 in parent-of-origin penetrance effects and autoantibody production (7). However, the relative significance of this locus in the overall predisposition to IgAD/CVID has not been clear. Likewise, there has been no consensus as to the precise location of IGAD1, with the disease locus placed to the HLA class I/III regions (6, 16, 17, 18, 19) or more centromerically to the HLA class II region (20, 21). Could this decade-lasting controversy (16, 20, 22) be resolved using genetic means?

In the present study, we have typed an extensive set of families with IgAD/CVID using MHC short tandem repeats (STR) and single nucleotide polymorphisms (SNPs), constructed haplotypes in the region, and monitored family-based allelic associations in the MHC. In addition, we have determined HLA specificities in representative carriers of STR haplotypes and provide in this report a reference panel of STR haplotypes/alleles on the most prevalent HLA class II haplotypes in a Caucasoid population. We also show a significant risk of developing IgAD/CVID conferred by homozygous stretches in the class II region. A maximum excess of the observed homozygosity over expected under the Hardy-Weinberg equilibrium (HWE) was in the region identical to that implicated by LD fine-mapping methods, with HLA-DQ/DR as the only expressed genes. In addition, the genome-wide linkage analysis of 101 multiplex families with IgAD/CVID showed the highest excess of allele sharing at 6p and suggested putative non-MHC loci at 4p, 12p, and 14q. Finally, we propose the autoimmune pathogenesis of IgAD/CVID and discuss how the previously observed lymphocyte abnormalities in CVID can be explained in the context of peptide-DQ/DR interactions with TCR.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Ascertainment of probands and families

Diagnosis of patients with IgAD and CVID was established in accordance with accepted recommendations as reported earlier (7, 23). The affection status and the ascertainment of index cases were defined previously (7). Multiplex families were identified through proband by screening blood relatives of index cases for serum Ig levels by nephelometry (7). The measurements were conducted blind to family relationship and affection status. The ethnic origin of multiplex families was shown previously (21). The Ig measurements of samples from outside Sweden were repeated in the Swedish laboratory to guarantee that the same method defined the phenotype and to ensure that a single blood sample could serve both for the determination of the affection status and DNA extraction, decreasing the probability of a laboratory mix-up. Bilineal families, rare cases with previously identified deletions in the Ig genes (24), drug-induced IgAD, and two multiplex families with IgAD in which the phenotype could not be regarded as categorical, were excluded from the study. Although no multiple-case family showed male-to-male inheritance of CVID, suggesting an X-linked defect, CVID samples were analyzed for mutations in BTK (25), AID (J. Kralovicova and I. Vorechovsky, unpublished observations), and HIGM1 (A. D. B. Webster et al., unpublished observations) to exclude misdiagnosed PIDs and/or reveal putative genetic heterogeneity of CVID. Samples from Swedish control trios (father-mother-child) were ascertained as described (7). The study was approved by the Institutional Review Boards.

Sample size

DNA samples were extracted from PBLs as described previously (26) from a total of 210 apparently unrelated families (242 nuclear families), consisting of 101 multiplex families and 109 single-case families with both unaffected parents. Multiplex families comprised 553 family members informative for linkage, 258 of them affected, 94 affected sibling pairs, and 42 nonaffected sibling pairs. The pedigree structure of each multiplex family is shown as a supplemental information.4 In addition, we used a control sample of 256 parent-child trios ascertained through Swedish children with cystic fibrosis and phenylketonurea. These trios would not be expected to show a biased segregation at IGAD1 and were used to estimate the relative risk and attributable fraction conferred by the 5-1-7-6 (AH8.1) haplotype. Assuming the relative risk of developing IgAD for siblings to be of ~50 (the recurrence risk for siblings was estimated to be ~8–9%; see Ref.5), the available sample and STR set had a sufficient power to detect a major predisposing locus.

Marker loci and genotyping

The location of HLA STRs on the physical map of a human MHC and the relative position to other MHC genes is shown in Fig. 1. Oligonucleotide primers and PCR conditions, Mg2+ concentrations, and annealing temperatures are shown as supplemental information or were reported previously (27, 28).4 Allelic sizes, numbers and their frequencies, observed heterozygosity, and polymorphism information content are shown as supplementary information.4 In addition, genotypes at three cSNPs in the BTLNL2 gene (939A->G, 1050G->A, and 1078A->G), located ~20 kb centromeric of LH1 (Fig. 1) (29), were determined using nucleotide sequencing with BigDye terminators (Applera, Norwalk, CT) as described (26) to establish a telomeric limit of the candidate region on a subset of haplotypes. The amplification primers were 5'-CCC CAC CTC ACC TAA G-3' and 5'-AGA GAA ATT GTC CAG GAA CTA-3'.



View larger version (54K):
[in this window]
[in a new window]
  		FIGURE 1. Family-based associations at the IGAD1 candidate region in 242 nuclear families. The map is drawn to scale, reflecting a particular haplotype sequenced (for sequence, see supplementary information).4 Only a subset of MHC genes relative to tested STRs is shown. The STRs used for allelic association are in bold as they were found to show correct Mendelian inheritance, unambiguous clustering and scoring, and the absence of null alleles (28 ). The FBAT program (32 ) was used under the additive and dominant models in biallelic and multiallelic tests. The empirical variance option of FBAT was used for computing Z scores. In all tests, additive models performed better for most marker loci than dominant models.

 
For the genome-wide scan, we used a total of 383 STR loci with an average intermarker genetic distance of 11 cM. Oligonucleotide primers and PCR conditions are available as supplementary information online.4 Genotyping was performed using fluorescence-labeled primers on an ABI 377 DNA Sequencer using the Genescan 672 (version 3.0) and Genotyper (version 2.0) software packages (Applera). Allelic sizes were converted to allele numbers by the ALSIZE module of the Genetic Analysis System (provided by A. Young, University of Oxford, Oxford, U.K.) and the accuracy of conversion was checked manually. A strict global binning of allele sizes was ensured using identical internal size standards (TAMRA 500; Applera).

Linkage analysis and linkage disequilibrium (LD) mapping

The Genehunter (version 2.0) (30) and Allegro (version 1.1) (31) programs were used for computing NPL and Zlr scores, respectively. Haplotypes were constructed using the Genehunter and Simwalk2 algorithms. HLA haplotypes were inspected and corrected manually against the Genotyper plots. Samples from pedigrees with variant or inconsistent alleles on the major susceptibility haplotypes were retyped to confirm the data.

Family-based allelic associations were computed using the extended X option of transmission disequilibrium test (TDT) (7) and family-based allelic association test (FBAT) (32). The trimmed haplotype method was used as implemented in the Trimhap program (33). The numerical and graphical representations of LD in the class I region are available as supplemental material online.4 The pair-wise measures of LD in the class II/III regions were shown previously for the analyzed sample (28).

Risk calculations and HWE testing

Population-attributable risk, or a fraction of affected cases in the studied population that would have not occurred had the risk factor been absent, was estimated as f(RR-1)/RR (34), where f is the proportion of cases carrying the risk haplotype and RR is the relative risk of developing IgAD/CVID.

The excess of homozygosity over expected was calculated from the observed (Ho) and expected (He) values in affected individuals as a parameter F = (Ho - He)(1 - He) (35). He was computed from allelic frequencies assuming HWE.

Mutation analysis and sequence-based HLA typing

PCR single-strand conformation polymorphism analysis was used for mutation detection in non-MHC genes as reported previously (26). Oligonucleotide primer sequences for PCR amplification are shown as supplementary information.4 Two single-strand conformation polymorphism gel-running conditions were used for the ICOS and BST1 screening. HLA specificities were determined using sequence-based typing as described (28) in representative carriers of STR haplotypes to relate HLA specificities and STR haplotypes in close proximity of the HLA class II genes.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Family-based associations and fine mapping at IGAD1

To narrow down the IGAD1 candidate region, we genotyped 879 individuals from 210 families at 36 HLA STR loci selected from a large number of HLA STRs for correct Mendelian inheritance, reliable size clustering, and the absence of null alleles. We first determined the degree of family-based allelic association with IgAD/CVID using the FBAT algorithm (Fig. 1). We found the most significant associations in the telomeric part of the class II region, decreasing rapidly in the area of known recombination hot spots centromeric of HLA-DQB1, but less precipitously in the opposite direction (Fig. 1), partly reflecting the progression of LD in the region (28, 36). These results illustrated that a mere degree of association does not permit a safe placement of the disease gene to a smaller area in this high LD, selection-driven, and gene-rich region, prompting us to study haplotypes in pedigrees.

We next used the trimmed haplotype method using the analysis of 2-, 3-, and 4-locus STR haplotypes that could have descended from the same ancestral founder but have been trimmed in succeeding generations (33). We show these results as a comparison of two IGAD1 subregions (Table I). Region 1 contains genes encoding HLA-DQ and -DR molecules, whereas region 2, suggested in previous studies (6, 16, 17, 18, 19), is in the telomeric part of the class III region and spans a similar physical distance (Fig. 1). We found that the region 1 haplotypes were always ranked at the top of trimmed haplotype tables with highly significant likelihood ratio/regression/model-dependent statistics and constituted the majority of the 5% best performing ancestral haplotypes tested (Table I). The interval between G51152 and 9-99431, which contains only HLA-DQB1, -DQA1, -DRB1, and -DRA (-DRB3 is present on HLA-DQB1*0201, DRB1*03011 haplotypes) and few HLA pseudogenes (Fig. 1), was implicated by all Trimhap routines as the best candidate, including model-dependent algorithms with varying program parameters (Table I). For 2-locus haplotypes, the regression analysis indicated the DQCAR-DQCARII interval telomeric of DQB1 as the most likely location (Fig. 1), whereas statistics 1 and 3 favored the region between G51152-DQCAR, containing only DQB1, suggesting limitations of the method for such small regions. The HLA class I haplotypes were at the bottom of trimmed haplotype tables, consistent with weaker associations in TDTs (data not shown), thus, not supporting the presence of a major disease gene. Consistent with the trimmed haplotype analysis, the extended TDT (37) using the same data set indicated that the most likely location of the disease gene is in the HLA-DQ/DR region between STRs G51152 and LH1 (Figs. 1 and 2A).


View this table:
[in this window]
[in a new window]
  		Table I. Trimmed haplotype analysis of 210 families with IgAD/CVIDa

 


View larger version (62K):
[in this window]
[in a new window]
  		FIGURE 2. Fine-scale mapping of IGAD1. The extended TDT (A) and a proportional excess of homozygosity (F) (B) in the HLA class II and class III regions.

 
Homozygosity at HLA-DQ/DR

Because previous association studies suggested an increase of homozygosity in affected individuals vs controls both for HLA-DQ/DR (38, 39) and HLA STRs (21), we estimated the relative risk of IgAD/CVID conferred by homozygous stretches at adjacent STRs and examined which region was associated with the highest risk. We chose to test 4- and 5-STR sliding windows because the average probability of being homozygous by chance was sufficiently low (~2 x 10-3 and 3 x 10-4, respectively), minimizing random homozygosity at adjacent STRs. Among 210 probands, we found 34 (16.2%) cases (17 sporadic and 17 familial; 15 CVID and 19 IgAD) homozygous at four or more adjacent marker loci compared with only 7.7% (17/220) in unaffected founder controls ({lambda}2 = 7.3, p < 0.01, odds ratio = 2.3, 95% confidence interval 1.3–4.2), indicating that homozygosity at adjacent STRs at IGAD1 confers susceptibility. We also identified 46 such defined homozygotes among a total of 258 affected individuals in multiplex families (17.8%) as compared with only 22 in 231 (9.5%) nonaffected relatives, consistent with a higher number of susceptibility haplotypes segregating in related family members.

We next examined the distribution of homozygous regions across IGAD1. We found that the interval between G51152 and 9-99431 (region 1), for which the expected homozygosity was only 6 x 10-4, was encompassed in 31/34 index cases homozygous in at least four adjacent HLA STRs. In contrast, only three probands showed such homozygosity in the HLA class III region and not in the HLA class II region, with only two of them between 24-140297 and MIB (region 2, Fig. 1). Ten of 34 probands were homozygous for the whole tested region. We also analyzed all affected homozygotes in families and found a similar bias toward the G51152/9-99431 interval (Fig. 3a, b, d, and f), thus strongly implicating region 1 as containing disease gene(s).



View larger version (142K):
[in this window]
[in a new window]
  		FIGURE 3. Alignment of susceptibility haplotypes in affected family members. Shared haplotype portions are highlighted in gray (haplotypes extending into the class I region) or blue (haplotypes terminating in the class III/I regions). Variant alleles/haplotypes are in light background colors. Haplotypes were constructed using the Genehunter program (version 2.0) and edited manually after retyping. Family designation starting with cv denotes a multiplex family, whereas the remaining cases were sporadic. CVID patients are boxed; the remaining family members had IgAD. We show only homozygous carriers of the most prevalent susceptibility haplotypes (a and d) and both homozygous and heterozygous carriers for less-prevalent haplotypes (b, c, e, and f). Full haplotypes in each family member are available as supplementary information online.

 
For the five-marker sliding window, the frequencies of homozygous index cases vs controls were 13.1 and 5.9% (p < 0.01). Of 26 homozygous probands, only a single case exhibited homozygosity in the telomeric part of the class III region and not in the HLA-DQ/DR region. The frequencies of unaffected homozygous individuals (7.7% for 4-STRs and 5.9% for 5-STRs) were comparable to ~6–8% homozygosity observed in large numbers of population controls typed for HLA-DR or -DQ specificities (38, 39).

To exclude that the excess of homozygosity in the class II region was due to other factors than the segregation of disease alleles in affected family members, we constructed HLA STR haplotypes for 128 control "child-father-mother" trios ascertained through Swedish children with cystic fibrosis and phenylketonurea. In contrast to affected individuals, we found no bias in the distribution of homozygous stretches toward the HLA class II region. All haplotypes in IgAD/CVID families and control trios are shown as supplementary information online.4

If there is evidence for a significant increase of homozygosity in affected individuals as compared with unaffected controls at IGAD1 and if there is a support for an additive disease model (Fig. 1), then testing for deviation from the HWE may be helpful in fine localization of disease genes in traits with presumed genetic heterogeneity (35, 40). To measure this, we calculated the excess of homozygosity over expected in affected individuals (Fig. 2B). We observed the largest proportionate increase of homozygosity (F) at 9-99431, with the Fmax value between this marker and DQCARII, implicating the same location as by LD and haplotype analyses, and providing independent support for HLA-DQ/DR as the disease susceptibility locus.

We then determined the nucleotide sequence of gene-coding and promoter regions of HLA-DRB3 (present only on ancestral haplotypes AH8.1), -DQB1, -DQA1, -DRB1, and -DRA in affected homozygotes. As expected, we observed no sequence changes as compared with the published HLA polymorphisms, consistent with our proposal of the autoimmune pathogenesis of IgAD/CVID.

STR vs HLA haplotypes

To relate STR haplotypes to HLA specificities, we determined the nucleotide sequence of exons 2 and 3 of HLA-DQB1 and -DRB1 in 86 individuals carrying representative STR haplotypes between G51152 and 9-99431 (Table II). We identified only 11 (6.4%) variant alleles, indicating that the majority of tested STR haplotypes carried identical high-resolution HLA-DQB1, -DRB1 haplotypes. Only a single allelic variant was found at HLA-DQB1 for the most predisposing haplotypes, suggesting a very high predictive value of STR haplotyping for deducing HLA-DQ/DR specificities. Using multipoint TDTs, we ordered STR haplotypes from the most disease predisposing to the most protective (Table II). The most protective haplotype HLA-DQB1*0602, DRB1*15011 was observed in trans with the majority of STR haplotypes in affected individuals, except for those carrying DQB1*06 (Table II). The absence of deduced DQB1*0602/DQB1*06xx genotypes among 367 affected individuals suggests that DQB1*0602 homozygosity may confer absolute protection from IgAD/CVID and indicates the existence of a dosage effect both for protective and susceptibility HLA-DQ/DR haplotypes. In addition, we found no obvious preference for the susceptibility/protective haplotypes to occur in cis/trans combination with each other in affected individuals, consistent with a less stringent HLA restriction than that found, for example, in celiac disease, which occurs almost exclusively on the background of the HLA-DQB1*0201 alleles.


View this table:
[in this window]
[in a new window]
  		Table II. Sequence-based HLA typing of representative STR haplotypes

 
Because the HLA-DQB1*0201, -DRB1*03011 haplotypes could be unambiguously predicted using STR haplotypes between G51152 and 9-99431 (Table II), we estimated the relative risk of developing IgAD/CVID conferred by the 5-1-7-6 haplotype in Swedish cases and controls (n = 256). We found that the proportion of the 5-1-7-6 carriers was increased among IgAD/CVID probands as compared with controls (40.7 vs 16.0%, odds ratio 3.60, 95% confidence interval 2.09–6.13, {chi}2 = 22.7, p < 5 x 10-5). This figure translates into a population-attributable fraction of 29.4%, indicating that this haplotype is a major risk factor for IgAD/CVID in the analyzed population. The haplotype frequency was not significantly different between Swedish female and male probands (22.6 vs 22.7%, respectively).

STR/HLA haplotype correlates (Table II) will facilitate fine mapping of the disease loci in the MHC, the prediction of HLA specificities using low-cost STR haplotyping, and the identification of HLA recombinants.

Differential value of susceptibility HLA haplotypes for the fine-scale mapping

The alignment of susceptibility STR haplotypes in affected family members (Fig. 3 and supplementary online information) showed that the majority of the DQB1*02-carrying haplotypes, including the most prevalent Caucasian AH8.1 (Fig. 3A), extended from the G51152/9-99431 interval into the class I region. In contrast, most of the DQB1*05-positive haplotypes, except for those carrying the DRB1*01021 allele (Fig. 3B), were broken down in the telomeric part of the class II region and had variant haplotypes in the class III (Fig. 3, D–F) and class I (data not shown) regions. Among the affected 10-3-6-8 homozygotes, only 9 of 28 haplotypes shared telomeric class III alleles and only 5 of them class I alleles (Fig. 3C). To map the telomeric border of this predisposing haplotype, we analyzed cSNP haplotypes in BTNL2 (BTLII, Fig. 1) and found that the 10-3-6-8-carrying haplotypes diversified in a region of ~40 kb between exon 5 of BTNL2 and 9-99431 located just telomeric of HLA-DRA (Fig. 1, 3d and data not shown). Thus, the haplotype analysis clearly showed that only a subset of predisposing haplotypes permitted to narrow down IGAD1 to a smaller region and was thus informative for the fine-scale mapping (Fig. 3).

HLA-DQ/DR is the strongest predisposing locus

Although the quasi-random nature of peptide MHC class II interactions with TCR (pMHCII/TCR) may fully account for a complex inheritance of IgAD/CVID and other autoimmune disorders in families, non-MHC genes involved in cellular processes triggered by such interactions may modify the disease risk. To test this possibility and to see if the MHC is the strongest locus, we conducted an STR-based genetic linkage study with 101 multiplex families. The genome-wide scan of IgAD/CVID showed that the highest multipoint nonparametric linkage scores at 6p were not matched anywhere else in the genome (Fig. 4, Table III), implicating IGAD1 as the major disease locus. In addition to IGAD1, we found a suggestive linkage (NPL > 1.8) to non-MHC regions at 4p, 12p, 14q, 16q, 7p, and 1q (Fig. 4), and we tested these loci with a denser set of STRs and additional 109 single-case families and family-based allelic associations (Table III). Although the suggestive linkage was not supported with additional STRs at 1q and 16q (data not shown), we observed an increased sharing for adjacent STRs at 4p (NPLall = 2.45, information content 0.74, p = 0.002), 12p (NPLall = 2.23, information content 0.87, p = 0.005), and 14q (NPLall = 1.97, information content 0.81, p = 0.01; Fig. 4 and Table III). Although the p values were below the significance limit recommended for genome-wide studies, these loci were further supported by positive TDTs for STRs at 4p and 12p (Table III). Suggestive linkage at 14q and on chromosome 7 (Fig. 4), which may implicate allelic variants of the TCR genes in predisposition, is subject to further analysis in our laboratory (I. Vorechovsky, L. Hammarström, and A. D. B. Webster, manuscript in preparation).



View larger version (54K):
[in this window]
[in a new window]
  		FIGURE 4. Genetic linkage analysis of IgAD/CVID with 371 autosomal and 12 X-linked chromosomal markers in 101 pedigrees. Black and blue lines show multipoint NPL scores for all and pairs options of the Genehunter program, respectively (scale on the left). Gray lines show the information content (scale on the right). Definitive and putative susceptibility loci are denoted with filled and open bars, respectively.

 

View this table:
[in this window]
[in a new window]
  		Table III. IGAD1 confers the strongest risk of IgAD/CVID

 
A search for locus heterogeneity of CVID and IgAD

In HLA, we found no clear evidence for locus heterogeneity for either IgAD or CVID. Even STR haplotypes characteristic of AH8.1 were overrepresented at HLA-DQ/DR as compared with region 2 in affected family members, and we observed more family members homozygous for AH8.1 at HLA-DQ/DR than in region 2.4 In addition, CVID patients were represented among carriers of all susceptibility haplotypes (Fig. 3), supporting earlier proposals of independent investigators (5, 6, 13, 14) that the two PIDs represent a spectrum in disease severity.

Although previously reported allelic associations at IGAD1 were weaker for CVID than IgAD (7), this may have been due to a smaller number of analyzed CVID patients, occasional misdiagnosis of CVID for other PIDs (25), and locus heterogeneity, perhaps with a greater contribution from non-MHC genes in CVID than IgAD. Because rare cases of CVID-like phenotypes have been shown to result from Mendelian loss-of-function mutations (25, 41), we addressed the latter possibility by analyzing candidate genes previously implicated humoral defects for mutations (Table IV). This was conducted also for genes located in chromosome regions not exhibiting an increased allele sharing (Fig. 4), because rare sporadic cases or families would not be expected to contribute to detectable association or linkage. The mutation screening of the gene-coding regions did not identify any disease-specific changes for XBP-1, ICOS1 (AILIM), OBF-1, CD38, and CTSS (Table IV), lending further support to the notion that most phenotypic variability observed in IgAD/CVID is, as for many autoimmune conditions, controlled by the interactions of class II molecules with antigenic peptides and Ag receptors on thymocytes and CD4+ T cells.


View this table:
[in this window]
[in a new window]
  		Table IV. Mutation analysis of candidate genes for CVID

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our results indicate that the HLA-DQ/DR locus is the major hereditary determinant of susceptibility to IgAD/CVID. The evidence is provided collectively by 1) the highest NPL/Zlr scores at 6p21 in the genome-wide scan (Fig. 4); 2) the strongest allelic associations in the class II/III regions in family-based (Fig. 1) (21) and case-control studies (for review see Ref.2); 3) haplotype analysis combined with the high-resolution HLA typing (Fig. 3 and Table II); 4) a higher risk of developing IgAD/CVID in HLA class II homozygotes and a significant bias in the distribution of homozygous stretches toward the class II region (Fig. 2B); and 5) the absence of other expressed genes than HLA-DQ/DR in the interval G51152/9-99431 implicated by all fine-mapping methods (Figs. 1–3, Table I).

A number of functional lymphocyte abnormalities in vitro and substantial clinical heterogeneity suggested that CVID is a collection of Mendelian defects affecting the Ab production (42). However, neither linkage nor allelic association would have been consistently detected at IGAD1 in our patients and families (Fig. 1, Tables III–IV) and in previous case-control studies with a more limited sample size (2), had there been an extensive genetic heterogeneity of IgAD/CVID. Very few patients diagnosed as CVID have been shown to have a Mendelian defect (25), consistent with our negative mutation screening (Table IV). Furthermore, as in other autoimmune disorders, the markers at 6p showed the highest excess of allelic sharing in the genome-wide scan, indicating that HLA-DQ/DR is the strongest predisposing locus (Fig. 4), with the 5-1-7-6 and 10-3-6-8 haplotypes contributing most to the disease predisposition in the analyzed population. This is consistent with the previous observations that the frequency of susceptibility haplotype(s) in several populations is mirrored by a gradient in the population prevalence of IgAD (reviewed in Ref.43).

In HLA, our results point to HLA-DR/DQ and not to the class III region, although not excluding a contribution of telomeric gene(s) as proposed for AH8.1 (44) and as hinted by an occasional class III homozygosity in affected individuals heterozygous for the class II alleles and by a second peak of allelic association for STRs around HLA-B (Fig. 1). However, the former observation can be due to chance and the latter finding may be fully explained by a high LD of the overrepresented AH8.1. Also, allelic associations with the disease were consistently lower for the HLA class I STRs than in the HLA-DQB1/DRA region, as was reported in previous case-control studies of the HLA class I Ags (reviewed in Ref.2). A high LD of AH8.1 renders this haplotype much less informative for the fine-mapping studies (Fig. 3A), and differential LD of HLA haplotypes well-illustrated by our results should favor gene-mapping strategies that consider all founder haplotypes, such as the trimmed haplotype method combined with homozygosity mapping (Table I, Fig. 3). The high LD of AH8.1 may have contributed to provisional placements of the susceptibility locus to more telomeric regions (6, 16, 17, 18).

In addition to the strong genetic evidence for the major role of the HLA-DQ/DR locus, both laboratory and clinical features of CVID provide indirect support for these molecules. Because functional correlates of antigenic pMHCII/TCR interactions are distinct in mature T cells (activation, anergy) and developing thymocytes (positive and negative selection), one would expect phenotypic features reflecting both developmental stages. Indeed, a number of functional abnormalities has been observed in CVID, including restricted TCR repertoires (45, 46), increased apoptosis (47, 48, 49, 50, 51), enhanced sensitivity of T cells to corticosteroids (47), an impairment of early signaling events triggered by TCR such as calcium flux, phospholipid metabolism and kinase activity (52, 53, 54), peripheral anergy, and dysregulation of the cytokine network including Th1 skewing (55, 56). However, the nature of pMHC/TCR interactions, particularly the longevity of pMHC/TCR complexes and kinetic thresholds between negative and positive selection, may fully account for many, if not all, abnormalities, including the modulation of Th1/Th2 differentiation (57, 58, 59, 60). Altered signaling via pMHC/TCR can result in differential phosphorylation patterns in downstream kinases, such as ZAP-70, which shows defective activation and recruitment in CVID (52, 53, 54). Ab deficiency in IgAD/CVID may develop as a result of impaired T cell activation and insufficient T cell help to B cells, consistent with IgA being the most T cell-dependent isotype and with the impairment of T cell-dependent processes in germinal centers observed in some CVID patients, such as affinity maturation (62) and memory formation (63). Although functional defects may be secondary to thymic pMHCII/TCR interactions, transgenic models of self-tolerance point to complex mechanisms involving peripheral and central selection coupled with a dysregulation of many immune effector genes, rather than processes mediated purely by thymic deletion (64). It will be important to identify autoantigenic epitopes in future studies, but it will be difficult to dissect the network of processes pertinent to autoimmunity, hypogammaglobulinemia, infection, and inflammation in symptomatic patients.

Clinical features of IgAD/CVID also resemble those found in other autoimmune disorders, including 1) the occurrence of anti-IgA Ab that are present at a higher frequency in patients with autoimmune complications than in asymptomatic IgAD (65); 2) the increased prevalence of autoimmune disease in IgAD/CVID (65); 3) frequent disease manifestation after an episode of other autoimmune phenomena; 4) variable disease severity; and 5) a slow, gradual appearance of hypogammaglobulinemia over many years (12). A transmission of IgAD to a healthy sibling by bone marrow transplantation (66) may have been due to the transfer of autoreactive T cell clones. IgAD can be transmitted from mother to fetus (7, 67), as has been observed for a number of autoimmune disorders. An autoreactive pathogenesis can also explain occasional recovery of CVID, including HIV-induced remission (68), that may have been due to the down-regulation of proinflammatory cytokines, reduction of the surface expression of HLA-DQ/DR (68), and/or altered {beta}-chain usage of the variable portion of TCR by deleting specific V{beta} families (69). The central role of HLA-DQ/DR in IgAD/CVID predisposition is also indirectly supported by the existence of humoral defects induced by drugs that interfere with Ag presentation as proposed earlier (7) and by hypogammaglobulinemia induced by the targeted inactivation of several components of Ag presentation pathways in the mouse, such as cathepsin S (70), indicating the critical dependence of humoral immunity on how Ags access the immune system. A diminished number of B cells observed in a subset of CVID may be due to the autodestruction of B cells expressing IgA/IgG, but it might also develop as a result of an impaired balance between class II chain pairing (71).

Given the central predisposing role of HLA-DQ/DR, susceptibility to IgAD/CVID may be influenced by genes modifying pMHCII/TCR interactions and shaping the T cell repertoire, including molecules involved in amplifying or inhibiting TCR signals, apoptosis, adhesion molecules, and coreceptors. We previously tested the aberrant splicing of PTPRC, a large phosphatase involved in downstream TCR signaling, but no apparent association was found in CVID (72). Although variable humoral defects have been found for targeted genes involved in DNA repair (73), candidates supported by occasional chromosomal radiosensitivity in CVID (74), this phenomenon is perhaps more likely to reflect alterations of T cell subpopulations exhibiting differential radiosensitivity, such as mature and immature T cells (75). The presence of hypogammaglobulinemia in a subset of patients with the Wolf-Hirschhorn syndrome (76), a contiguous gene syndrome mapped to 4p16, suggests that a modifying gene and a gene closely linked to Wolf-Hirschhorn syndrome may be identical. In fact, several genes in the region have been implicated in humoral immune responses, including CD38 and CD157 (Table IV). A putative locus at 12p, which shows association/suggestive linkage in other autoimmune disorders (77), contains the NK gene complex, and thus is of interest since NK cell abnormalities have been reported in CVID (78). A proposal linking IL-10 to the granulomatous form of CVID (79) may gain support from the NPL scores at 1q31-q32 (Fig. 4), but previous allelic associations at IGHG1 (chromosome region 14q32) (80, 81) and MBL1 (10q11-q21) (82) were not supported by allele-sharing data (Fig. 4). Clearly, large and independent association studies will be required to confirm all putative non-MHC loci in different populations, preferably supported by collaborative linkage studies.

Finally, the mounting evidence for autoimmune pathogenesis of IgAD/CVID is likely to influence current therapeutic strategies, shifting the attention from replacement Ig therapy to more active immunomodulating and, paradoxically, immunosuppressive approaches in symptomatic CVID. Because of the potential availability of both target and effector lymphocyte populations, CVID may become a useful model for gene therapy of autoimmune disease.


   Acknowledgments
 
We thank Michael D. Laycock and Shuang Liu for technical help, Renee Engqvist and Anne Bryan for their enthusiastic work with family members, and referring physicians for their support of the European Society for Immunodeficiencies Registry and this project.


   Footnotes
 
1 This work was supported by grants from the Swedish Foundation for Strategic Research (to I.V.), the British and Swedish Medical Research Councils (to L.H. and A.D.B.W.), and the Primary Immunodeficiency Association of the United Kingdom (to I.V., L.H., and A.D.B.W.). Back

2 Address correspondence and reprint requests to Dr. Igor Vorechovsky, Division of Human Genetics, Southampton University Hospital, School of Medicine, University of Southampton, Tremona Road, Southampton SO16 6YD, U.K. E-mail address: igvo{at}soton.ac.uk Back

3 Abbreviations used in this paper: PID, primary immunodeficiency; IgAD, selective IgA deficiency; CVID, common variable immunodeficiency; LD, linkage disequilibrium; STR, short tandem repeat; TDT, transmission disequilibrium test; FBAT, family-based allelic association test; SNP, single nucleotide polymorphism; HWE, Hardy-Weinberg equilibrium. Back

4 The online version of this article contains supplemental material. Back

Received for publication September 18, 2002. Accepted for publication December 18, 2002.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ochs, H. D., C. I. E. Smith, J. M. Puck. 1999. Primary Immunodeficiency Diseases Oxford University Press, Oxford.
  2. Burrows, P. D., M. D. Cooper. 1997. IgA deficiency. Adv. Immunol. 65:245.[Medline]
  3. Cunningham-Rundles, C., C. Bodian. 1999. Common variable immunodeficiency: clinical and immunological features of 248 patients. J. Clin. Immunol. 92:34.
  4. Koistinen, J.. 1976. Familial clustering of IgA deficiency. Vox Sang. 30:181.[Medline]
  5. Vorechovsky, I., H. Zetterquist, R. Paganelli, S. Koskinen, A. D. Webster, J. Bjorkander, C. I. Smith, L. Hammarström. 1995. Family and linkage study of selective IgA deficiency and common variable immunodeficiency. Clin. Immunol. Immunopathol. 77:185.[Medline]
  6. Volanakis, J. E., Z.-B. Zhu, F. M. Schaffer, K. J. Macon, J. Palermos, B. O. Berger, R. Go, R. D. Campbell, H. W. J. Schroeder, M. D. Cooper. 1992. Major histocompatibility complex class III genes and susceptibility to immunoglobulin A deficiency and common variable immunodeficiency. J. Clin. Invest. 89:1914.
  7. Vorechovsky, I., A. D. B. Webster, A. Plebani, L. Hammarström. 1999. Genetic linkage of IgA deficiency to the major histocompatibility complex: evidence for allele segregation distortion, parent-of-origin penetrance differences and the role of anti-IgA antibodies in disease predisposition. Am. J. Hum. Genet. 64:1096.[Medline]
  8. Ishizaka, A., M. Nakanishi, S. Yamada, Y. Sakiyama, S. Matsumoto. 1989. Development of hypogammaglobulinaemia in a patient with common variable immunodeficiency. Eur. J. Pediatr. 149:175.[Medline]
  9. Sanal, O., F. Ersoy, A. Metin, I. Tezcan, A. I. Berkel, L. Yel. 1995. Selective IgA deficiency with unusual features: development of common variable immunodeficiency, Sjogren’s syndrome, autoimmune hemolytic anemia and immune thrombocytopenic purpura. Acta Paediatr. Jpn. 37:526.[Medline]
  10. Español, T., M. Catala, M. Hernandez, I. Caragol, J. M. Bertran. 1996. Development of a common variable immunodeficiency in IgA-deficient patients. Clin. Immunol. Immunopathol. 80:333.[Medline]
  11. Slyper, A. H., D. Pietryga. 1997. Conversion of selective IgA deficiency to common variable immunodeficiency in an adolescent female with 18q deletion syndrome. Eur. J. Pediatr. 156:155.
  12. Johnson, M. L., L. G. Keeton, Z.-B. Zhu, J. E. Volanakis, M. D. Cooper, H. W. Schroeder, Jr.. 1997. Age-related changes in serum immunoglobulins in patients with familial IgA deficiency and common variable immunodeficiency (CVID). Clin. Exp. Immunol. 108:477.[Medline]
  13. Schaffer, F. M., J. Palermos, Z. B. Zhu, B. O. Barger, M. D. Cooper, J. E. Volanakis. 1989. Individuals with IgA deficiency and common variable immunodeficiency share polymorphisms of major histocompatibility complex class III genes. Proc. Natl. Acad. Sci. USA 86:8015.[Abstract/Free Full Text]
  14. Olerup, O., C. I. Smith, J. Björkander, L. Hammarström. 1992. Shared HLA class II-associated genetic susceptibility and resistance, related to the HLA-DQB1 gene, in IgA deficiency and common variable immunodeficiency. Proc. Natl. Acad. Sci. USA 89:10653.[Abstract/Free Full Text]
  15. Eibl, M. M., H. M. Wolf. 1995. Common variable immunodeficiency: clinical aspects and recent progress in identifying the immunological defect(s). Folia Microbiol. 40:360.
  16. French, M. A., R. L. Dawkins. 1990. Central MHC genes, IgA deficiency and autoimmune disease. Immunol. Today 11:271.[Medline]
  17. Cucca, F., Z. B. Zhu, A. Khanna, F. Cossu, M. Congia, M. Badiali, R. Lampis, F. Frau, S. De Virgiliis, A. Cao, et al 1998. Evaluation of IgA deficiency in Sardinians indicates a susceptibility gene is encoded within the HLA class III region. Clin. Exp. Immunol. 111:76.[Medline]
  18. Schroeder, H. W., Jr., Z. B. Zhu, R. E. March, R. D. Campbell, S. M. Berney, S. A. Nedospasov, R. L. Turetskaya, T. P. Atkinson, R. C. Go, M. D. Cooper, J. E. Volanakis. 1998. Susceptibility locus for IgA deficiency and common variable immunodeficiency in the HLA-DR3, -B8, -A1 haplotypes. Mol. Med. 4:72.[Medline]
  19. Matthews, V. B., C. S. Witt, M. A. French, H. K. Machulla, E. G. De la Concha, K. Y. Cheong, P. Vigil, P. N. Hollingsworth, K. J. Warr, F. T. Christiansen, P. Price. 2002. Central MHC genes affect IgA levels in the human: reciprocal effects in IgA deficiency and IgA nephropathy. Hum. Immunol. 63:424.[Medline]
  20. Olerup, O., C. I. Smith, L. Hammarstrom. 1991. Is selective IgA deficiency associated with central HLA genes or alleles of the DR-DQ region?. Immunol. Today 12:134.[Medline]
  21. Vorechovsky, I., M. Cullen, M. Carrington, L. Hammarstrom, A. D. Webster. 2000. Fine mapping of IGAD1 in IgA deficiency and common variable immunodeficiency: identification and characterization of haplotypes shared by affected members of 101 multiple-case families. J. Immunol. 164:4408.[Abstract/Free Full Text]
  22. French, M., R. Dawkins, F. T. Christiansen, W. Zhang, M. A. Degli-Esposti, G. Saueracker. 1991. Is selective IgA deficiency associated with central HLA genes or alleles of the DR-DQ region?. Immunol. Today 12:135.
  23. Group, R. o. a. W. S. 1995. Primary immunodeficiency diseases. Clin. Exp. Immunol. 1:(Suppl 99):1.
  24. Plebani, A., A. O. Carbonara, A. Bottaro, R. Gallina, C. Boccazzi, P. Crispino, L. Ruggeri, F. Salvioni, M. Duina, A. Negrini, et al 1993. Gene deletion as a cause of associated deficiency of IgA1, IgG2, IgG4 and IgE. Immunodeficiency 4:245.[Medline]
  25. Vorechovsky, I., J. N. Zhou, D. Vetrie, D. Bentley, J. Bjorkander, L. Hammarstrom, C. I. Smith. 1993. Molecular diagnosis of X-linked agammaglobulinaemia. Lancet 341:1153.
  26. Vorechovsky, I., L. Luo, M. J. Dyer, D. Catovsky, P. L. Amlot, J. C. Yaxley, L. Foroni, L. Hammarstrom, A. D. Webster, M. A. Yuille. 1997. Clustering of missense mutations in the ataxia-telangiectasia gene in a sporadic T-cell leukaemia. Nat. Genet. 17:96.[Medline]
  27. Tamiya, G., T. Shiina, A. Oka, M. Tomizawa, M. Ota, Y. Katsuyama, M. Yoshitome, S. Makino, M. Kimura, H. Inoko. 1999. New polymorphic microsatellite markers in the human MHC class I region. Tissue Antigens 54:221.[Medline]
  28. Vorechovsky, I., J. Kralovicova, M. D. Laycock, A. D. B. Webster, S. G. E. Marsh, A. Madrigal, L. Hammarstrom. 2001. Short tandem repeat (STR) haplotypes in HLA: an integrated 50-kb STR/linkage disequilibrium/gene map between the RING3 and HLA-B genes and identification of STR haplotype diversification in the class III region. Eur. J. Hum. Genet. 9:590.[Medline]
  29. Stammers, M., L. Rowen, D. Rhodes, J. Trowsdale, S. Beck. 2000. BTL-II: a polymorphic locus with homology to the butyrophilin gene family, located at the border of the major histocompatibility complex class II and class III regions in human and mouse. Immunogenetics 51:373.[Medline]
  30. Kruglyak, L., M. J. Daly, M. P. Reeve-Daly, E. S. Lander. 1996. Parametric and nonparametric linkage analysis: a unified multipoint approach. Am. J. Hum. Genet. 58:1347.[Medline]
  31. Gudbjartsson, D. F., K. Jonasson, M. L. Frigge, A. Kong. 2000. Allegro, a new computer program for multipoint linkage analysis. Nat. Genet. 25:12.[Medline]
  32. Laird, N. M., S. Horvath, X. Xu. 2000. Implementing a unified approach to family based tests of association. Genet. Epidemiol. 1:S36.
  33. MacLean, C. J., R. B. Martin, P. C. Sham, H. Wang, R. E. Straub, K. S. Kendler. 2000. The trimmed-haplotype test for linkage disequilibrium. Am. J. Hum. Genet. 66:1062.[Medline]
  34. Miettinen, O. S.. 1974. Proportion of disease caused or prevented by a given exposure, trait or intervention. Am. J. Epidemiol. 99:325.[Abstract/Free Full Text]
  35. Feder, J. N., A. Gnirke, W. Thomas, Z. Tsuchihashi, D. A. Ruddy, A. Basava, F. Dormishian, R. Domingo, Jr., M. C. Ellis, A. Fullan, et al 1996. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat. Genet. 13:399.[Medline]
  36. Herr, M., F. Dudbridge, P. Zavattari, F. Cucca, C. Guja, R. March, R. D. Campbell, A. H. Barnett, S. C. Bain, J. A. Todd, B. P. Koeleman. 2000. Evaluation of fine mapping strategies for a multifactorial disease locus: systematic linkage and association analysis of IDDM1 in the HLA region on chromosome 6p21. Hum. Mol. Genet. 9:1291.[Abstract/Free Full Text]
  37. Sham, P. C., D. Curtis. 1995. An extended transmission/disequilibrium test (TDT) for multi-allele marker loci. Ann. Hum. Genet. 59:323.[Medline]
  38. De La Concha, E. G., M. Fernandez-Arquero, A. Martinez, F. Vidal, P. Vigil, L. Conejero, M. C. Garcia-Rodriguez, G. Fontan. 1999. HLA class II homozygosity confers susceptibility to common variable immunodeficiency (CVID). Clin. Exp. Immunol. 116:516.[Medline]
  39. Machulla, H. K., U. Schonermarck, A. Schaaf, L. P. Muller, C. Kloss, J. Kruger, G. Kunze, G. Schonermarck, J. Langner. 2000. HLA-A, B, Cw and DRB1, DRB3/4/5, DQB1, DPB1 frequencies in German immunoglobulin A-deficient individuals. Scand. J. Immunol. 52:207.[Medline]
  40. Nielsen, D. M., M. G. Ehm, B. S. Weir. 1998. Detecting marker-disease association by testing for Hardy-Weinberg disequilibrium at a marker locus. Am. J. Hum. Genet. 63:1531.[Medline]
  41. Stewart, D. M., L. Tian, D. L. Nelson. 2001. A case of X-linked agammaglobulinemia diagnosed in adulthood. Clin. Immunol. 99:94.[Medline]
  42. Spickett, G. P., J. Farrant, M. E. North, J. G. Zhang, L. Morgan, A. D. Webster. 1997. Common variable immunodeficiency: how many diseases?. Immunol. Today 18:325.[Medline]
  43. Vorechovsky, I., A. D. Webster, L. Hammarstrom. 2001. Mapping genes underlying complex disorders: progress on IgA deficiency and common variable immunodeficiency. Adv. Exp. Med. Biol. 495:183.[Medline]
  44. De La Concha, E. G., M. Fernandez-Arquero, L. Gual, P. Vigil, A. Martinez, E. Urcelay, A. Ferreira, M. C. Garcia-Rodriguez, G. Fontan. 2002. MHC susceptibility genes to IgA deficiency are located in different regions on different HLA haplotypes. J. Immunol. 169:4637.[Abstract/Free Full Text]
  45. Duchmann, R., J. Jaffe, R. Ehrhardt, D. W. Alling, W. Strober. 1996. Differential usage of T-cell receptor V{beta} gene families by CD4+ and CD8+ T cells in patients with CD8hi common variable immunodeficiency: evidence of a post-thymic effect. Immunology 87:99.[Medline]
  46. Serrano, D., K. Becker, C. Cunningham-Rundles, L. Mayer. 2000. Characterization of the T cell receptor repertoire in patients with common variable immunodeficiency: oligoclonal expansion of CD8+ T cells. Clin. Immunol. 97:248.[Medline]
  47. Vukmanovic, S., S. Vuckovic, S. Stosic-Grujicic, Z. Ramic, M. Abinun. 1992. An unusual T-cell surface phenotype in vivo correlates with the failure to proliferate and produce IL-2 in vitro in a patient with common variable immunodeficiency. Clin. Immunol. Immunopathol. 65:261.[Medline]
  48. Iglesias, J., N. Matamoros, S. Raga, J. M. Ferrer, J. Mila. 1999. CD95 expression and function on lymphocyte subpopulations in common variable immunodeficiency (CVID); related to increased apoptosis. Clin. Exp. Immunol. 117:138.[Medline]
  49. Di Renzo, M., Z. Zhou, I. George, K. Becker, C. Cunningham-Rundles. 2000. Enhanced apoptosis of T cells in common variable immunodeficiency (CVID): role of defective CD28 co-stimulation. Clin. Exp. Immunol. 120:503.[Medline]
  50. Washington, K., T. T. Stenzel, R. H. Buckley, M. R. Gottfried. 1996. Gastrointestinal pathology in patients with common variable immunodeficiency and X-linked agammaglobulinemia. Am. J. Surg. Pathol. 20:1240.[Medline]
  51. Saxon, A., B. Keld, D. Diaz-Sanchez, B. C. Guo, N. Sidell. 1995. B cells from a distinct subset of patients with common variable immunodeficiency (CVID) have increased CD95 (Apo-1/fas), diminished CD38 expression, and undergo enhanced apoptosis. Clin. Exp. Immunol. 102:17.[Medline]
  52. Fischer, M. B., I. Hauber, H. Eggenbauer, V. Thon, E. Vogel, E. Schaffer, J. Lokaj, J. Litzman, H. M. Wolf, J. W. Mannhalter, et al 1994. A defect in the early phase of T-cell receptor-mediated T-cell activation in patients with common variable immunodeficiency. Blood 84:4234.[Abstract/Free Full Text]
  53. Majolini, M. B., M. M. D’Elios, M. Boncristiano, P. Galieni, G. Del Prete, J. L. Telford, C. T. Baldari. 1997. Uncoupling of T-cell antigen receptor and downstream protein tyrosine kinases in common variable immunodeficiency. Clin. Immunol. Immunopathol. 84:98.[Medline]
  54. Boncristiano, M., M. B. Majolini, M. M. D’Elios, S. Pacini, S. Valensin, C. Ulivieri, A. Amedei, B. Falini, G. Del Prete, J. L. Telford, C. T. Baldari. 2000. Defective recruitment and activation of ZAP-70 in common variable immunodeficiency patients with T cell defects. Eur. J. Immunol. 30:2632.[Medline]
  55. Sneller, M. C., W. Strober. 1990. Abnormalities of lymphokine gene expression in patients with common variable immunodeficiency. J. Immunol. 144:3762.[Abstract]
  56. Cambronero, R., W. A. Sewell, M. E. North, A. D. Webster, J. Farrant. 2000. Up-regulation of IL-12 in monocytes: a fundamental defect in common variable immunodeficiency. J. Immunol. 164:488.[Abstract/Free Full Text]
  57. Constant, S. L., K. Bottomly. 1997. Induction of Th1 and Th2 CD4+ T cell responses: the alternative approaches. Annu. Rev. Immunol. 15:297.[Medline]
  58. Tao, X., S. Constant, P. Jorritsma, K. Bottomly. 1997. Strength of TCR signal determines the costimulatory requirements for Th1 and Th2 CD4+ T cell differentiation. J. Immunol. 159:5956.[Abstract]
  59. Iezzi, G., E. Scotet, D. Scheidegger, A. Lanzavecchia. 1999. The interplay between the duration of TCR and cytokine signaling determines T cell polarization. Eur. J. Immunol. 29:4092.[Medline]
  60. Williams, C. B., D. L. Engle, G. J. Kersh, J. M. White, P. M. Allen. 1999. A kinetic threshold between negative and positive selection based on the longevity of the T cell receptor-ligand complex. J. Exp. Med. 189:1531.[Abstract/Free Full Text]
  61. Islam, K. B., B. Baskin, L. Nilsson, L. Hammarstrom, P. Sideras, C. I. Smith. 1994. Molecular analysis of IgA deficiency: evidence for impaired switching to IgA. J. Immunol. 152:1442.[Abstract]
  62. Levy, Y., N. Gupta, F. Le Deist, C. Garcia, A. Fischer, J. C. Weill, C. A. Reynaud. 1998. Defect in IgV gene somatic hypermutation in common variable immuno-deficiency syndrome. Proc. Natl. Acad. Sci. USA 95:13135.[Abstract/Free Full Text]
  63. Kondratenko, I., P. L. Amlot, A. D. Webster, J. Farrant. 1997. Lack of specific antibody response in common variable immunodeficiency (CVID) associated with failure in production of antigen-specific memory T cells. Clin. Exp. Immunol. 108:9.[Medline]
  64. Boyton, R. J., D. M. Altmann. 2002. Transgenic models of autoimmune disease. Clin. Exp. Immunol. 127:4.[Medline]
  65. Petty, R. E., N. R. Palmer, J. T. Cassidy, D. G. Tubergen, D. B. Sullivan. 1979. The association of autoimmune diseases and anti-IgA antibodies in patients with selective IgA deficiency. Clin. Exp. Immunol. 37:83.[Medline]
  66. Hammarstrom, L., B. Lonnqvist, O. Ringden, C. I. Smith, T. Wiebe. 1985. Transfer of IgA deficiency to a bone-marrow-grafted patient with aplastic anaemia. Lancet 1:778.[Medline]
  67. Oen, K., R. E. Petty, M. L. Schroeder. 1982. Immunoglobulin A deficiency: genetic studies. Tissue Antigens 19:174.[Medline]
  68. Wright, J. J., D. L. Birx, D. K. Wagner, T. A. Waldmann, R. M. Blaese, T. A. Fleisher. 1987. Normalization of antibody responsiveness in a patient with common variable hypogammaglobulinemia and HIV infection. N. Engl. J. Med. 317:1516.[Medline]
  69. Imberti, L., A. Sottini, A. Bettinardi, M. Puoti, D. Primi. 1991. Selective depletion in HIV infection of T cells that bear specific T cell receptor V{beta} sequences. Science 254:860.[Abstract/Free Full Text]
  70. Shi, G. P., J. A. Villadangos, G. Dranoff, C. Small, L. Gu, K. J. Haley, R. Riese, H. L. Ploegh, H. A. Chapman. 1999. Cathepsin S required for normal MHC class II peptide loading and germinal center development. Immunity 10:197.[Medline]
  71. Labrecque, N., L. Madsen, L. Fugger, C. Benoist, D. Mathis. 1999. Toxic MHC class II {beta} chains. Immunity 11:515.[Medline]
  72. Vorechovsky, I., J. Kralovicova, E. Tchilian, T. Masterman, Z. Zhang, B. Ferry, S. Misbah, H. Chapel, D. Webster, D. Hellgren, et al 2001. Does 77C->G in PTPRC modify susceptibility to autoimmune disorders?. Nat. Genet. 29:22.[Medline]
  73. Schrader, C. E., W. Edelmann, R. Kucherlapati, J. Stavnezer. 1999. Reduced isotype switching in splenic B cells from mice deficient in mismatch repair enzymes. J. Exp. Med. 190:323.[Abstract/Free Full Text]
  74. Vorechovsky, I., D. Scott, M. R. Haeney, D. A. Webster. 1993. Chromosomal radiosensitivity in common variable immune deficiency. Mutat. Res. 290:255.[Medline]
  75. Bhandoola, A., B. Dolnick, N. Fayad, A. Nussenzweig, A. Singer. 2000. Immature thymocytes undergoing receptor rearrangements are resistant to an Atm-dependent death pathway activated in mature T cells by double-stranded DNA breaks. J. Exp. Med. 192:891.[Abstract/Free Full Text]
  76. Hanley-Lopez, J., L. L. Estabrooks, R. Stiehm. 1998. Antibody deficiency in Wolf-Hirschhorn syndrome. J. Pediatr. 133:141.[Medline]
  77. Wanstrat, A., E. Wakeland. 2001. The genetics of complex autoimmune diseases: non-MHC susceptibility genes. Nat. Immunol. 2:802.[Medline]
  78. Aspalter, R. M., W. A. Sewell, K. Dolman, J. Farrant, A. D. Webster. 2000. Deficiency in circulating natural killer (NK) cell subsets in common variable immunodeficiency and X-linked agammaglobulinaemia. Clin. Exp. Immunol. 121:506.[Medline]
  79. Mullighan, C. G., S. E. Marshall, M. Bunce, K. I. Welsh. 1999. Variation in immunoregulatory genes determines the clinical phenotype of common variable immunodeficiency. Genes Immun. 1:137.[Medline]
  80. Oxelius, V. A.. 1996. Suppressed G2m(n) levels from IGHCG2 in IgA deficiency. Exp. Clin. Immunogenet. 13:1.[Medline]
  81. Olsson, P. G., L. Hammarstrom, D. W. Cox, C. I. Smith. 1992. Involvement of both HLA and Ig heavy chain haplotypes in human IgA deficiency. Immunogenetics 36:389.[Medline]
  82. Mullighan, C. G., S. E. Marshall, K. I. Welsh. 2000. Mannose binding lectin polymorphisms are associated with early age of disease onset and autoimmunity in common variable immunodeficiency. Scand. J. Immunol. 51:111.[Medline]
  83. Nielsen, P. J., O. Georgiev, N. Lorenz, W. Schaffner. 1996. B lymphocytes are impaired in mice lacking the transcriptional co-activator Bob1/OCA-B/OBF1. Eur. J. Immunol. 26:3214.[Medline]
  84. Cockayne, D. A., T. Muchamuel, J. C. Grimaldi, H. Muller-Steffner, T. D. Randall, F. E. Lund, R. Murray, F. Schuber, M. C. Howard. 1998. Mice deficient for the ecto-nicotinamide adenine dinucleotide glycohydrolase CD38 exhibit altered humoral immune responses. Blood 92:1324.[Abstract/Free Full Text]
  85. Itoh, M., K. Ishihara, T. Hiroi, B. O. Lee, H. Maeda, H. Iijima, M. Yanagita, H. Kiyono, T. Hirano. 1998. Deletion of bone marrow stromal cell antigen-1 (CD157) gene impaired systemic thymus independent-2 antigen-induced IgG3 and mucosal TD antigen-elicited IgA responses. J. Immunol. 161:3974.[Abstract/Free Full Text]
  86. Dong, C., A. E. Juedes, U. A. Temann, S. Shresta, J. P. Allison, N. H. Ruddle, R. A. Flavell. 2001. ICOS co-stimulatory receptor is essential for T-cell activation and function. Nature 409:97.[Medline]
  87. Reimold, A. M., N. N. Iwakoshi, J. Manis, P. Vallabhajosyula, E. Szomolanyi-Tsuda, E. M. Gravallese, D. Friend, M. J. Grusby, F. Alt, L. H. Glimcher. 2001. Plasma cell differentiation requires the transcription factor XBP-1. Nature 412:300.[Medline]



This article has been cited by other articles:


Home page
 EndocrinologyHome page
H. Kimura, M. Kimura, S.-C. Tzou, Y.-C. Chen, K. Suzuki, N. R. Rose, and P. Caturegli
Expression of Class II Major Histocompatibility Complex Molecules on Thyrocytes Does Not Cause Spontaneous Thyroiditis but Mildly Increases Its Severity after Immunization
Endocrinology, March 1, 2005; 146(3): 1154 - 1162.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kralovicova, J.
Right arrow Articles by Vorechovsky, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kralovicova, J.
Right arrow Articles by Vorechovsky, I.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS