Alopecia areata: the disease

It is estimated that AA affects between 2% and 4% of patients within average dermatology practices (^{Price, 1991}). It usually begins as one or several oval patches of nonscarring hair loss in the scalp that can appear suddenly or more gradually over several days or weeks. The hair loss may regress, or the patches can coalesce and progress to cover the entire scalp (AT) and eventually the entire body (AU). AA is sometimes accompanied by nail changes in the form of pitting, brittleness, and splitting. In all, the prognosis of AA is unpredictable and there is no definitive treatment.

The pathogenetic basis and etiology of AA are largely unknown. In the initial stages, the number of hair follicles appears to remain the same; however, in the more advanced stages this number decreases and miniaturization of the anagen hair follicles is observed (^{McDonagh and Messenger, 1996}). A cardinal feature of AA is the presence of a lymphocytic infiltrate in the scalp biopsies of affected patients. This, together with the response of patients to steroid treatment, has led to the suggestion of an autoimmune mechanism (^{Khoury et al, 1988;Welsh et al, 1994}). It was recently hypothesized that melanocyte-associated antigens could induce hair loss in AA (^{Gilhar et al, 2001}). However, an autoantigen has never been identified. On the other hand, haplotype association studies suggest that a permissive HLA status may potentiate the development of the AA phenotype (^{Kianto et al, 1977;Kuntz et al, 1977;Hacham-Zadeh et al, 1981;Morling et al, 1991;Welsh et al, 1994;Colombe et al, 1995;de Andrade et al, 1999}). Although studies describing a positive association between HLA and AA are numerous, there is a lack of biological data implicating specific alleles in the disease phenotype. Moreover, association between HLA and AA has been excluded in some familial cases of this disease (^{Zlotogorski et al, 1990}).

Alopecia areata: a complex trait

The term "complex trait" is used to describe those phenotypes that do not exhibit classic Mendelian inheritance attributable to a single gene locus but that does have a genetic component, as demonstrated by twin, adoption, and epidemiological studies (^{Lander and Schork, 1994}).

Polygenic inheritance of alopecia areata

No single resource of evidence for polygenic inheritance exists for AA, although it is becoming increasingly evident that a significant genetic predisposition underlies the AA phenotype (^{Green and Sinclair, 2000;McElwee et al, 2001;McDonagh and Tazi-Ahnini, 2002;McElwee and Hoffmann, 2002}). There are several independent lines of evidence in favor of polygenic inheritance (^{Aita and Christiano, 2001}), among them (i) the high prevalence of the trait, typical of complex traits for which the predisposing alleles are more common than the relatively rare mutations identified for Mendelian disorders; (ii) the Gaussian curve of distribution for both the stages of disease progression and the distribution of the disease, with a threshold effect that could be lowered by, for example, the presence of a particular HLA haplotype or autoimmune susceptibility (^{Welsh et al, 1994}); (iii) the heritability as defined by both the frequency of affected family members, ranging from 3% to 42% (^{Green and Sinclair, 2000}), and concordance in twins (^{Jackow et al, 1998}); and (iv) the presence of congenital AA, strongly suggesting the contribution of genetic factors (^{de Viragh et al, 1997;Bardazzi et al, 1999;Bereket et al, 2001;Crowder et al, 2002}). In addition, Sundberg and coworkers (Sundberg, this issue) reported the identification of potential susceptibility loci underlying the AA phenotype in the C3H/HeJ mouse model.

Association studies in alopecia areata

In a first attempt to identify the genetic basis of AA, a number of association studies for candidate genes have been conducted. Numerous reports indicate a significant association between AA and the HLA alleles DQB1^*0301 (for severe AA) and DRB1^*1104 (^{Morling et al, 1991;Welsh et al, 1994}). In a family-based study, DQB1^*03 alleles were shown to be present in 85% of AA patients as compared to 46% of controls (^{de Andrade et al, 1999}). Several authors have also suggested association with the IL-1 receptor antagonist gene (IL-1RN) and the IL-1 receptor antagonist homolog IL-1L1 (^{Tarlow et al, 1994;Barahamani et al, 2002;Tazi-Ahnini et al, 2002}), as well as with the MX1 gene on chromosome 21 (^{Tazi-Ahnini et al, 2000}). Finally, a strong association between autoimmune polyglandular syndrome type 1 (APS1), caused by mutations in the AIRE gene on chromosome 21, and AA has also been reported (^{Betterle et al, 1998}). In their study, Betterle and colleagues observed AA in 37% of their APS1 patients.

Linkage analysis in complex traits

A genome-wide scan on large cohorts of patients has been previously performed with success for the study of complex traits. For example, a number of groups have performed genome-wide scans in families affected with psoriasis and have identified at least three predisposing genetic loci on chromosomes 4, 6, and 17 (^{Bhalerao and Bowcock, 1998}). Atopic dermatitis, another complex skin disease, has also been the subject of genetic studies (^{Lee et al, 2000;Söderhäll et al, 2001;Bradley et al, 2002}). Lee and colleagues reported the identification of a major susceptibility locus on chromosome 3 (^{Lee et al, 2000}). It is anticipated that this type of study will lead to the identification of the pathomechanisms for these common diseases, as has been the case for Crohn's disease, with the identification of the actual alleles conferring susceptibility in the NOD2 gene (^{Hugot et al, 2001;Ogura et al, 2001}).

Identifying susceptibility genes for alopecia areata: The Genome-wide scan

As discussed above, the genome-wide linkage strategy has been previously applied to the study of complex traits. The design of a complex trait study is dependent on three variables: (i) the collection of families; (ii) the number and spacing of genetic markers; and (iii) the statistical power to identify a locus as a function of these choices. These three considerations of the study design are addressed in the following sections.

Diagnostic criteria and ascertainment of alopecia areata families

Critical to the success of any genome-wide initiative for a complex disease is a patient collection that fulfills a number of requirements: (i) accurate and uniform diagnosis; (ii) a large size for the results to be significant; (iii) a sample in which the contribution of genetic factors is enriched; and (iv) a sample amenable to analysis as a single group or as smaller, more homogeneous subgroups.

Although AA presents with some cardinal morphological and histological features, its diagnosis can be complicated by the variability in hair loss and the characteristic waxing and waning nature of the disease. To achieve homogeneity in the collection of patients, ascertainment, diagnosis, and collection of the families in our study are being undertaken by dermatologists with long-standing expertise in AA using the diagnostic questionnaire developed by the NIH AA Registry. The sources of the AA families recruited for this study comprise a large group of families from the Israeli Registry ascertained and diagnosed personally by A.Z., families from the NIH AA Registry, and families referred by physicians throughout the United States.

An example of the patient collection used in our study is shown in Figure 1 and summarized in Table 1, showing a family history in 35% of the patients examined. The genetic dissection of complex traits has traditionally been focused on large collections of small families (affected sib pairs, for example). It was recently shown, however, that a small sample of larger pedigrees can potentially derive better results (^{Tomfohrde et al, 1994;Hugot et al, 1996;Matthews et al, 1996}). The rationale for this is the following: (i) the use of a small sample reduces the level of genetic heterogeneity ("noise") among the pedigrees, one of the hallmarks of complex traits; (ii) a major contribution of genetic factors is suspected in pedigrees with several affec-ted family members (i.e., such families are "enriched" for predisposing alleles); and (iii) knowing the relationship between the individuals in a large pedigree results in higher statistical power (^{Terwilliger and Goring, 2000}). In order to maximize the success of our study, we have focused our initial efforts of the genome-wide scan in pedigrees with three or more affec-ted family members. Importantly, such a collection of families is amenable to the establishment of different subsets, based on severity of the disease (patchy AA, AT, or AU) or ethnic origin, among other criteria.

Figure 1.

Representative example of AA pedigrees available for this study.

Full figure and legend (30K)

Table 1 - Summary of AA patients examined.

Full table

Genetic markers

In order to achieve statistically significant results, a sufficient number of highly polymorphic markers must be genotyped. There are some reports on successful results with a low number of markers, such as the identification of the PSOR1 and IBD1 loci conferring susceptibility to psoriasis and Crohn's disease, respectively (^{Matthews et al, 1996;Trembath et al, 1997;Hugot et al, 2001}). In our study, however, we have chosen to perform a genome-wide scan using a panel of 324 microsatellite markers, with an average marker spacing of 10 cM and a semiautomated fluorescence-based genotyping system. Most of the markers are chosen from version 8.0 of the Marshfield fluorescence-labeled genome-screening set. This approach has been described in detail elsewhere (^{Aita et al, 1999;Liu et al, 2001;Liu et al, 2002}) and utilized by our group in collaboration with the Columbia University Genome Center (^{Ahmad et al, 1998a;Ahmad et al, 1998b;Martinez-Mir et al, 2002}).

Statistical analysis

Because of the inherent nature of complex traits, it is expected that a number of genetic components will contribute to the final presentation of the phenotype. However, different combinations of genetic factors can contribute to the disease in each family. For this reason, once the data-set from the genome-wide scan is obtained, it is extremely critical to perform a thorough and exhaustive statistical analysis to extract all possible information. This data-set is amenable to analysis with a large number and wide variety of statistical tests.

As a first approach, we will apply the following test statistics to the data-set obtained in the genome-wide scan: (i) the heterogeneity LOD score (^{Smith, 1963;Ott, 1999}), maximized over four settings of the penetrance parameters (MAXHLOD); and (ii) the mean test for affected sib pairs, as implemented in the ANALYZE program (^{Terwilliger and Ott, 1994}) (ASP), a test of allele sharing that uses all sibs (^{Terwilliger and Ott, 1994}) (ALLSIBS) and a likelihood version of the transmission disequilibrium test (^{Spielman et al, 1993}) as developed by Terwilliger (^{Terwilliger, 1995}) (TDT-LIKE).

With the MAXHLOD calculations, we will apply a model-based linkage analysis in which the LOD score is calculated under both autosomal dominant and autosomal recessive patterns of inheritance. For both models, two different values of penetrance will be considered: 50% and 80%. ASP, ALLSIBS, and TDT-LIKE tests have been chosen because they are all genetic model–free (^{Elston, 1989;Hodge and Elston, 1994}) in the sense that they do not require a specification of the genetic model parameters (penetrance and disease allele frequency). MAXHLOD has been chosen because it has been shown that it is at least as powerful in localizing disease loci as are tests like ASP and ALLSIBS (^{Abreu et al, 1999}), and under certain circumstances it is a more precise indicator of the location of a disease locus than are statistics like ASP and ALLSIBS (^{Finch et al, 2001}). Finally, TDT-LIKE will be chosen because it has been shown that it may be more powerful than linkage tests (MAXHLOD, ASP, ALLSIBS) when there is linkage and linkage disequilibrium between a disease and a marker locus. Some of the AA families collected by A.Z. in Israel are Ashkenazi Jewish (AJ). The AJ population is considered genetically isolated, and the extent of linkage disequilibrium is thought to extend over larger regions of the human genome (^{Ostrer, 2001}). The strategy to be used here is similar to exploratory methods employed by other researchers analyzing genome scan data for complex traits (^{Wise and Lewis, 1999;Wise et al, 1999}), with the exception that we will study a whole chromosome at a time rather than subdivide each chromosome into bins. We consider the whole chromosome as the unit of measure because in simulated data-sets it has been observed that the methods we plan to employ are more powerful at determining the correct chromosome rather than a particular region of a chromosome harboring a disease susceptibility locus (^{Gordon et al, 2001}).

Identification of genes and DNA variants conferring genetic susceptibility for alopecia areata

Once a complex trait has been mapped to one or several susceptibility loci, the task of identifying the specific alleles conferring susceptibility for the phenotype of interest still remains. First, a small interval of amenable size for positional cloning is rarely identified in initial linkage studies. Second, the nature of a complex disease implies that the alleles predisposing to the final phenotype can be numerous. Finally, DNA variants also present in the general population, rather than disease-restricted mutations, are the expected gene variants contributing to the etiology of complex traits. For this reason, multiple approaches will be utilized in the identification of candidate susceptibility genes and alleles for AA.

Fine mapping of susceptibility loci for alopecia areata

The results from the first stage of the genome-wide scan, as described above, are expected to indicate the most likely location for the AA susceptibility genes. In order to confirm the results from this first stage, as well as to exclude false positive results, a second stage of analysis will be performed. A dense map of polymorphic markers in the regions of interest will be developed using both microsatellite markers and SNPs. These are available in public databases: UCSC, NCBI, Ensembl, and deCODE Genetics (^{Kong et al, 2002}) for map and sequence information on microsatellites and SNPs; the Marshfield genetic map (^{Broman et al, 1998}) for map information on microsatellites; and the SNP consortium database (^{Sachidanandam et al, 2001}) for SNP selection. The entire collection of AA families, together with newly collected pedigrees, can then be genotyped. If SNP analysis is required for a large genomic region, and in order to optimize the genotyping effort in this refinement stage, the presence of haplotype blocks can be determined (^{Daly et al, 2001;Gabriel et al, 2002}). The use of haplotype blocks can considerably reduce the cost of genotyping, because a few SNPs are typed that represent the entire haplotype. Statistical analyses similar to those described above can be applied to exclude regions in the genome that could represent false positives and to map the actual susceptibility loci for AA.

Identification of candidate genes

Once the candidate regions cannot be further refined, it will be necessary to proceed with the analysis of the genes therein by both in silico and experimental procedures. We now have access to continuous updated sequence data of the human genome at three main public databases: UCSC, NCBI, and Ensembl. Based on the information available from these databases, a detailed physical map of the region of interest will be built in order to prioritize the candidate genes to be analyzed. This usually includes previously known genes, novel genes, ESTs, and predicted genes. As a second but parallel approach in the case of AA, we can cross-reference the positional data from the genome-wide scan with expression data. Carroll and coworkers have performed expression studies in human AA patients and in a mouse model for AA (^{Carroll et al, 2002}). They have distinguished two groups of differentially expressed genes for the initial and late stages of disease development. Finally, several animal models for AA have been described. Among them, the C3H/HeJ mouse and the DEBR rat have been extensively characterized (^{McElwee and Hoffmann, 2002}). Sundberg and coworkers (Sundberg, this issue) have reported the first attempt at identifying genes underlying the AA phenotype in the C3H/HeJ mouse model, with the identification of suggestive loci on chromosomes 9 and 17. The study of the syntenic regions containing AA loci in mice and humans will help in defining candidate genes.

Gene characterization and detection of DNA variants

Once a candidate gene is identified, the final goal will be the characterization of the actual alleles conferring susceptibility to the phenotype of interest. With this aim, the candidate gene will be sequenced in a collection of patients and control individuals.

Because genomic DNA is the material used in a genome-wide scan, it is necessary to know the genomic structure of the candidate gene for PCR primer design. The strategy followed will depend on the available information on each gene. For those genes fully sequenced as part of the Human Genome Project, the intron/exon boundaries are known or can otherwise be determined by cDNA and genomic sequence comparison. For those genes partially sequenced, a PCR-based cloning strategy can be followed. Finally, in the case of predicted genes the predicted gene structure must be verified by RT-PCR.

For each candidate gene, intron/exon boundaries will eventually be identified, as well as flanking intron sequences for the design of exon-specific PCR amplification primers for the detection of DNA variants. For each gene, both coding and noncoding regions—mainly the promoter region—can then be analyzed. The first samples to be sequenced will usually be those corresponding to the pedigrees linked to a particular chromosomal region. Once DNA variants are identified, both in coding and noncoding sequences, it will be necessary to test whether they are significantly associated with the disease phenotype. With this goal, their frequency can be determined in the complete collection of AA families, isolated AA cases, and a control-matched population. If needed, the cohort of AA pedigrees can be further subdivided according to severity and ethnic origin to test for associations between the variants identified and particular subgroups of the patient collection.

We anticipate that this type of study will provide a foundation for understanding the interactions of these genes with each other and with other variables such as the immune system and environmental triggers. Ultimately, it is expected that this study will help to define therapeutic targets for the future and eventually provide an effective treatment for this psychologically devastating dermatologic disorder.

Electronic database information

The URLs for data presented herein are as follows:

Center for Medical Genetics, Marshfield Medical Research Foundation, http://research.marshfieldclinic.org/genetics/

Ensembl Genome Server, http://us.ensembl.org/

National Alopecia Registry (NAAR) http://www.AlopeciaAreataRegistry.org

National Center for Biotechnology Information (NCBI), http://www.ncbi.nlm.nih.gov

TDT-AE Software, http://linkage.rockefeller.edu/soft/list4.htmltdtae

UCSC Human Genome Project Working Draft ('Golden Path'), http://genome.ucsc.edu/index.html

References

1. Abreu, PC, Greenberg, DA, Hodge, SE: Direct power comparisons between simple LOD scores and NPL scores for linkage analysis in complex diseases. Am J Hum Genet 65:847–857, 1999 | Article | PubMed | ISI | ChemPort |
2. Ahmad, W, Brancolini, V, ul Haque, MF, et al: A locus for autosomal recessive hypodontia with associated dental anomalies maps to chromosome16q12.1. Am J Hum Genet 62:987–991, 1998a
3. Ahmad, W, ul Haque, MF, Brancolini, V, et al: Alopecia universalis associated with a mutation in the human hairless gene. Science 279:720–724, 1998b | Article | PubMed | ISI | ChemPort |
4. Aita, VM, Christiano, AM: The genetics of alopecia areata. Dermatol Ther 14:329–339, 2001
5. Aita, VM, Liu, J, Knowles, JA, et al: A comprehensive linkage analysis of chromosome 21q22 supports prior evidence for a putative bipolar affective disorder locus. Am J Hum Genet 64:210–217, 1999 | Article | PubMed | ISI | ChemPort |
6. de Andrade, M, Jackow, CM, Dahm, N, Hordinsky, M, Reveille, JD, Duvic, M: Alopecia areata in families: Association with the HLA locus. J Invest Dermatol Symp Proc 4:220–223, 1999 | ISI | ChemPort |
7. Barahamani, N, de Andrade, M, Slusser, J, Zhang, Q, Duvic, M: Interleukin-1 receptor antagonist allele 2 and familial alopecia areata. J Invest Dermatol 118:335–337, 2002 | Article | PubMed | ISI | ChemPort |
8. Bardazzi, F, Neri, I, Raone, B, Patrizi, A: Congenital alopecia areata: Another case. Dermatology 199:369, 1999 | Article | PubMed | ISI | ChemPort |
9. Bereket, A, Turan, S, Alper, G, Comu, S, Alpay, H, Akalin, F: Two patients with Kabuki syndrome presenting with endocrine problems. J Pediatr Endocrinol Metab 14:215–220, 2001
10. Betterle, C, Greggio, NA, Volpato, M: Clinical review 93: Autoimmune polyglandular syndrome type 1. J Clin Endocrinol Metab 83:1049–1055, 1998 | Article | PubMed | ISI | ChemPort |
11. Bhalerao, J, Bowcock, AM: The genetics of psoriasis: A complex disorder of the skin and immune system. Hum Mol Genet 7:1537–1545, 1998 | Article | PubMed | ISI | ChemPort |
12. Bradley, M, Soderhall, C, Luthman, H, Wahlgren, CF, Kockum, I, Nordenskjold, M: Susceptibility loci for atopic dermatitis on chromosomes 3, 13, 15, 17 and 18 in a Swedish population. Hum Mol Genet 11:1539–1548, 2002 | Article | PubMed | ISI | ChemPort |
13. Broman, KW, Murray, JC, Sheffield, VC, White, RL, Weber, JL: Comprehensive human genetic maps: Individual and sex-specific variation in recombination. Am J Hum Genet 63:861–869, 1998 | Article | PubMed | ISI | ChemPort |
14. Carroll, JM, McElwee, KJ, King, LE, Byrne, MC, Sundberg, JP: Gene array profiling and immunomodulation studies define a cell-mediated immune response underlying the pathogenesis of alopecia areata in a mouse model and humans. J Invest Dermatol 119:392–402, 2002 | Article | PubMed | ISI | ChemPort |
15. Colombe, BW, Price, VH, Khoury, EL, Garovoy, MR, Lou, CD: HLA class II antigen associations help to define two types of alopecia areata. J Am Acad Dermatol 33:757–764, 1995 | ISI | ChemPort |
16. Crowder, JA, Frieden, IJ, Price, VH: Alopecia areata in infants and newborns. Pediatr Dermatol 19:155–158, 2002
17. Daly, MJ, Rioux, JD, Schaffner, SF, Hudson, TJ, Lander, ES: High-resolution haplotype structure in the human genome. Nat Genet 29:229–232, 2001 | Article | PubMed | ISI | ChemPort |
18. Elston, RC: Man bites dog? The validity of maximizing LOD scores to determine mode of inheritance. Am J Med Genet 34:487–488, 1989
19. Finch, SJ, Chen, CH, Gordon, D, Mendell, NR: A study comparing the precision of the maximum heterogeneity LOD statistic to two model free linkage methods. Genetic Epidemiol 21:315–325, 2001
20. Gabriel, SB, Schaffner, SF, Nguyen, H, et al: The structure of haplotype blocks in the human genome. Science 296:2225–2229, 2002 | Article | PubMed | ISI | ChemPort |
21. Gilhar, A, Landau, M, Assy, B, Shalaginov, R, Serafimovich, S, Kalish, RS: Melanocyte-associated T cell epitopes can function as autoantigens for transfer of alopecia areata to human scalp explants on Prkdc (SCID) mice. J Invest Dermatol 117:1357–1362, 2001 | Article | PubMed | ISI | ChemPort |
22. Gordon, D, Hoh, J, Finch, SJ, et al: Two approaches for consolidating results from genome scans of complex traits: selection methods and scan statistics. Genet Epidemiol Supplement 21:S396–S402, 2001
23. Green, J, Sinclair, RD: Genetics of alopecia areata. Australas J Dermatol 41:213–218, 2000
24. Hacham-Zadeh, S, Brautbar, C, Cohen, CA, Cohen, T: HLA and alopecia areata in Jerusalem. Tissue Antigens 18:71–74, 1981 | PubMed | ChemPort |
25. Hodge, SE, Elston, RC: Lods, wrods, and mods: the interpretation of lod scores calculated under different models. Genet Epidemiol 11:329–342, 1994 | Article | PubMed | ISI | ChemPort |
26. Hugot, JP, Chamaillard, M, Zouali, H, et al: Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 411:599–603, 2001 | Article | PubMed | ISI | ChemPort |
27. Hugot, JP, Laurent-Puig, P, Gower-Rousseau, C, et al: Mapping of a susceptibility locus for Crohn's disease on chromosome 16. Nature 379:821–823, 1996 | Article | PubMed | ISI | ChemPort |
28. Jackow, C, Puffer, N, Hordinsky, M, Nelson, J, Tarrand, J, Duvic, M: Alopecia areata and cytomegalovirus infection in twins: genes versus environment? J Am Acad Dermatol 38:418–425, 1998 | PubMed | ISI | ChemPort |
29. Khoury, EL, Price, VH, Greenspan, JS: HLA-DR expression by hair follicle keratinocytes in alopecia areata: evidence that it is secondary to the lymphoid infiltration. J Invest Dermatol 90:193–200, 1988 | Article | PubMed | ISI | ChemPort |
30. Kianto, U, Reunala, T, Karvonen, J, Lassus, A, Tiilikainen, A: HLA-B12 in alopecia areata. Arch Dermatol 113:1716, 1977 | Article | PubMed | ISI | ChemPort |
31. Kong, A, Gudbjartsson, DF, Sainz, J, et al: A high-resolution recombination map of the human genome. Nat Genet 31:241–247, 2002 | Article | PubMed | ISI | ChemPort |
32. Kuntz, BM, Selzle, D, Braun-Falco, O, Scholz, S, Albert, ED: HLA antigens in alopecia areata. Arch Dermatol 113:1717, 1977
33. Lander, ES, Schork, NJ: Genetic dissection of complex traits. Science 265:2037–2048, 1994 | PubMed | ISI | ChemPort |
34. Lee, YA, Wahn, U, Kehrt, R, et al: A major susceptibility locus for atopic dermatitis maps to chromosome 3q21. Nat Genet 26:470–473, 2000 | Article | PubMed | ISI | ChemPort |
35. Liu, J, Juo, SH, Holopainen, P, et al: Genomewide linkage analysis of celiac disease in Finnish families. Am J Hum Genet 70:51–59, 2002 | Article | PubMed | ISI | ChemPort |
36. Liu, J, Nyholt, DR, Magnussen, P, et al: A genomewide screen for autism susceptibility loci. Am J Hum Genet 69:327–340, 2001 | Article | PubMed | ISI | ChemPort |
37. Madani, S, Shapiro, J: Alopecia areata update. J Am Acad Dermatol 42:549–566, 2000 | Article | PubMed | ISI | ChemPort |
38. Martinez-Mir, A, Liu, J, Gordon, D, et al: EB simplex superficialis resulting from a mutation in the type VII collagen gene. J Invest Dermatol 118:547–549, 2002
39. Matthews, D, Fry, L, Powles, A, et al: Evidence that a locus for familial psoriasis maps to chromosome 4q. Nat Genet 14:231–233, 1996 | Article | PubMed | ISI | ChemPort |
40. McDonagh, AJ, Messenger, AG: The pathogenesis of alopecia areata. Dermatol Clin 14:661–670, 1996 | PubMed | ISI | ChemPort |
41. McDonagh, AJ, Tazi-Ahnini, R: Epidemiology and genetics of alopecia areata. Clin Exp Dermatol 27:409–413, 2002
42. McElwee, KJ, Freyschmidt-Paul, P, Ziegler, A, Happle, R, Hoffmann, R: Genetic susceptibility and severity of alopecia areata in human and animal models. Eur J Dermatol 11:11–16, 2001 | PubMed | ISI | ChemPort |
43. McElwee, KJ, Hoffmann, R: Alopecia areata – animal models. Clin Exp Dermatol 27:414–421, 2002
44. Morling, N, Frentz, G, Fugger, L, Georgsen, J, Jakobsen, B, Odum, N, Svejgaard, A: DNA polymorphism of HLA class II genes in alopecia areata. Dis Markers 9:35–42, 1991 | PubMed | ISI | ChemPort |
45. Ogura, Y, Bonen, DK, Inohara, N, et al: A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 411:603–606, 2001 | Article | PubMed | ISI | ChemPort |
46. Ostrer, H: A genetic profile of contemporary Jewish populations. Nat Rev Genet 2:891–898, 2001 | Article | PubMed |
47. Ott, J: Analysis of Human Genetic Linkage. Baltimore: The Johns Hopkins University Press, 1999
48. Price, VH: Alopecia areata: Clinical aspects. J Invest Dermatol 96:68S, 1991
49. Sachidanandam, R, Weissman, D, Schmidt, SC, et al: A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 409:928–933, 2001 | Article | PubMed | ISI | ChemPort |
50. Sawaya, ME, Hordinsky, MK: Advances in alopecia areata and androgenetic alopecia. Adv Dermatol 7:211–226, 1992
51. Schwartz, RA, Janniger, CK: Alopecia areata. Cutis 59:238–241, 1997
52. Smith, CAB: Testing for heterogeneity of recombination fraction values in human genetics. Ann Hum Genet 27:175–182, 1963 | PubMed | ISI | ChemPort |
53. Söderhäll, C, Bradley, M, Kockum, I, Wahlgren, C-F, Luthman, H, Nordenskjöld, M: Linkage and association to candidate regions in Swedish atopic dermatitis families. Hum Genet 109:129–135, 2001
54. Spielman, RS, McGinnis, RE, Ewens, WJ: Transmission test for linkage disequilibrium. The insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am J Hum Genet 52:506–516, 1993 | PubMed | ISI | ChemPort |
55. Tarlow, JK, Clay, FE, Cork, MJ, Blakemore, AI, McDonagh, AJ, Messenger, AG, Duff, GW: Severity of alopecia areata is associated with a polymorphism in the interleukin-1 receptor antagonist gene. J Invest Dermatol 103:387–390, 1994 | Article | PubMed | ISI | ChemPort |
56. Tazi-Ahnini, R, Cox, A, McDonagh, AJ, et al: Genetic analysis of the interleukin-1 receptor antagonist and its homologue IL-1L1 in alopecia areata: Strong severity association and possible gene interaction. Eur J Immunogenet 29:25–30, 2002 | Article | PubMed | ISI | ChemPort |
57. Tazi-Ahnini, R, di Giovine, FS, McDonagh, AJ, et al: Structure and polymorphism of the human gene for the interferon-induced p78 protein (MX1): Evidence of association with alopecia areata in the Down syndrome region. Hum Genet 106:639–645, 2000 | PubMed |
58. Terwilliger, JD: A powerful likelihood method for the analysis of linkage disequilibrium between trait loci and one or more polymorphic marker loci. Am J Hum Genet 56:777–787, 1995 | PubMed | ISI | ChemPort |
59. Terwilliger, JD, Goring, HH: Gene mapping in the 20th and 21st centuries. Statistical methods, data analysis, and experimental design. Hum Biol 72:63–132, 2000 | PubMed | ISI | ChemPort |
60. Terwilliger, JD, Ott, J: Handbook of human genetic linkage. Baltimore: The Johns Hopkins University Press, 1994
61. Tomfohrde, J, Silverman, A, Barnes, R, et al: Gene for familial psoriasis susceptibility mapped to the distal end of human chromosome 17q. Science 264:1141–1145, 1994 | PubMed | ISI | ChemPort |
62. Trembath, RC, Clough, RL, Rosbotham, JL, et al: Identification of a major susceptibility locus on chromosome 6p and evidence for further disease loci revealed by a two stage genome-wide search in psoriasis. Hum Mol Genet 6:813–820, 1997 | Article | PubMed | ISI | ChemPort |
63. de Viragh, PA, Gianadda, B, Levy, ML: Congenital alopecia areata. Dermatology 195:96–98, 1997 | PubMed | ChemPort |
64. Welsh, EA, Clark, HH, Epstein, SZ, Reveille, JD, Duvic, M: Human leukocyte antigen-DQB1*03 alleles are associated with alopecia areata. J Invest Dermatol 103:758–763, 1994 | Article | PubMed | ISI | ChemPort |
65. Wise, LH, Lanchbury, JS, Lewis, CM: Meta-analysis of genome searches. Ann Hum Genet 63:263–272, 1999 | Article | PubMed | ISI |
66. Wise, LH, Lewis, CM: A method for meta-analysis of genome searches: Application to simulated data. Genet Epidemiol Supplement 17:S767–S771, 1999
67. Zlotogorski, A, Weinrauch, L, Brautbar, C: Familial alopecia areata: No linkage with HLA. Tissue Antigens 36:40–41, 1990

Acknowledgements

The authors are grateful to the NIH AA Registry for its invaluable contribution to this study. This study was supported in part by grants from the National Alopecia Areata Foundation (AMC and AZ), by the General Administrator, Israel (AZ), and NIH grants N01-AR-0–2249 (NIH AA Registry), K01-HG00055 (DG), MH59492 (JO), and HG00008 (JO).