The Role of Hemochromatosis Susceptibility Gene Mutations in Protecting Against Iron Deficiency in Celiac Disease

Transcription

1 GASTROENTEROLOGY 2002;123: The Role of Hemochromatosis Susceptibility Gene Mutations in Protecting Against Iron Deficiency in Celiac Disease JEFFREY R. BUTTERWORTH,* BRIAN T. COOPER,* WILLIAM M. C. ROSENBERG, MICHAEL PURKISS, SHIRLEY JOBSON, MARK HATHAWAY, DAVID BRIGGS, W. MARTIN HOWELL, GORDON M. WOOD, DAVID H. ADAMS, # and TARIQ H. IQBAL* *Gastroenterology Unit, City Hospital, Birmingham; Liver Group, Division of Infection, Inflammation, and Repair, and Divisionof Human Genetics, Southampton General Hospital, University of Southampton, Southampton; National Blood Service, Birmingham; Department of Medicine, George Elliot Hospital, Nuneaton; and # Liver Research Laboratories, MRC Centre for Immune Regulation, University of Birmingham, Birmingham, England Background & Aims: Celiac disease and hereditary hemochromatosis are common HLA-defined conditions in northwestern Europe. We sought to determine whether there is a genetic relationship between the 2 diseases and if hemochromatosis susceptibility gene (HFE) mutations are protective against iron deficiency in celiac disease. Methods: Polymerase chain reaction amplification using sequence-specific primers capable of identifying the 2 HFE gene mutations (H63D and C282Y) and the HLA class I and II alleles was used to type 145 white patients with celiac disease and 187 matched controls. Hemoglobin and fasting serum iron levels in celiac patients were measured at diagnosis. Results: HFE gene mutations, H63D or C282Y, were identified in 70 celiac patients (48.3%) and 61 controls (32.6%) (P 0.004). The C282Y mutation was associated with HLA-A*03 and B*07 alleles in controls and with A*01, A*03, B*08, and DRB1*0301 alleles in celiac patients; the H63D mutation was associated with HLA-A*25 and DRB1*03 alleles in controls and A*29 and DRB1*03 alleles in celiac patients. At diagnosis, celiac patients with the C282Y mutation had higher mean hemoglobin and fasting serum iron levels compared with the HFE wild type (P and 0.006, respectively). This was not observed with the H63D mutation. Conclusions: In celiac disease, HFE gene mutations are common and are in linkage disequilibrium with different HLA alleles compared with controls. A disease-specific haplotype that carries C282Y and DQB1*02 is suggested. We propose that HFE gene mutations provide a survival advantage by ameliorating the iron deficiency seen in celiac patients. Celiac disease and hereditary hemochromatosis (HHC) are both most prevalent in Celtic populations, and susceptibility to both is associated with genes in the major histocompatibility complex (MHC) on the short arm of chromosome 6.However, their MHC associations are quite distinct.celiac disease is most strongly associated with HLA-DQ2, 1 3 determined by DQB1*0201 and DQA1*0501, which are in strong linkage disequilibrium with HLA-A1, B8, and DR3. 4,5 Despite the strong association with the HLA-DQ2 allele, the inheritance pattern of celiac disease is not clear-cut, with the percentage contribution of the MHC region to the development of celiac disease between 30% and 40%. 6,7 HHC is an autosomal-recessive disease in which susceptibility is conferred by mutations of the hemochromatosis susceptibility gene (HFE) locus, which is telomeric to the classical HLA class I loci but exhibits strong linkage disequilibrium with the HLA-A3 and B7 alleles. 8 The C282Y mutation of HFE results in an abnormality in protein trafficking and absent cell surface expression of the HFE protein.iron modulation at the level of the small bowel crypt is disrupted, resulting in an increase in duodenal iron absorption (via the divalent metal transporter protein 9 ) and leading over decades to iron overload in organs such as the liver, heart, and pancreas. The allele frequency of the C282Y mutation varies among populations, with the highest frequencies seen in individuals of northwestern European origin (frequency, 6.4%) compared with South Asians and native, nonwhite Americans and Australasians (frequencies up to 0.7%). 10 Homozygosity for C282Y is responsible for between 80% and 100% of cases of HHC. 11 The H63D mutation is the most common mutation identified within the HFE gene and is more frequent in those of Celtic descent (allele frequency, 10% 13% compared with 0% 3% for Afro-Caribbean and Asian populations). 10 It has recently been found to be in linkage Abbreviations used in this paper: HFE, hemochromatosis susceptibility gene; HHC, hereditary hemochromatosis; MHC, major histocompatibility complex by the American Gastroenterological Association /02/$35.00 doi: /gast

2 August 2002 HFE MUTATIONS IN CELIAC DISEASE 445 disequilibrium with the HLA-A*29 allele. 12 Unlike the C282Y mutation, the HFE protein with this H63D mutation is not prevented from associating with the cell surface transferrin receptor 2-microglobulin complex. Although the physiologic consequences of the H63D heterozygous state are not fully known, North American and European studies have shown a highly significant 2-fold enrichment of HHC cases with H63D compared with random control cases. 11,13 21 When all the data from Europe and North America (including the U.K. Haemochromatosis Consortium 22 ) are combined, the association of H63D with HHC is highly significant. Testing the effect of the H63D mutation independently of the C282Y mutation has shown a 3-fold enrichment of the H63D homozygotes in cases of HHC compared with a control population. 23 Therefore, the H63D mutation can be considered a hemochromatosis-associated allele. The observation that H63D and C282Y have never been found on the same chromosome provides further evidence to support this view. Thus, celiac disease and HHC are both relatively common, inheritable diseases associated with the HLA region on chromosome 6, with celiac disease associated with the HLA-DQ2 allele as well as non-hla genes and HHC associated with the HLA-A3 and B7 alleles of the MHC region.celiac disease is characterized by iron deficiency at presentation, 24 and it might be expected that genetic pressures would operate to select mutations in the HFE gene to ameliorate this iron deficiency and provide a nutritional advantage in an iron-poor environment.however, the 2 conditions are associated with different HLA haplotypes and therefore would be expected to be negatively associated.patients with both celiac disease and hemochromatosis have been reported. 25,26 In the absence of any previous data, we sought to determine if there is a genetic relationship between the 2 diseases and if HFE gene mutations are protective against iron deficiency in patients with celiac disease. Materials and Methods Subjects The study included 145 white adult patients with celiac disease diagnosed at City Hospital, Birmingham, Southampton General Hospital, and the George Elliot Hospital (Nuneaton, Warwickshire) based on the characteristic histologic appearances of subtotal or total villous atrophy on small bowel biopsy.the population control group included 187 consecutive ethnically matched cadaveric organ donors whose livers had been transplanted at the Liver Unit, Queen Elizabeth Hospital (Edgbaston, Birmingham), between 1994 and 1996.Controls were matched with the patient group by selecting only those of northern European descent.serum hemoglobin and iron levels at diagnosis were available for 105 and 50 patients with celiac disease, respectively.the study was approved by the research ethics committees of the hospitals involved.all patients with celiac disease provided informed consent to participate in the study. Methods Genomic DNA was extracted from EDTA anticoagulated peripheral blood by a salting out method. 27 Polymerase chain reaction amplifications to detect the 2 common HFE gene mutations (C282Y and H63D) were performed in 145 patients with celiac disease using sequence-specific primers (supplied by Alta Bioscience, University of Birmingham) as previously described by Mullighan et al. 28 HLA typing was performed on 99 patients using sequence-specific primers as previously described by Bunce et al. 29 Serum iron and hemoglobin levels were determined by standard laboratory techniques. Data Analysis and Statistics Mutation frequencies were analyzed as carrier or phenotype frequencies (the proportion for an allele or mutation) and allele frequencies (proportion of positive chromosomes, determined by gene counting).phenotype and allele frequencies of the HFE gene mutations among patients and controls were compared by Fisher exact test.linkage disequilibrium between HFE alleles and HLA class I and II alleles (Table 1) was assessed by 2 analysis with Yates correction. 2 P value was calculated by comparing the number of positive haplotypes separately within the celiac and control groups. Differences between the mean of the hemoglobin and serum iron levels between those patients with celiac disease with and without HFE gene mutations were calculated by Student t test. A 2-sided P value of 0.05 was considered to indicate statistical significance. Results Significantly more patients with celiac disease (70 of 145, 48.3%) had HFE gene mutations compared with the control population (61 of 187, 32.6%; P 0.004). Table 2 shows the HFE genotype and haplotype frequencies in the patients with celiac disease and the control population.the carrier frequency for the C282Y mutation was 0.17 for the patients with celiac disease, which is significantly higher than for the control population (0.09; P 0.04). For the H63D mutation, the carrier frequency was 0.33 among the patients with celiac disease, which was significantly greater than that for the controls (0.23; P 0.05). The allele frequency of the C282Y mutation was not statistically significant in the patients with celiac disease compared with the controls

3 446 BUTTERWORTH ET AL. GASTROENTEROLOGY Vol. 123, No. 2 Table 1. AssociationBetweenHFE Mutations and HLA Alleles in 198 Haplotypes in Patients With Celiac Disease and 374 Control Haplotypes Patients with celiac disease Control population Haplotype combination No. Frequency 2 2 P No. Frequency 2 2 P C282Y-A* C282Y-A* NS C282Y-B* NS C282Y-B* NS C282Y-DRB1* NS C282Y-DQB1* NS NS H63D-A* NS H63D-A* NS H63D-DRB1* H63D-DQB1* NS NS NOTE. Numbers for the patients with celiac disease and controls refer to those positive for both alleles indicated at the 2 loci. Frequencies were calculated by dividing the numbers of counted haplotypes by the total number of chromosomes studied (198 for the patients with celiac disease and 374 for the control population). Haplotypes are listed as combinations of class I or II alleles with C282Y or H63D alleles. 2 analysis was used to assess the differences between the different categories within the celiac and control groups. Only specific combinations are shown (i.e., those showing significance or previously known to be relevant). (0.08 vs. 0.05, respectively; P 0.07). The allele frequency of the H63D mutation was 0.17 for the patients with celiac disease and 0.13 for the controls (P 0.17). Table 3 shows the frequencies of individuals with selected HLA alleles in the celiac and control populations.the patients with celiac disease have the expected associations with each of the HLA alleles: A*01, B*08, DRB1*03, and DQB1*02.Sixty-four (65.3%) of the patients carry these alleles compared with 30 (16%) of the controls (P ). The frequency of B*07 is significantly reduced in the patient group (P ). Linkage disequilibrium data for H63D and C282Y with HLA class I and II alleles are presented in Table 1. Among the control population, the expected linkage disequilibrium between C282Y and HLA-A*03 (P ) and between C282Y and HLA-B*07 (P ) is confirmed. The H63D mutation was found to be in linkage disequilibrium with HLA-A*25 (P ) and HLA-DRB1*03 (P 0.005), which has not been previously reported.in the patients with celiac disease, there is a statistically significant link between the H63D mutation and the A*29 and DRB1*03 alleles. In addition, there is significant linkage disequilibrium between the C282Y mutation and the HLA-A*01 (P 0.006), A*03 (P 0.014), B*08 (P 0.02), and DRB1*03 (P 0.02) alleles. No linkage disequilibrium Table 2. HFE Genotype and Haplotype Frequencies in the Patient and Control Groups Patients with celiac disease Control population No. (n 145) Frequency No. (n 187) Frequency Genotype Codon63 63HH HD DD Codon CC CY YY Haplotype Codons H-282C D-282C H-282Y D-282Y Compound heterozygotes (63H-282Y and H63D-282C)

4 August 2002 HFE MUTATIONS IN CELIAC DISEASE 447 Table 3. Frequencies of Selected HLA Alleles in the Patient and Control Groups Estimated by Gene Counting Allele Haplotype counts Patients with celiac disease (n 198) Control population (n 374) was shown between the HLA-B*07 allele and either HFE gene mutation in the patients. A comparison of the patients with celiac disease and the control population shows that, among the patients with celiac disease, the C282Y mutation is in linkage disequilibrium more frequently with the HLA-A*01 (P 0.001), B*08 (P ), DRB1*03 (P ), and DQB1*02 (P 0.007) alleles. For those patients with celiac disease with the H63D mutation, statistically more have HLA-A*29 (P 0.04), DRB1*03 (P ), and DQB1*02 (P 0.03) alleles in combination with H63D compared with controls (information from Table 1). The fasting serum iron and hemoglobin levels at diagnosis were significantly higher in those patients with celiac disease with the C282Y mutation compared with the HFE wild-type genotype (Table 4).In contrast, the H63D mutations were not found to be associated with significantly higher fasting serum iron and hemoglobin levels compared with the HFE wild-type genotype (Table 4). Discussion This report describes the novel observation of a significantly high frequency of mutations of the MHClinked locus HFE in a cohort of patients with celiac disease.patients with both HHC and celiac disease have been described, 25,26 and our genetic analysis is consistent with and perhaps may explain this.none of our patients was found to have hemochromatosis, but we speculate that abnormalities of HFE are likely to be beneficial to patients with celiac disease; indeed, this was seen to be the case.thus, we showed C282Y mutation to be protective against the iron deficiency anemia commonly seen 2 2 P A* A* NS B* B* DRB1* DQB1* NOTE. Elevenof 15 patients with celiac disease with a C282Y mutationwere HLA-A*01, B*08, DRB1*03, and DQB1*02 positive compared with 3 of 15 patients in the control group (P ). n the number of chromosomes (2 number of individuals). Homozygotes were counted twice. in patients with celiac disease.the results of this study are therefore of clinical significance, with the C282Y mutation affording protection from iron deficiency and possibly delaying the onset of presentation of the disease. We confirmed the widely reported strong association of DQB1*02 with celiac disease together with high frequencies of A*01, B*08, and DRB1*03.We also identified a significantly low frequency of B*07, suggesting a protective effect of this allele or with B*07-associated genes.within the control population, strong linkage disequilibrium was shown between the C282Y mutation and the HLA-A*03 and B*07 alleles as expected. 12,28 A new finding was the observation of linkage disequilibrium between the H63D mutation and the HLA-A*25 and DRB1*03 alleles in a control group. Previous studies have described linkage between the H63D mutation and the HLA-A*29 allele, 12,28 which we also have been able to confirm among our cohort of patients with celiac disease. In light of the HLA and HFE allelic associations seen in the control group (Table 1), an increased frequency of HFE mutations in a group of patients with celiac disease is unexpected.first, the HLA alleles known to be linked to HFE mutations (i.e., HLA-A*03 and B*07) are not associated with this patient group (in which the HLA- A*01, B*08, DRB1*03, and DQB1*02 alleles predominate).second, in the general population, the MHC alleles associated with celiac disease are not known to be in linkage disequilibrium with HFE C282Y or H63D.A reduced frequency of HFE mutations would therefore be expected in a patient group characterized by such a high frequency of DQ2.This implies a mechanism selecting 2 functionally independent genetic factors against negative linkage disequilibrium.the first factor, DQ2, is detrimental in that it predisposes to the disease, whereas HFE C282Y seems to be advantageous in the context of this disease.it will be important to determine whether these factors are inherited on the same or different haplotypes in patients with celiac disease.in the former case, this Table 4. Fasting Iron and Hemoglobin Levels at Diagnosis inthe Patients With Celiac Disease Meaniron( SD) at diagnosis (mmol/l) (n 50) Meanhemoglobin( SD) at diagnosis (g/dl) (n 105) HFE wild type C282Y mutation a b H63D mutation c c NOTE. Difference compared with HFE wild type. a P b P c P 0.45 (mean iron) and P 0.12 (meanhemoglobin).

5 448 BUTTERWORTH ET AL. GASTROENTEROLOGY Vol. 123, No. 2 would imply a disease-specific haplotype comprising HFE C282Y, HLA-A*01, B*08, DRBI*03, and DQB1*02.This would be distinct from the common Caucasoid haplotype, which contains the same HLA alleles but lacks the HFE mutation.there is evidence for this in our data in that 11 of 15 patients with C282Y also carry HLA-A*01, B*08, DRBI*03, and DQB1*02 compared with 3 of 15 in the control group (P ). Furthermore, in the control group, C282Y associates with HLA-A*03 and B*07, but this is almost absent in the patients (one case). Iron deficiency anemia is common at diagnosis in patients with celiac disease 24 as a consequence of iron malabsorption resulting from villous atrophy. HFE gene mutations may be expected to ameliorate this iron deficiency by enhancing iron absorption from the small bowel, protecting against the development of anemia. Healthy women heterozygous for the C282Y mutation have higher hemoglobin levels, serum iron levels, and transferrin saturation compared with women with the HFE wild type, and this mutation offers protection from iron deficiency anemia. 13,23 We have shown that, at the time of diagnosis, patients with celiac disease with the C282Y mutation have higher fasting serum iron and hemoglobin levels than those with the wild-type HFE genotype.this was not seen in patients with celiac disease with the H63D mutation.the lack of an association between the H63D mutation and higher serum iron or hemoglobin levels at diagnosis does not exclude the possible role for this mutation in modulating body iron homeostasis.heterozygosity for either HFE gene mutation has been shown to be significantly associated with increased serum iron levels and transferrin saturation. 30 In conclusion, the current study provides novel observations concerning HFE gene mutations in patients with celiac disease.we have shown that, in these patients, the C282Y HFE gene mutation is not in linkage disequilibrium with HLA-B*07 as normally found in northwestern European populations.linkage disequilibrium between C282Y and HLA-A*01 and B*08 is not seen in the normal population but was observed in the patient group.it remains to be determined whether there is linkage disequilibrium between these alleles in this disease or whether disease susceptibility through DQ2 and protection against iron deficiency due to HFE mutations are inherited in separate gametes.the former would suggest a disease-specific haplotype extending from HFE to DQ, and this will be tested formally in family studies. Should this be the case, it will be of interest to investigate the origins of the genetic linkage, particularly in Celtic populations in which there is a peculiarly high incidence of both celiac disease and hemochromatosis. References 1. Mazzilli MC, Ferrante P, Mariani P, Martone E, Petronzelli F, Triglione P, Bonamico M. A study of Italian pediatric coeliac disease patients confirms that the primary HLA association is to the DQ ( 1*0501, 1*0201) heterodimer. Hum Immunol 1992; 33: Tighe MR, Hall MA, Barbado M, Cardi E, Welsh KI, Ciclitira PJ. HLA class II alleles associated with coeliac disease susceptibility ina southern European population. Tissue Antigens 1992;40: Sollid LM, MarkussenG, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a primary association of coeliac disease to a particular HLA-DQ alpha/beta heterodimer. J Exp Med 1989;169: Falchuk ZM, Rogentine GN, Strober W. Predominance of histocompatibility antigen HLA-B8 in patients with gluten-sensitive enteropathy. J Clin Invest 1972;51: Stokes PL, Asquith P, Holmes GKT, Mackintosh P, Cooke WT. Histocompatibility antigens associated with coeliac disease. Lancet 1972;2: Petronzelli F, Bonamico M, Ferrante P, Grillo R, Mora B, Mariani P, Apollonio I, Gemme G, Mazzilli MC. Genetic contribution of the HLA region to the familial clustering of coeliac disease. Ann Hum Genet 1997;61: BevanS, Popat S, Braegger CP, Busch A, O Donoghue D, Falth- MagnussonK, FergusonA, GodkinA, Hogberg L, Holmes G, Hosie KB, Howdle PD, Jenkins H, Jewell D, Johnston S, Kennedy NP, Kerr G, Kumar P, LoganRFA, Love AHG, Marsh M, Mulder CJJ, Sjoberg K, Stenhammer L, Walker-Smith J, Marossy AM, Houlston RS. Contributionof the MHC regionto the familial risk of coeliac disease. J Med Genet 1999;36: Edwards CQ, Dadone MM, Skolnick MH, Kushner JP. Hereditary hemochromatosis. ClinHaematol 1982;11: Gunshin H, MacKenzie B, Berger UV, Gunshin Y, Romero MF, Boron WF, Nussberger S, Gollan JL, Hediger MA. Cloning and characterisationof a mammalianproton-coupled metal-iontransporter. Nature 1997;388: Merryweather-Clarke AT, PointonJJ, ShearmanJD, RobsonKJ. Global prevalence of putative haemochromatosis mutations. J Med Genet 1997;34: Feder JN, Gnirke A, Thomas W, Tsuchihashi Z, Ruddy DA, Basava A, DormishianF, Domingo R Jr, Ellis MC, FullanA, HintonLM, Jones NL, Kimmel BE, Kronmal GS, Lauer P, Lee VK, Loeb DB, Mapa FA, McClelland E, Meyer NC, Mintier GA, Moeller N, Moore T, Morikang E, Wolff RK, et al. A novel MHC class I-like gene is mutated inpatients with hereditary haemochromatosis. Nat Genet 1996;13: Murphy S, CurranMD, McDougall N, Callender ME, O BrienCJ, MiddletonD. High incidence of the Cys 282 Tyr mutationinthe HFE gene in the Irish population implications for haemochromatosis. Tissue Antigens 1998;52: Beutler E, Gelbart T, West C, Lee P, Adams M, Blackstone R, Pockros P, Kosty M, Venditti CP, Phatak PD, Seese NK, Chorney KA, TenElshof AE, Gerhard GS, Chorney M. Mutationanalysis in hereditary hemochromatosis. Blood Cells Mol Dis 1996;23: BartonJC, Shih WW, Sawada-Hirai R, ActonRT, HarmonL, Rivers C, Rothenberg BE. Genetic and clinical description of hemochromatosis probands and heterozygotes: evidence that multiple genes linked to the major histocompatibility complex are responsible for hemochromatosis. Blood Cells Mol Dis 1997;23: Brandhagen DJ, Fairbanks VF, Baldus WP, et al. Prevalence of

6 August 2002 HFE MUTATIONS IN CELIAC DISEASE 449 HFE gene mutations in normal blood donors and in patients with ironoverload (abstr). Gastroenterology 1998;114:A Sham RL, Ou CY, Braggins C, Dunnigan K, Phatak PD. Correlation between genotype and phenotype in hereditary hemochromatosis. Blood Cells Mol Dis 1997;23: Borot N, Roth M, Malfroy L, Demangel C, Vinel JP, Pascal JP, Coppin H. Mutations in the MHC class I-like candidate gene for hemochromatosis in French patients. Immunogenetics 1997;45: Jouanolle AM, Gandon G, Jezequel P, Blayau M, Campion ML, Yaouanq J, Mosser J, Fergelot P, Chauvel B, Bouric P, Carn G, Andrieux N, Gicquel I, Le Gall JY, David V. Haemochromatosis and HLA-H. Nat Genet 1996;14: Martinez PA, Guillard A, Blanc F, Eiperier JM, Perney G, Biron C, JeanjeanPh, MasmejeanH, Michel H, Schved JF. Diagnosis of hemochromatosis gene mutations: the H63D mutation must be tested (abstr). Blood 1997;90(Suppl 1):7A. 20. Carella M, D Ambrosio L, Totaro A, Grifa A, Valentino MA, Piperno A, Girelli D, Roetto A, Franco B, Gasparini P, Camaschella C. Mutationanalysis of the HLA-H gene initalianhemochromatosis patients. Am J Hum Genet 1997;60: Gottschalk R, Seidl C, Loffler T, Seifried E, Kaltwasser JP, Hoelzer D. Analysis of HFE-codon 63/282 (H63D/282Y) gene variants ingermanhemochromatosis patients (abstr). Blood 1997; 90(Suppl 1):9A. 22. The UK Haemochromatosis Consortium. Worwood M, Shearman JD, Wallace DF, Dooley JS, Merryweather-Clarke AT, Pointon JJ, Rosenberg WMC, Bowen DJ, Burnett AK, Jackson HA, Lawless S, Raha-Chowdhury R, Partridge J, Williams R, Bomford A, Walker AP, Robson KJH. A simple genetic test identifies 90% of UK patients with haemochromatosis. Gut 1997;41: Fairbanks VF, Brandhagen DJ, Thibodeau SN, Snow K, Wollan PC. H63D is anhaemochromatosis associated allele. Gut 1998;43: Cluysenaer OJJ, vantongerenjhm. Malabsorptionincoeliac sprue. The Hague: Martinus Nijhoff, Turcu A, Lévêque L, Bielefeld P, Besancenot J-F, Hillon P. Adult celiac disease and hemochromatosis. Am J Gastroenterol 2000; 95: Heneghan MA, Feeley KM, Stevens FM, Little MPG, McCarthy CF. Precipitationof ironoverload and hereditary hemochromatosis after successful treatment of celiac disease. Am J Gastroenterol 2000;95: Miller S, Dykes D, Polesky H. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16: Mullighan CG, Bunce M, Fanning GC, Marshall SE, Welsh KI. A rapid method of haplotyping HFE mutations and linkage disequilibrium ina Caucasoid population. Gut 1998;42: Bunce M, O Neill CM, Barnardo MCNM, Krausa P, Browning MJ, Morris PJ, Welsh KI. Phototyping: comprehensive DNA typing for HLA-A,B,C,DRB1, DRB3, DRB4, DRB5 and DQB1 by PCR with 144 primer mixes utilizing sequence specific primers (PCR-SSP). Tissue Antigens 1995;46: Burt MJ, George PM, UptonJD, Collett JA, FramptonCMA, ChapmanTM, Walmsley TA, ChapmanBA. The significance of haemochromatosis gene mutations in the general population: implications for screening. Gut 1998;43: Received March 12, Accepted May 9, Address requests for reprints to: Jeffrey R. Butterworth, MRCP (UK), 1 Forty Acre Close, Loggerheads, Near Market Drayton, Staffordshire, TF9 4RQ England. j.r.butterworth@btinternet.com; fax: (44) Supported by a City Hospital, Birmingham, National Health Service Research and Development grant and by European Union Consortium grant QLRT J.R.B., D.H.A., B.T.C., and T.H.I. developed and designed the study, analyzed the data, and prepared the final manuscript. G.M.W. allowed the study of patients with celiac disease from Nuneaton, England. W.M.C.R., M.P., and W.M.H. provided HLA and HFE data from patients with celiac disease in Southampton and helped prepare the final document. S.J. and D.B. performed the HLA analysis on the patients with celiac disease from Birmingham and prepared the final document. M.H. performed the HLA analysis on the control population.