Editor—Primary biliary cirrhosis (PBC) is a chronic, slowly progressive cholestatic liver disease believed to result from autoimmune mechanisms. The initiation of the disease is likely to be multifactorial with genetic, infectious, and environmental factors contributing. A familial predisposition to PBC has been reported, but studies to investigate an association between PBC and polymorphisms at a number of genetic loci have not been conclusive. The aetiology of the disease remains unknown but it has been suggested that R forms of E coli 1 andMycobacterium gordonae 2 3 may play a potentially pathogenic role in PBC, though this has not been established. A common characteristic feature of PBC is the presence of granulomas and it is interesting to note that these tend to disappear as the lesions progress and fibrosis and cholestasis appear, that is, secondary effects of tissue damage.

NRAMP1 (natural resistance associated macrophage protein 1) was isolated as the human homologue of the mousenramp1 gene (previously designatedIty/Lsh/Bcg) which, when mutated, is responsible for susceptibility to a number of macrophage trophic intracellular pathogens including Mycobacterium bovis, Salmonella typhimurium, andLeishmania donovani.4Expression of the gene is restricted to cells of the mononuclear phagocytic system (macrophages and granulocytes)5 and it plays an important role in the activation of macrophages and innate immunity. When nramp1 is mutated, mice fail to control pathogen growth in the early stages of infection. Sequence analysis of nramp1 and recent functional studies suggest that the gene encodes a multispanning transmembrane transporter protein6 with specificity for divalent metal cations,7 but its physiological role in relation to macrophage function is still poorly understood. In the human gene at least 10 polymorphic sites have been described,8-10 one of which spans a microsatellite repeat region in the 5′ untranscribed promoter region of the gene. This is a functional polymorphism affecting levels of NRAMP1expression.11 The alleles at this site have been inconsistently reported; Blackwell et al 9 initially reported identification of four alleles where alleles 1, 2, 3, and 4 had 11, 10, nine, and four repeats of the final dinucleotide repeat in the sequence t(gt)₅ac(gt)₅ac(gt)_ng. In subsequent reports they have designated the sequence t(gt)₅ac(gt)₁₀g as allele 411-13and have most recently reported allele 4 to have the sequence t(gt)₅ac(gt)₉g14. Liuet al 8 identified three alleles at this site which they allocated alleles 1, 2, and 3 with 2 bp increments of increasing size and allele 1 having the sequence t(gt)₇ac(gt)₅ac(gt)₉g.8We adopted and added to the allele naming system used by Blackwellet al 9 11-14 as we have shown evidence from restriction enzyme digestion and sequencing that three of the alleles we have identified are consistent with their reported alleles 1, 2, and 3.

Associations with NRAMP1 and susceptibility to a number of diseases including Crohn's disease,15tuberculosis,16 leprosy,17 and rheumatoid arthritis (RA)12 13 have been reported. The reportedNRAMP1 and RA association is with the allele driving highest levels of NRAMP1 expression (allele 3) of the microsatellite repeat region polymorphism discussed above. Allele 3 was found to be transmitted to RA affected children in preference to allele 2 in a study using identity by descent sib pair analysis.12 13 We investigated this polymorphic site in patients with PBC.

Using restriction fragment length polymorphism (RFLP) analysis we genotyped 46 PBC patients, 76 alcoholic liver disease (ALD) patients, 39 hepatitis C patients, and 78 normal, healthy, non-cirrhotic subjects for the microsatellite repeat polymorphism in the promoter region of the NRAMP1 gene. PCR products of approximately 194 bp, generated using NRAMPFpol GGACATGAAGACTCGCATTAGG (59-70 bp Genbank Accession number X82016) and NRAMPRpol TTAGCTCT- GATTTCAGATGCTTCC (240-217 bp Genbank Accession number X82016) PCR primers, were digested separately withRsaI and MnlI, resolved on 6% denaturing polyacrylamide gel by electrophoresis, and silver stained.18 We found three alleles as previously reported,9 11-14 but in addition two new alleles which we named allele 5 and allele 6.

The PCR products from cases with these novel genotypes were cloned into the vector pCR2.1 using Invitrogen TA Cloning System Version 2.0 and the plasmids used to transform competent XL-2 blue cells. The cloned DNA inserts from successful transformants showing the appropriate RFLP pattern were then sequenced. Allele 5 was found to have the sequence t(gt)₄ac(gt)₅ac(gt)₁₀g with a gt deletion in the first (gt)₅ dinucleotide repeat identified in the other alleles. Allele 6 had the sequence t(gt)₅ac(gt)₅ac(gt)₄at(gt)₄ggcaga(g)₇which had an extra guanidine inserted in a hexanucleotide g repeat 7 bp 3′ of the final gt of the polymorphic site in addition to the final polymorphic gt repeat being interrupted. Despite finding these novel alleles we have not identified a pattern representing any of the sequences published as allele 49 11-14 in any of the 246 cases analysed.

The sequence of the five alleles we have detected in our study (alleles 1, 2, 3, 5, and 6) and allele 4 reported by Blackwellet al 9 11-14 are shown in table 1. In the PBC population we genotyped, allele 5 was significantly more frequent (Fisher's exact test) in the PBC patients (8/53) we studied than in normal controls (3/78) (p<0.024), ALD (2/76) (p<0.012), or hepatitis C patients (0/39) (p<0.012) but was still uncommon.

The repetitive nature of the sequence we analysed requires a stringent, sensitive, and reproducible detection method, owing to the possibility of slippage or infidelity of the Taqpolymerase enzyme during PCR amplification. The method we adopted gave consistent results when we analysed DNA extracted from the same blood sample at different times, DNA extracted from blood taken from the same person at different times, and DNA extracted from blood and paraffin embedded material from the same person, and we are confident that our findings are genuine. A number of alleles at this site have previously been reported by sizing this polymorphic region as a single fragment. Our strategy used restriction enzyme digests designed to size both the whole polymorphic site as a single fragment (MnlI) and the size of fragments containing the 5′ dinucleotide and 3′ dinucleotide repeats of the region as distinct entities (RsaI) as shown in fig 1. It was the combined results of these digests which allowed detection of alleles 5 and 6 and confirmation of alleles 1, 2, and 3 reported by Blackwell et al. 9 11-14 It is important to note that had we been relying solely on results from a single analysis of the size of the microsatellite repeat as a whole, as previous reports have done, allele 5 would have been mistaken for allele 3, the allele which is at the highest frequency in all previously published reports.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Sequence of the promoter region of the human NRAMP1 (Accession number X82016, 59-240 bp) amplified by PCR to genotype the polymorphic microsatellite region shown in bold. The RsaI and MnlI restriction enzyme sites used to digest the amplified DNA are shown. The enzymes recognise the following sites and cut at the position shown by the vertical slash. n represents any base. MnlI∣(n)₆gagg ∣(n)₇ctcc RsaIgt∣ac. The sequence for allele 3 is used in this figure. Sequences of the other alleles identified in this study and those previously published are shown in table 1.

Our finding of two new alleles at this functional polymorphic site warrants further analysis in other populations and investigation into the effect of allele 5 on levels of NRAMP1expression. Though we may not have identified a biologically significant association with PBC and this site, it is interesting to consider whether previous studies of this polymorphic site may have contained cases with unrecognised allele 5.