Analysis of the promoter of human bilirubin UDP-glucuronosyltransferase gene (UGT1*1) in relevance to Gilbert's syndrome

Abstract

Bilirubin UDP-glucuronosyltransferase 1 (B-UGT₁) is an enzyme responsible for hepatic bilirubin glucuronidation and in Gilbert's syndrome, its activity is decreased to 30% of normal. About one third of the patients we analyzed with the syndrome had a homozygous TATA box mutation in UGT1*1 coding for B-UGT₁, i.e. (TA)₇TAA instead of (TA)₆TAA. Since the frequency was much higher than in the normal Japanese population (≈1%), the mutation was considered to be closely associated with the etiology of the syndrome. Since these patients did not have any structural mutations in UGT1*1, we suspected a transcriptional defect of the gene. We analyzed the promoter region (up to −3190) of UGT1*1 by transient transfection assay to identify transcriptionally regulatory sequences and found two elements; one was between −1362 and −1220 and the other between −113 and −70. The proximal one (PE) consisted of two elements, E-box (−104 to −95) and an HNF-1 site (−91 to −79). The two TATA boxes were compared in promoter activity, but no substantial difference was observed. Distal element (DE) and PE were then analyzed in these patients' DNA (n=6) by PCR and direct sequencing. The same homozygous mutation (C-1353A) was found in two of them, but the reduction in promoter activity was only 15%. These results suggest that the TATA box mutation itself is not the major cause of the syndrome and may be genetically linked to an, as yet, unidentified defect in the further upstream or in the intronic region of UGT1*1.

Introduction

Bilirubin UDP-glucuronosyltransferase 1 (B-UGT₁), is responsible for hepatic bilirubin glucuronidation [1]and its activity is reduced to 30% of normal in Gilbert's syndrome 2, 3. We have been analyzing UGT1*1, which encodes B-UGT₁, in the syndrome by PCR and direct sequencing 4, 5, 6, 7, 8. Five patients with the syndrome had a heterozygous missense mutation in UGT1*1 [4]. An in vitro expression experiment carried out on one of the patients who had a mutation of C696A (Pro229Gln) showed a severe decrease in B-UGT₁ activity (≈14% of normal) [5]. These results suggested that, at least in this case, the missense mutation is the cause of the disease.

On the other hand, Bosma et al. [9]and Monaghan et al. [10]reported that patients with Gilbert's syndrome have an extended TATA box, i.e. (TA)₇TAA instead of (TA)₆TAA. We also found that about one third of the patients we analyzed were homozygous for this mutation [8]: among 19 patients, six patients were homozygous and two were heterozygous for the mutated TATA box without any structural mutations, three had simultaneous mutations in the TATA box and in the coding region (Gly71Arg or Arg367Gly) and the other eight patients were either homozygous for Gly71Arg or heterozygous for Gly71Arg, Pro229Gln, Arg367Gly or Tyr486Asp. Since the (TA)₇TAA allele is rare in the Japanese population (10.5%, this study), such an accumulation of homozygous patients for the allele suggested a close association of the TATA box mutation with the etiology of the syndrome. As these patients did not have any mutations in the exons or in the exon/intron boundaries [8], we speculated that the TATA box mutation may affect transcriptional efficiency or it may be genetically linked to another abnormality present in the promoter of UGT1*1. In this study, we first identified transcriptionally regulatory regions of the gene, compared promoter activity between (TA)₆TAA and (TA)₇TAA and then analyzed the patient's DNAs with regard to the upstream regulatory regions.