Common variant p.D19 H of the hepatobiliary sterol transporter ABCG5/8 increases the gallstone risk and affects cholesterol homeostasis in children

Introduction:

Gallstones are believed to be caused via different pathomechanisms in children and in adults. In children black pigment stones are common, whereas in adults stones consist mainly of cholesterol. Serum concentrations of cholesterol precursors and plant sterols represent valid surrogate markers for cholesterol biosynthesis and intestinal absorption, respectively. Genetic studies demonstrated that “adult” gallstones are more frequent in carriers of the p.D19 H variant in the hepatobiliary cholesterol transporter ABCG5/8. The Gilbert polymorphism of the UGT1A1gene further increases this risk. We now investigated the role of the ABCG8 and UGT1A1*28 variants in the development of gallstones in paediatric patients.

Patients and methods:

In total, 214 children with gallstone disease (age at diagnosis 1 month – 17 years, 107 boys) were recruited. Children with history of total parenteral nutrition (TPN) were excluded. Symptomatic stones were present in 138 patients, 47 underwent cholecystectomy, and ERCP was required in 10 cases. The control cohort was composed of 172 adults (age 40 – 92 years, 70 males) with gallstones excluded by abdominal ultrasound. The ABCG8 p.D19 H and UGT1A1*28variants were genotyped using TaqMan assays and Sanger sequencing, respectively. Serum sterol levels were measured using gas chromatography/mass spectrometry (GC/MS) in 52 children.

Results:

The ABCG8risk allele was more frequent in children with gallstones (14.9%, 2 homozygotes and 30 heterozygotes) than in controls (7.5%, 1 homozygote, 12 heterozygotes). The lithogenic genotype was associated with an increased risk of stones (common OR = 1.82, P = 0.03). Of note, even carriers of one lithogenic allele demonstrated an increased gallstone prevalence (OR = 2.18, 95%CI 1.08 – 4.40, P = 0.02). The population attributable risk was 4% of the total gallstone risk in children. Patients carrying the lithogenic allele presented with significantly lower concentrations of the natural phytosterol sitosterol (P = 0.045) and serum phytostanols, i.e. campestanol (P = 0.028) and sitostanol (0.029). In line with these results, the ABCG5/8 p.D19 H variant was associated with decreased ratios of phytosterols to cholesterol precursors (sitosterol:desmosterol, P = 0.008; campesterol:desmosterol, P = 0.013). The UGT1A1*28variant was neither associated with increased prevalence of gallstone disease in the entire cohort (P = 0.20) nor in separate analyses of males or females (both P > 0.05).

Discussion:

Thelithogenic ABCG5/8 variant p.D19 H affects cholesterol homeostasis and increases gallstone risk in children. Overall, cholesterol gallstones might be more frequent in paediatric patients than estimated. Our results point to the presence of a common lithogenic pathway in adults and in children, for which precise therapeutic strategies might be envisioned in the future.