Genetic Variants in the Promoter Region of the Macrophage Migration Inhibitory Factor are Associated with the Severity of Hepatitis C Virus-Induced Liver Fibrosis

Abstract

Two polymorphisms in the promoter region of macrophage migration inhibitory factor (MIF) - rs755622 and rs5844572 - exhibit prognostic relevance in inflammatory diseases. The aim of this study was to investigate a correlation between these MIF promoter polymorphisms and the severity of hepatitis C virus (HCV)-induced liver fibrosis. Our analysis included two independent patient cohorts with HCV-induced liver fibrosis (504 and 443 patients, respectively). The genotype of the single nucleotide polymorphism (SNP) -173 G/C and the repeat number of the microsatellite polymorphism -794 CATT_5-8 were determined in DNA samples and correlated with fibrosis severity. In the first cohort, homozygous carriers of the C allele in the rs755622 had lower fibrosis stages compared to heterozygous carriers or wild types (1.25 vs. 2.0 vs. 2.0; p = 0.03). Additionally, ≥7 microsatellite repeats were associated with lower fibrosis stages (<F2) (p = 0.04). Comparable tendencies were observed in the second independent cohort, where fibrosis was assessed using transient elastography. However, once cirrhosis had been established, the C/C genotype and higher microsatellite repeats correlated with impaired liver function and a higher prevalence of hepatocellular carcinoma. Our study demonstrates that specific MIF polymorphisms are associated with disease severity and complications of HCV-induced fibrosis in a stage- and context-dependent manner.

Keywords: HCV; biomarker; hepatocellular carcinoma; liver fibrosis; macrophage migration inhibitory factor; promoter polymorphisms.

Figure 1

The C/C genotype in the -173 G/C-SNP and higher microsatellite repeat numbers in the CATT_5–8 microsatellite significantly correlated with lower fibrosis stages in HCV-induced liver fibrosis. Only patients with fibrosis stage ≥F0.5 were included. (A) Correlation of the SNP genotype and fibrosis staging F1–F4 showed a significantly lower median fibrosis stage in patients with a C/C genotype compared to those with a C/G genotype (1.25 vs. 2.0) and G/G genotype (1.25 vs. 2.0, p = 0.030 in the Kruskal–Wallis test). (B) Histological grading did not significantly differ between the SNP-genotypes. (C) Proportions of patients with low or high fibrosis stages (<F2 vs. ≥F2) within the three SNP genotype subgroups. In the C/C subgroup, significantly more patients showed lower levels of fibrosis as Fisher´s exact test (*; p = 0.043) and linear-by-linear association test (**; p = 0.016) shows. (D) Histological grading did not significantly differ between the SNP-genotypes. (E) High repeat numbers on at least one allele of the microsatellite (“CATT 7/X; 8/X”) significantly correlated with lower fibrosis stages (p = 0.040). (F) There was no significant difference in the inflammatory activity of patients with low (“CATT 5/5, 5/6, 6/6”) to high repeat numbers. Data are expressed as the median ± range (A,B) or contingency (C,D–F) and were considered significant if p < 0.05 (*).

Figure 2

The C/C genotype of the -173 G/C SNP and higher microsatellite repeat numbers in the CATT_5–8 microsatellite correlated with lower fibrosis stages in HCV-induced liver fibrosis in the HCV 2018 validation cohort. (A) Correlation of the SNP genotype and transient elastography (TE) results showed a tendency toward lower mean fibrosis measurements in patients with the C/C genotype compared the G/G genotype (p = 0.40). (B) High repeat numbers on at least one allele of the microsatellite (“CATT 7/X; 8/X”) did not significantly correlate with lower liver stiffness measurements. (C) The haplotype analysis of the G/G SNP genotype and low repeat numbers in the microsatellite (CATT 5/5, 5/6, 6/6) showed higher mean TE values than the C/C SNP genotype and high microsatellite repeat numbers (CATT 7/X, 8/X; p = 0.015). (D,E) Grading of the inflammatory activity based on levels of alaninaminotransferase (ALT) did not differ significantly between the SNP or microsatellite genotypes (patients were subgrouped by the ALT cutoff of 0.85 µkat/L). Data are expressed as the mean fibrosis level (A,C,D) or contingency (B,E) and were considered significant if p < 0.05 (*).

Figure 3

The G/C and C/C genotypes in the -173 G/C SNP and high microsatellite repeat counts in the -794 CATT_5–8 are associated with worse liver function in patients with liver cirrhosis. Liver cirrhosis was diagnosed based on TE ≥ 12 kPa, if available, or morphological evaluation using a CAT scan. (A,B) Bilirubin and albumin serum levels did not differ between the three SNP genotype subgroups. (C) The proportion of patients with higher Child–Pugh stadium B and C was significantly higher within the patients´ subgroup with a G/C and C/C genotype according to the chi-square test (p < 0.0001). In the C/C genotype subgroup, 20% of patients have already progressed to Child–Pugh stadium C. (D,E) In the patients´ subgroup with higher microsatellite repeat numbers (CATT 7/X and 8/X), significantly higher percentages show bilirubin levels above and albumin levels below the reference cutoff respectively (cutoff for bilirubin ≥ 19 µmol/L; albumin < 35 g/L). (F) There was a tendency toward a higher Child–Pugh stadium within the CATT 7X; 8/X genotype subgroup. Data are expressed as the mean ± SEM (A,B) or contingency (C–F) and were considered significant, if p < 0.05 (*).

Figure 4

The C/C genotype in the -173 G/C SNP was associated with the prevalence of HCC: there is a strong tendency towards a higher HCC prevalence within the C/C genotype subgroup (p = 0.065). Data are expressed as contingencies.