Liver fibrosis is a trait driven by genetic and environmental determinants [1
]. Since liver biopsy represents the “gold standard” method for quantifying liver fibrosis, the majority of previous studies aiming to define the genetic background of fibrosis were based on biopsy samples. Nevertheless, liver biopsy may be associated with complications [3
] and costs, which renders recruiting large cohorts for well-powered genetic analyses troublesome [4
]. Hence, in our latest candidate-gene study, we availed of transient elastography (TE) to phenotype a cohort of patients with chronic liver diseases [5
]. This study demonstrated that carriers of the common adiponutrin (PNPLA3
) variant, p.I148M, present with increased liver stiffness levels [5
]. The PNPLA3
polymorphism was previously identified in a genome-wide association study (GWAS), which showed that carriers of the risk allele are prone to hepatic fat accumulation and non-alcoholic fatty liver disease (NAFLD) [6
]. The association was then replicated in additional NAFLD cohorts [7
] and extended to alcoholic liver disease [13
], as well as severe liver phenotypes resulting in hepatic fibrosis and cirrhosis [5
]. Finally, cirrhotic carriers of the PNPLA3
variant are at increased risk of hepatocellular carcinoma (HCC) [18
], and in patients with HCC, this variant is associated with poor prognosis [22
]. The above results, substantiated in the latest meta-analysis [23
], establish the p.I148M adiponutrin variant as a common determinant of chronic liver injury leading to progressive fibrosis [24
]. The recent study by Kumari et al.
] demonstrated that PNPLA3
p.I148M is a “gain-of-function” variant, which results in higher lysophosphatidic acid acyltransferase activity and increased hepatic diacylglycerol synthesis. Moreover, further genetic studies associated the variant with some metabolic traits, in particular related to lipid metabolism [26
The above studies focused, however, on investigating primarily the PNPLA3
locus. Indeed, the analysis of selected single polymorphisms within candidate genes can reveal variants associated with common traits; nevertheless, the investigation of genetic susceptibility within specific pathways regulating the expression of these genes might provide further insights into disease predisposition. As highlighted lately, such regulatory pathways could play a pivotal role in the development of organ-specific fibrosis [30
]. Of note, Huang et al.
] described a pathway of adiponutrin expression in mouse liver that is induced by carbohydrates. As shown in their study [31
], the pathway includes the activation of nuclear receptors, namely, liver X receptor (LXR) and retinoid X receptor (RXR), which, in turn, increases hepatic expression of sterol regulatory element binding protein (SREBP1c). Upon ingestion of carbohydrates, SREBP1c induces the transcription of PNPLA3
and increases the synthesis of fatty acids (FA) in the liver [31
]. This observation in mice was substantiated by the analysis of human hepatocyte cell lines by Dubuquoy et al.
], who showed that SREBP1c is the master regulator of hepatic PNPLA3
expression in humans, as well. In short, SREBPs are transcription factors serving as central regulators of lipid and cholesterol metabolism [33
]. The major SREBP isoforms, SREBP1a, SREBP1c and SREBP2, are encoded by two different genes [37
]. Interestingly, previous studies demonstrated that two SREBP1c
variants, rs2297508 and rs11868035, affect glucose and lipid metabolism in humans [38
Hence, in the current study, we assess the influence of these common genetic variants within the SREBP1c gene on liver stiffness in a cohort of patients with chronic liver diseases.
Variable progression of liver fibrosis in individuals with comparable environmental risk profiles underscores the importance of genetic susceptibility in hepatic fibrogenesis [24
]. To date, the p.I148M adiponutrin variant represents the only well-established proinflammatory and profibrogenic polymorphism in patients with chronic liver diseases. Here, in contrast to previous studies that investigated solely the role of the adiponutrin risk variant in liver fibrosis, we focused on the key regulator of hepatic PNPLA3
expression. Using a candidate-gene approach, we showed that the common intronic SREBP1c
variant, rs11868035, localized between exons 18c and 19c, is associated with liver stiffness, especially in patients presenting with low TE values. Moreover, by introducing pathway analysis, we were able to investigate a combined effect of two SREBP1c
polymorphisms, underscoring the effects of this specific pathway on liver stiffness.
Genetic analysis of several variants within the pathway regulating hepatic adiponutrin expression represents the major novelty of our study. This design follows the hypothesis stating that if a gene is involved in a specific process or disease process, its direct interactors might also be suspected to play a role in the same process (“disease module”) [44
]. Pathway analysis is a novel approach, which is complementary to separate analyses of single genes, and tests if a given set of candidate genes within a metabolic pathway affects the occurrence of the disease. Given these assumptions, in the current analysis, we combined the genotyping results of the PNPLA3
variant from our previous study [5
] with the frequencies of two variants of, which is known to regulate the expression of PNPLA3
. Based on the key regulatory function of SREBP1c [31
], we chose two candidate variants within this locus and tested them as possible determinants of liver stiffness. Interestingly, the rs11868035 SREBP1c
polymorphism does not influence the structure of the protein, due to its intronic localization, but Liu et al.
] reported that it might affect mRNA turnover. Here, further functional studies should provide additional insights into the role of this variant in the progression of chronic liver diseases. On the other hand, others showed an involvement of SREBP1c in liver diseases, most of all, hepatic steatosis [46
], and pointed out that activation of SREBP might be critical for the development of fatty liver [49
]. Hence, one can hypothesize that carriers of the SREBP1c
risk variant could be prone to increased fat accumulation, due to a dysfunction of the protein.
Our analysis was performed in a cohort of patients phenotyped by TE. This approach was validated in our previous study, which demonstrated that the PNPLA3
variant is associated with increased liver stiffness [5
]. The application of TE in genetic studies provides the unique opportunity to analyze the relationship between genetic variation and a wide range of liver fibrosis phenotypes (i.e
., TE levels from 2.0 to 75.0 kPa). In the current study, this approach allowed us to identify a profibrogenic polymorphism that might be associated with early stages of liver fibrogenesis. The uni- and multi-variate analyses, including other potentially profibrogenic factors, further underscored this variant as an independent determinant of liver stiffness. Conversely to the SREBP1c
variant, the PNPLA3
p.I148M SNP is associated with more advanced stages of liver fibrosis and increases the risk of cirrhosis [5
]. Especially in the setting of alcoholic liver disease or chronic hepatitis C virus infection, carriers of the p.148M variant are at risk of developing fibrosis, cirrhosis and HCC [13
]. Of note, the TE ranges associated with the SREBP1c
variants do not overlap (see Tables 3
and Figure 3
). Since the PNPLA3
SNP has been previously associated with the development of HCC [18
], further studies in patients with HCC are needed to assess whether an association between the SREBP1c
variant and cancer susceptibility exists.
After stratification of our cohort, the association between the SREBP1c
variant and liver stiffness was replicated in the sub-cohort of patients with viral hepatitis, but not in patients with non-viral liver disease. This observation could be related to the relatively low number of patients in the latter group and/or a lower number of individuals with non-viral diseases in the cirrhotic group; the finding could also indicate that distinct mechanisms are involved in fibrogenesis in the setting of viral and non-viral liver diseases. We did not detect a significant association between the SREBP1c
polymorphism and histopathological fibrosis stages in the 229 individuals scheduled for this procedure, but a trend (p
= 0.052) for different genotype distributions between F0 and F1. This lack of association might be due to the lower number of patients with histopathological assessment; furthermore, most of these patients suffered from advanced liver fibrosis [5
], whereas the SREBP1c
variant seems to be associated with lower TE results. Due to the cross-sectional design of this study, only a single TE result is available for each patient. Hence, additional studies are warranted to prospectively assess the SREBP1c
variant and the association during disease progression. Moreover, non-invasive methods quantifying hepatic fat contents might provide additional functional insights.