See “Genetic Insight Into the Pathogenesis of Nonalcoholic Fatty Liver Disease” by Valenti et al on page 139.
Nonalcoholic fatty liver disease (NAFLD) is a common chronic liver disease that encompasses a spectrum of disease ranging from fatty infiltration of the liver (steatosis) to histologic evidence of inflammation (nonalcoholic steatohepatitis) to fibrosis or cirrhosis, without a history of excessive alcohol ingestion (1). In an individual with NAFLD, excess hepatic fat is associated with an increased risk of developing diabetes, hypertension, cardiovascular events, abnormal resting electrocardiography, and endothelial dysfunction. These findings have been corroborated in studies in teenagers as well as in adults (2).
NAFLD was previously confined to residents of industrialized Western countries (3); however, because of the rapid adoption of the Westernized lifestyle, such as high-fat and high-energy diets and less physical activity (4), even in emerging nations, increasing sections of the Asian population have been at risk for developing NAFLD during the last several decades (5). An epidemiological study indicated that the prevalence of nonalcoholic fatty liver in the general population of Shanghai is 15.35% (6).
The genetic underpinning of NAFLD has been supported by familial aggregation studies (7), heritability studies (8), candidate gene studies (9), genome-wide scans (10), and gene expression studies (11). Some genes were reported to be associated with NAFLD, but most of them have not been replicated in independent samples, so that the susceptible genes of NAFLD need further studies (12).
Because obesity is a risk factor for NAFLD (13), the common variants of obesity-susceptible genes may be associated with NAFLD. The rs9939609 and rs9930506 polymorphisms in fat mass and obesity-associated gene (FTO) were most frequently reported to be associated with obesity and body mass index (BMI) both in adults and children (14–17). Several genome-wide association studies (GWAS) found that rs17782313 and rs12970134 in melanocortin-4 receptor (MC4R) gene region were obesity-susceptible variants (18). Herbert et al (19) reported that a common variant (rs7566605) located 10 kb upstream of insulin-induced gene 2 (INSIG2) was associated with obesity. A meta-analysis supported its association with an extreme degree of obesity (20). Therefore, we performed the association study between the single-nucleotide polymorphisms (SNPs) of FTO (rs9939609, rs9930506, rs4783819), MC4R (rs12970134, rs17782313), INSIG2 (rs7566605, rs13428113, rs9308762), and NAFLD or ALT level, which aimed to identify whether the SNPs were BMI-independent genetic predictors of NAFLD and ALT elevation.
Subjects of the present study were selected from 1093 individuals who were 7 to 18 years old and participated in the Comprehensive Prevention Project for Overweight and Obese Adolescents in Beijing, China. They were recruited from 3 middle schools and 2 elementary schools of the Haidian District of Beijing. The ascertainment strategy for the study population has been described in detail previously (21). By asking medical history, we selected the subjects without any of the following conditions: alcohol consumption, a history of diseases or drugs (including herbal medicines) causing liver disease, common (HBV, HCV) or less common (autoimmune, Wilson disease, α1-antitrypsin deficiency) liver diseases, hepatic malignancies, infections, biliary tract disease, and any cardiovascular and metabolic diseases. Finally, 1027 children having liver ultrasound examination and blood samples were included in the study.
The subjects were classified into the NAFLD group and the non-NAFLD group, according to the diagnostic criteria of NAFLD (22). The NAFLD diagnosis is based on the result of a questionnaire about alcohol consumption and image of abdominal ultrasound examination. NAFLD can be defined by the presence of at least 2 of 3 abnormal findings on abdominal ultrasonography: diffusely increased echogenicity (“bright”) liver with liver echogenicity greater than kidney, with vascular blurring, and deep attenuation of ultrasound signal (22).
All of the participants gave their written informed consent. The study was approved by the ethics committee of Peking University Health Science Center.
Measurement of height and weight was performed according to the standard protocols. Fasting venous blood samples were taken for detection of total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), fasting glucose, and alanine transaminase (ALT) by using a biochemical autoanalyzer (Hitachi 7060, Tokyo, Japan). Liver ultrasound examination was performed by 1 senior doctor with an ultrasound system (SIEMENS Soneline G50, Berlin, Germany).
Genomic DNAs of subjects were extracted from blood leukocytes by the phenol-chloroform extraction method. Polymerase chain reaction with subsequent restriction fragment length polymorphism assay was performed for genotyping rs9930506, rs7566605, rs13428113, and rs9308762 polymorphisms. Tetra-primer amplification refractory mutation system analysis (tetra-primer ARMS-PCR) (23) was performed for genotyping the rs9939609, rs4783819, rs17782313, and rs12970134 polymorphisms. All of the primers and enzymes for genotyping the above variants are available on request.
All of the genotypes were determined by the corresponding patterns resolved by electrophoresis on 2.5% agarose gels stained with ethidium bromide. Reference individuals of genotypes identified by sequencing were included in each run. For validity of the genotypes, allele assignments were made by at least 2 experienced individuals independently. Discrepancies were solved unambiguously either by reaching consensus or by repeating. We genotyped 5% of samples twice for quality control and obtained a 100% genotype concordance rate.
The genotype data of the non-NAFLD group was tested for deviation from Hardy-Weinberg equilibrium. Differences in age, BMI, and ALT between the NAFLD group and the non-NAFLD group were tested with t test. χ 2 test was applied to compare the sex difference between the 2 groups. Logistic regression analysis adjusted for age, age-square, and sex was used to calculate the odds ratios of the variants for NAFLD. The relations between the variants and ALT level were tested by using the linear regression analysis with age, age-square, and sex as covariates. We analyzed the associations among 3 variants (rs7566605, rs13428113, and rs9308762) and NAFLD under the recessive model because we found it was more appropriate in the population, whereas other variants were analyzed under the additive model based on recent publications (24). A 2-sided P value <0.05 was considered nominally significant. Adjustment was made for multiple testing using Bonferroni correction, considered 0.00625 (0.05 divided by 8) as statistically significant. All of the statistical analyses were performed using SPSS 18.0 software (SPSS Inc, Chicago, IL).
The general characteristics of NAFLD and non-NAFLD groups were shown in Table 1. The NAFLD group consisted of 162 children (47 girls, mean age 11.81 ± 2.20 years, mean BMI 26.75 ± 3.85 kg/m2) and the non-NAFLD group consisted of 865 children (406 girls, mean age 11.44 ± 2.99 years, mean BMI 20.72 ± 3.61 kg/m2). Sex differed in 2 groups (P < 0.001), whereas the age distribution in the 2 groups was not significantly different (P = 0.060). The mean BMI and ALT were significantly higher in the NAFLD group than in the non-NAFLD group (all P values <0.001).
The genotypic distribution of the variants was in Hardy-Weinberg equilibrium in the non-NAFLD group (all P values >0.05). As shown in Table 2, the genotype distribution of FTO rs9939609 was significantly different between the NAFLD group and the non-NAFLD group under the additive model with nominally statistical significance, that is, each additional A-allele increased NAFLD risk by 43% (95% confidence interval [CI] of odds ratio 1.04–1.97; P = 0.029). The association did not exist after Bonferroni correction. After being further adjusted for BMI, the association was not statistically significant (P = 0.268). No significant difference was found between other variants and NAFLD.
Using linear regression adjusted for age, age-square, and sex, we detected significant association between 2 variants (MC4R rs12970134 and INSIG2 rs9308762) and ALT level (Table 3; the frequency distribution of all INSIG2 SNPs can be viewed at http://links.lww.com/MPG/A253). For MC4R rs12970134, each A-allele increased ALT level by 1.87 IU/L (95% CI 0.16–3.57; P = 0.032). For INSIG2 rs9308762, the homozygotes of C-allele had higher ALT level than T-allele carriers (β = 3.19; 95% CI 0.87–5.51; P = 0.007). After being further adjusted for BMI, the association between MC4R rs12970134 and ALT level did not exist (P = 0.064), but the association between INSIG2 rs9308762 and ALT level remained significant (P = 0.003). After Bonferroni correction, the BMI-independent association of INSIG2 rs9308762 with ALT level was still significant, but the association between MC4R rs12970134 and ALT level did not exist. Additionally, we analyzed association of the INSIG2 SNP (rs9308762) with the indicators of serum lipids (TC, TG, HDL-C, LDL-C), and fasting glucose. None of the INSIG2 variants was significantly associated with serum lipids or glucose (all P > 0.05, data not shown).
For the 8 gene variants in our study, there has been only 1 published study, which analyzed the relations between the MC4R rs17782313 and liver fat (25). This is the first study on the relations between the other 7 variants (rs9939609, rs9930506, and rs4783819 of FTO; rs12970134 of MC4R; rs7566605, rs13428113, and rs9308762 of INSIG2) and NAFLD.
We found a significant association of the A-allele of FTO rs9939609 with NAFLD; however, the association did not exist after being further adjusted for BMI, which indicated FTO rs9939609 A-allele predisposed to NAFLD through an effect on BMI.
Obesity is strongly associated with NAFLD. In the Dionysos study, fatty liver diagnosed by ultrasound examination was found in 10% to 15% of normal-weight individuals and in up to 76% of obese subjects not drinking alcohol in toxic amounts (26). Compared with healthy controls, risk for steatosis was 4.6-fold increased in obese subjects (13). The mechanisms showing how obesity increases the risk of NAFLD are not clear. Some studies suggested that the major pathways included increased dietary intake of carbohydrates and triglycerides, lipolysis from fat reservoirs because of insulin resistance, and prevention of triglyceride removal from the liver (26).
Because many studies reported that the ALT level of patients with NAFLD was significantly higher than that in non-NAFLD individuals (22,27), increased ALT activities were the most common abnormality in patients with nonalcoholic steatohepatitis, that is, the second stage of NAFLD (28). We detected the relations between 8 variants and ALT level. We found the significant association between the A-allele of MC4R rs12970134 and ALT level, with the effect size of 1.87 IU/L per A-allele. After being further adjusted for BMI, the association did not exist. Furthermore, the association between INSIG2 rs9308762 and ALT level was statistically significant and sustained even after adjustment for BMI. The results suggested that rs12970134 in MC4R was associated with ALT level through an effect on BMI, whereas the association between rs9308762 in INSIG2 and ALT level was independent of BMI. The BMI-independent association of INSIG2 rs9308762 with ALT level was still significant after Bonferroni correction.
In our study, the association between the INSIG2 rs9308762 and ALT level was statistically significant, but the same variant had no relations to NAFLD for several reasons. First, the NAFLD was a categorical variable and analyzed by using logistic regression, whereas ALT level was a continuous numeric variable and analyzed by using linear regression. These 2 regression models were statistically different. One factor may be significantly associated with a continuous indicator of a specific disease but not significantly associated with a categorical outcome of the same disease, such as some gene SNPs associated with BMI but not with obesity. Second, Szalman et al (29) demonstrated that the ALT level has been elevated in the group with grade 3 histological activity and there was a positive, statistically significant correlation (r = 0.35, P = 0.02) between ALT values and grade of histological activity. It indicated that ALT may serve as a parameter of the progress or severity of NAFLD. In our study, we diagnosed NAFLD based on ultrasonography. The limitation is lack of evaluation of liver damage by histology, thereby precluding the possibility of assessing the progress of NAFLD. The authors found that the SNP rs9308762 in INSIG2 was associated with ALT levels but not associated with NAFLD diagnosed by ultrasonography, indicating the SNP as a new potential modifier of the risk of progressive NAFLD (represented by ALT level) independent of BMI.
INSIG is located on chromosome 2q14.1, including INSIG1 and INSIG2. The INSIG2 protein can prevent the proteolytic processing of sterol regulatory element–binding proteins by Golgi enzymes and thereby block cholesterol synthesis (30), which plays an important role in lipid and glucose metabolization, but the changes of serum lipid or insulin resistance are not certain. It is well known that a close link exists between NAFLD and dyslipidemia, a constellation of abnormalities in lipoproteins, but several key questions remain to be answered. How does lipoprotein metabolism evolve as the progression of NAFLD? Are there changes in lipoprotein metabolism that one can identify in the serum to predict disease progression? (31) Cali et al (32) determined fasting lipoprotein subclasses using nuclear magnetic resonance spectroscopy. Although there was no difference in the level of total LDL between adolescents with and without hepatosteatosis, they found the presence of fatty liver was associated with a pronounced dyslipidemic profile characterized by large very-low-density lipoprotein cholesterol, small LDL, and decreased large HDL concentrations. In the present study, we did not identify the association of the INSIG2 SNP with the common clinical indicators of lipids (TC, TG, LDL-C, HDL-C, and glucose). There are 2 possible reasons. First, the effects of the INSIG2 SNP on serum lipids or glucose may be less than that on ALT level. Because the sample size of the present study is limited, we could not detect the effects. Second, the SNPs may have effects on the lipoprotein subclasses or insulin, which play roles in liver damage, but we did not measure the indicators. The possible mechanisms underpinning this genetic association between the SNP and ALT level await further studies.
The present study has several limitations. The first is the limited sample size and statistical power. Because no previous study reported the effects of these 8 gene variants on NAFLD, we could not estimate the statistical power. Second, the case-control study design did not permit us to make conclusions about causality. Third, the diagnosis of NAFLD was based on questionnaire and abdominal ultrasound examination, but not histologic examination; however, our study results provided evidence for identifying the genetic factors of NAFLD because we used the diagnostic criteria of NAFLD, which has been used worldwide (22).
In conclusion, we found the FTO rs9939609 A-allele increased the risk of NAFLD and the rs12970134 in MC4R was associated with ALT level through an effect on BMI. The association between INSIG2 rs9308762 and ALT level was independent of BMI. The results provided evidence for identifying genetic factors of NAFLD and may be useful for risk assessment and personalized medicine of NAFLD. Further large-scale studies in different ethnic populations are needed to validate the association between these gene variants and NAFLD.
1. Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease
2. Sanyal AJ. NASH: a global health problem. Hepatol Res
3. Angulo P. Nonalcoholic fatty liver disease
. N Engl J Med
4. Fan JG, Farrell GC. Epidemiology of non-alcoholic fatty liver disease in China. J Hepatol
5. Amarapurkar DN, Hashimoto E, Lesmana LA, et al. How common is non-alcoholic fatty liver disease in the Asia-Pacific region and are there local differences? J Gastroenterol Hepatol
6. Fan JG, Zhu J, Li XJ, et al. Prevalence of and risk factors for fatty liver in a general population of Shanghai, China. J Hepatol
7. Willner IR, Waters B, Patil SR, et al. Ninety patients with nonalcoholic steatohepatitis: insulin resistance, familial tendency, and severity of disease. Am J Gastroenterol
8. Schwimmer JB, Celedon MA, Lavine JE, et al. Heritability of nonalcoholic fatty liver disease
9. Daly AK, Ballestri S, Carulli L, et al. Genetic determinants of susceptibility and severity in nonalcoholic fatty liver disease
. Expert Rev Gastroenterol Hepatol
10. Chalasani N, Guo X, Loomba R, et al. Genome-wide association study identifies variants associated with histologic features of nonalcoholic fatty liver disease
2010; 139: 1567–1576 e1-6.
11. Miquilena-Colina ME, Lima-Cabello E, Sanchez-Campos S, et al. Hepatic fatty acid translocase CD36 upregulation is associated with insulin resistance, hyperinsulinaemia and increased steatosis in non-alcoholic steatohepatitis and chronic hepatitis C. Gut
12. Hernaez R. Genetic factors associated with the presence and progression of nonalcoholic fatty liver disease
: a narrative review. Gastroenterol Hepatol
13. Wree A, Kahraman A, Gerken G, et al. Obesity affects the liver–the link between adipocytes and hepatocytes. Digestion
14. Frayling TM, Timpson NJ, Weedon MN, et al. A common variant in the FTO gene
is associated with body mass index and predisposes to childhood and adult obesity. Science
15. Scuteri A, Sanna S, Chen WM, et al. Genome-wide association scan shows genetic variants in the FTO gene
are associated with obesity-related traits. PLoS Genet
16. Kilpelainen TO, Qi L, Brage S, et al. Physical activity attenuates the influence of FTO variants on obesity risk: a meta-analysis of 218,166 adults and 19,268 children. PLoS Med
17. Wei HY, Zhang YD, Tan LN. Meta-analysis of fat mass and obesity associated gene
rs9939609 polymorphism and childhood overweight and obesity. Chin J Clin
18. Loos RJF, Lindgren CM, Li S, et al. Common variants near MC4R are associated with fat mass, weight and risk of obesity. Nat Genet
19. Herbert A, Gerry NP, McQueen MB, et al. A common genetic variant is associated with adult and childhood obesity. Science
20. Heid IM, Huth C, Loos RJF, et al. Meta-analysis of the INSIG2 association with obesity including 74,345 individuals: does heterogeneity of estimates relate to study design? PLoS Genet
21. Wang HJ, Zhang H, Zhang SW, et al. Association of the common genetic variant upstream of INSIG2 gene
with obesity related phenotypes in Chinese children and adolescents. Biomed Environ Sci
22. Farrell GC, Chitturi S, Lau GKK, et al. Guidelines for the assessment and management of non-alcoholic fatty liver disease in the Asia–Pacific region: executive summary. J Gastroenterol Hepatol
23. Ye S, Dhillon S, Ke X, et al. An efficient procedure for genotyping single nucleotide polymorphisms. Nucleic Acids Res
24. Chambers JC, Elliott P, Zabaneh D, et al. Common genetic variation near MC4R is associated with waist circumference and insulin resistance. Nat Genet
25. Haupt A, Thamer C, Heni M, et al. Impact of variation near MC4R on whole-body fat distribution, liver fat, and weight loss. Obesity
26. Bedogni G, Miglioli L, Masutti F, et al. Prevalence of and risk factors for nonalcoholic fatty liver disease
: the Dionysos nutrition and liver study. Hepatology
27. Clark JM, Diehl AM. Nonalcoholic fatty liver disease
28. Diehl A, Goodman Z, Ishak K. Alcohol-like liver disease in nonalcoholics. A clinical and histologic comparison with alcohol-induced liver injury. Gastroenterology
29. Szalman K, Bancu L, Sin A. Correlations between anthropometric and serologic elements of metabolic syndrome and histopathologic features of nonalcoholic fatty liver disease
. Rom J Morphol Embryol
30. Yabe D, Brown MS, Goldstein JL. Insig-2, a second endoplasmic reticulum protein that binds SCAP and blocks export of sterol regulatory element-binding proteins. Proc Natl Acad Sci U S A
31. Jiang ZG, Robson SC, Yao Z. Lipoprotein metabolism in nonalcoholic fatty liver disease
. J Biomed Res
32. Cali AM, Zern TL, Taksali SE, et al. Intrahepatic fat accumulation and alterations in lipoprotein composition in obese adolescents: a perfect proatherogenic state. Diabetes Care