SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Objective

Mechanisms explaining the relationship in non-alcoholic fatty liver disease (NAFLD), obesity, and insulin resistance are poorly understood. A genetic basis has been suggested. We studied the association between the genes patatin-like phospholipase domain-containing protein 3 (PNPLA3) and apolipoprotein C3 (APOC3) and metabolic and histological parameters of NAFLD in obese patients.

Design and Methods

Overweight and obese patients underwent a metabolic and liver assessment. If NAFLD was suspected, liver biopsy was proposed. APOC3 variant rs2854117 and PNPLA3 variant rs738409 were genotyped.

Results

Four hundred seventy patients were included (61.1% had liver biopsy). The percentage of patients with non-alcoholic steatohepatitis (NASH) was significantly different according to the PNPLA3 variant. After adjustment for age and body mass index, the PNPLA3 variant was associated with alanine aminotransferase (P < 0.001) and aspartate aminotransferase (P < 0.001). The PNPLA3 variant was associated with more severe features of steatohepatitis: steatosis (P < 0.001), lobular inflammation (P < 0.001), and ballooning (P = 0.002), but not with liver fibrosis, anthropometry, or insulin resistance. No significant difference in liver histology, anthropometric, or metabolic parameters was found between carriers and non-carriers of the APOC3 variant.

Conclusions

PNPLA3 polymorphism rs738409 was associated with NASH and the severity of necroinflammatory changes independently of metabolic factors. No association between APOC3 gene variant rs2854117 and histological or metabolic parameters of NAFLD was found.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Together with the obesity epidemic [1], incidences of non-alcoholic fatty liver disease (NAFLD) rise across countries. NAFLD is defined as an accumulation of fat, mainly triglycerides (TG), in hepatocytes exceeding 5-10% of the liver weight and this in the absence of significant alcohol consumption or any other cause of secondary liver steatosis or disease [2]. NAFLD comprises a spectrum of hepatic disorders ranging from simple steatosis, non-alcoholic steatohepatitis (NASH) to cirrhosis, and hepatocellular carcinoma [3]. NAFLD is affecting approximately one-third of the adult population in the Western world and relates with obesity and its metabolic consequences such as insulin resistance and dyslipidemia [4], but underlying mechanisms explaining this relationship are still poorly understood.

The genetic contribution to NAFLD has been largely ignored because alterations in lifestyle have been blamed as major culprit for the fatty liver epidemic. Most genetic studies involve evaluation of candidate genes that are chosen for their established association with insulin resistance, metabolic syndrome, adipocytokine response, and hepatic fibrogenesis [5]. The first genome-wide association study (GWAS) on NAFLD [6] identified patatin-like phospholipase domain containing protein 3 (PNPLA3), also known as adiponutrin. A PNPLA3 variant was initially associated with fatty liver, and further robustly replicated in independent candidate gene studies [7]. A variant in PNPLA3 (rs738409 encoding I148M) is associated with elevation of serum liver enzymes alanine aminotransferase (ALT) [6], aspartate aminotransferase (AST) [6, 8], or the AST/ALT ratio [9]. Several studies report a relationship between PNPLA3 and increased hepatic triglyceride content, measured by proton magnetic resonance spectroscopy (H-MRS) [6, 8, 10]. This association has also been confirmed in subjects with histologically characterized NAFLD [11, 12]. No association could be observed with anthropometric parameters such as body mass index (BMI) [12] or metabolic characteristics such as insulin resistance, glucose tolerance [6, 8, 10, 12], and plasma TG [12].

Another gene studied in association with NAFLD is apolipoprotein C3 (APOC3) [13]. APOC3, synthesized in liver and intestine [14], inhibits the catabolism of lipoproteins and delays their clearance from plasma [15]. APOC3 gene expression is increased by glucose stimulation, suggesting a role in the development of diabetic hypertriglyceridemia [16]. Petersen et al. [13] reported that two sequence variants (rs2854116 and rs2854117) in the promoter region of the gene encoding APOC3 were linked to hypertriglyceridemia, hepatic fat, and insulin resistance in healthy Asian men. This association between the APOC3 variants and NAFLD prevalence or severity has not been replicated in any study to date [17], so results are discordant.

In the present study, we investigated cross-sectionally the association of a PNPLA3 variant (rs738409) and an APOC3 variant (rs2854117) with liver histology and metabolic parameters of NAFLD in a large consecutively recruited cohort of Caucasian obese patients without a priori suspicion of liver disease. In most other studies patients are highly selected, as they are referred to specialized hepatology clinics because of elevated transaminases or, in case of bariatric surgery series, resulting in patients with relative high prevalence of more advanced disease.

Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Metabolic work-up

Patients visiting the obesity clinic of the Antwerp University Hospital (a tertiary referral facility) for a problem of overweight (BMI ≥ 25-29.9 kg/m2) or obesity (BMI ≥ 30 kg/m2) presenting at their own initiative or who were referred by their treating physician were consecutively recruited. Every patient underwent a standard metabolic work-up combined with a liver-specific program, both approved by the Ethics Committee of the Antwerp University Hospital and requiring written informed consent of the patient. The metabolic work-up includes a detailed questionnaire and a clinical examination with anthropometry. All measurements were performed in the morning, with patients in fasting conditions and undressed. Height was measured to the nearest 0.5 cm and body weight was measured with a digital scale to the nearest 0.2 kg. BMI was calculated as weight in kilograms over height in meters squared. Waist circumference was measured at the mid-level between the lower rib margin and the iliac crest. Hip circumference was measured at the level of the trochanter major. Waist to hip ratio (WHR) was calculated by dividing waist circumference by hip circumference. A blood analysis included blood cell count, coagulation tests, electrolytes and kidney function tests, lipid profile (total and high density lipoprotein cholesterol (HDL-C), and TG), liver tests [AST, ALT, lactate dehydrogenase (LDH), gamma glutamyl transpeptidase (GGT), alkaline phosphatase (ALP), total bilirubin and fractions], high-sensitive C-reactive protein (hs-CRP), creatinine kinase, total protein, protein electrophoresis, thyroid function, ferritin, vitamin B12 and folic acid. Low-density lipoprotein cholesterol (LDL-C) was calculated using the Friedewald formula [22]. A 3-h oral glucose tolerance test (75 g of glucose) including insulin and c-peptide analysis was also performed. Insulin resistance was estimated using homeostasis model assessment (HOMA) as described by Matthews et al. [23], and was calculated as [insulin (mU/l) × glucose (mmol/l)]/22.5, with 1 as reference value for normal insulin sensitivity. Area under the curve (AUC) was calculated for glucose, insulin, and C-peptide. Plasma glucose, total cholesterol, and TG were measured on Vitros 750 XRC (Ortho Clinical Diagnostics, Johnson & Johnson, UK). HDL-C was measured on Hitachi 912 (Roche Diagnostics, Germany). Insulin levels were measured with the Medgenic two-site IRMA assay (BioSource, Belgium). AST, ALT, ALP, and GGT were all measured by dry chemistry on Vitros Fusion 5.1FS (Ortho Clinical Diagnostics, Beerse, Belgium). Hs-CRP was assayed with nephelometry on BNII (Siemens Healthcare Diagnostics, Brussels, Belgium). The cross-sectional area of total abdominal adipose tissue (TAT), visceral abdominal adipose tissue (VAT), and subcutaneous abdominal adipose tissue (SAT) was measured by computed tomography (CT) at level L4-L5 according to previously described methods [24].

Hepatological work-up

The liver specific program included additional blood analysis, to exclude the classical etiologies of liver disease (e.g., viral hepatitis and autoimmune disease), [s-choline-esterase, carcino-embryonic antigen, α-fetoprotein, antinuclear factor, antineutrophil cytoplasm antigen antibodies, antismooth muscle antibodies, antimitochondrial antibodies, antiliver–kidney microsome antibodies, serum copper and coeruloplasmin, alpha-1-antitrypsin, anti-Hepatitis B core antibodies, Hepatitis B surface antigen, anti-Hepatitis C virus antibodies], a Doppler ultrasound of the abdomen with parameters of liver and spleen volume and liver vascularization, a liver–spleen scintigraphy using technetium tin colloid (99mTC) [25], and an aminopyrine breath test as a measure for liver metabolic reserve [26].

As diabetes has a specific clinical picture that might substantially influence the obtained results, patients already known to have diabetes or with a de novo diagnosis according to the criteria of the American Diabetes Association were not included in the protocol. Patients were also excluded from further analysis if they refused informed consent, if they had significant alcohol consumption (>20 g/day) [27], or if another liver disease was diagnosed. The possibility of liver involvement was defined by one or more of the following elements: abnormal liver tests (AST and/or ALT and/or GGT and/or ALP), ultrasound abnormality of the liver [enlarged liver [28], steatotic liver (scored 0-3) [29]], signs of parenchymal liver disease on liver–spleen scintigraphy [25], and abnormal aminopyrine breath test [26].

Liver biopsy

When one or more of the above criteria were met, a liver biopsy was proposed [30]. A separate informed consent for liver biopsy was required. In patients who subsequently were referred to bariatric surgery, the liver biopsy was performed peri-operatively. The remaining patients were proposed for transparietal liver biopsy.

The liver biopsy specimen was stored in formalin aldehyde. Haematoxylin–eosin stain, Sirius red (Fouchet) stain, periodic acid Schiff stain after diastase, reticulin stain (Gordon–Sweets), and Perl's iron stain were routinely performed on all biopsies and subsequently analyzed by an experienced pathologist (blinded for the clinical characteristics of the patients). The diagnosis of NASH was made according to Brunt et al. requiring the association of some degree of steatosis, some degree of ballooning, and some degree of lobular inflammation [31]. The different features were scored according to the NASH Clinical Research Network Scoring System [32]. Steatosis was graded as follows: less than 5% of liver parenchyma: 0, 5-33%: 1, more than 33-66%: 2, more than 66%: 3. Lobular inflammation is scored as: no foci: 0, less than 2 foci per ×200 field: 1, 2-4 foci per ×200 field: 2, more than 4 foci per ×200 field: 3. Ballooning was scored as: none: 0, few balloon cells: 1, many cells/prominent ballooning: 2. Fibrosis was staged: none: 0, perisinusoidal or periportal: 1 (divided into 1A, 1B, or 1C according to the localization and extent), perisinusoidal and portal/periportal: 2, bridging fibrosis: 3, cirrhosis: 4. Other features (e.g., portal inflammation, Mallory's hyaline) were even so assessed in the same way [32]. The NASH Activity Score was calculated as the sum of the scores for steatosis, ballooning, and lobular inflammation [32]. The length of the biopsy and the number of portal tracts were equally so reported by the pathologist.

SNP selection and genotyping

Genomic deoxyribonucleic acid was extracted from whole blood [33]. We selected the rs738409 (c.444C>G, encoding pI148M) polymorphism of PNPLA3 and the rs2854117 (c.1157C>T) polymorphism of APOC3 for analysis. TaqMan Pre-Designed Genotyping Assays (Applied Biosystems Inc., Foster City, CA, USA) were used to genotype the selected single-nucleotide polymorphisms (SNPs), according to the manufacturer's protocol, on a Lightcycler 480 Real-Time PCR System (Roche, Penzberg, Germany).

RNA extraction and real time quantitative PCR

RNA was isolated from human liver by guanidinium thiocyanate/phenol/chloroform extraction [34]. Reverse transcription was performed using the High Capacity Reverse Transcription kit (Applied Biosystems, Life Technologies, Carlsbad, USA). PCR was performed with Brilliant II SYBR Green QPCR Master Mix (Agilent Technologies, Santa Clara, USA) on a Stratagene Mx3005P system (Agilent Technologies) using specific primers. mRNA levels were subsequently normalized to those of cyclophilin and fold induction was calculated using the ΔCt method.

Statistical analyses

Statistical analyses were performed with SPSS 18.0. Log transformation was applied to BMI, ALP, HDL, TG, fasting insulin, fasting C-peptide, AUC-insulin, and HOMA-IR. Data are expressed as mean ± SD for normally distributed variables or as median (interquartile range) when distribution is skewed. Hardy–Weinberg equilibrium (HWE) was checked for all individual SNPs using the HWE program from LINKUTIL [35]. Comparison of population characteristics was performed with an independent samples t-test. To quantify the effect of a SNP on anthropometric and metabolic parameters, we performed linear regressions with age and gender as covariates. Multinominal logistic regression analysis was performed to detect an effect of a SNP on liver histology. Genotypes were coded as 1 (WT), 2 (heterozygotes), or 3 (homozygous mutants) with 1 = CC, 2 = CG, 3 = GG for rs738409 and 1 = CC, 2 = CT, 3 = TT for rs2854117. The association was tested under the assumption of an additive model. Multinomial logistic regression with age and sex as covariables was used to calculate odds ratios (OR). Due to the lack of patients with genotype 3 and steatosis grade less than 5%, no foci and no ballooning, genotypes were re-coded into 1 (homozygous wild-type (WT) allele carriers) and 2 (heterozygotes and homozygous mutant allele carriers). Power calculations were performed using Quanto, with our population we have 80% power to detect an effect of 1.66% [36]. Significance level was set at 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Main characteristics

Between August 1, 2006 and June 1, 2010, 470 consecutively included patients followed the above described protocol and were genotyped for the PNPLA3 variant rs738409 and the APOC3 variant rs2854117. Subjects were all of Caucasian origin. Clinical characteristics of the study objects are reported in Table 1. All subjects were overweight or obese (median BMI 37.6 kg/m2, inter-quartile range 33.7-41.2). Mean age was 44 ± 13 year. Median blood pressure, serum lipids and glucose parameters were within normal ranges. As could be expected in obese subjects, insulin resistance (as defined by HOMA-IR) was indicative for reduced insulin sensitivity. Median liver tests (AST, ALT, ALP, and GGT) were also within normal limits.

Table 1. Clinical characteristics of the study population
  1. BMI, body mass index; VAT, visceral adipose tissue; TG, triglycerides; HOMA-IR, insulin resistance calculated with homeostasis model assessment; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma glutamyl transpeptidase; NAS, non-alcoholic steatohepatitis activity score; NASH, non-alcoholic steatohepatitis.

  2. Data are expressed as mean ± SD (range) for normally distributed variables or as median (interquartile range) when distribution of the variable is skewed. The absolute numbers and percentages of cases are given for the different scores of steatosis grade, lobular inflammation, ballooning, NAS, NASH diagnosis, and fibrosis.

N470
Female/male331/139
Age (years)44 ± 13
BMI (kg/m2)37.6 (33.7–41.2)
VAT (cm2)182 (136–253)
Systolic blood pressure (mmHg)127 (119–138)
Diastolic blood pressure (mmHg)75 (69–83)
Fasting TG (mg/dl)127 (97–180)
Fasting glucose (mg/dl)82 (75–90)
Fasting insulin (mIU/l)13.6 (9.1–18.8)
HOMA-IR2.75 (1.84–4.09)
AST (U/l)27 (23–35)
ALT (U/l)37 (28–49)
GGT (U/l)29 (22–43)
Liver biopsy performed287
NAS (median-range)3 (0–8)
Steatosis grade
0 (<5%)79 (27.5%)
1 (5–33%)94 (32.8%)
2 (33–66%)64 (22.3%)
3 (>66%)50 (17.4%)
Lobular inflammation
0 (no foci)103 (35.9%)
1 (<2 foci per 200× field)108 (37.6%)
2 (2–4 foci per 200× field)51 (17.8%)
3 (>4 foci per 200× field)25 (8.7%)
Ballooning
0 (none)94 (32.8%)
1 (few balloon cells)99 (34.5%)
2 (many cells/prominent ballooning)94 (32.8%)
NASH diagnosis
No NASH136 (47.4%)
NASH151 (52.6%)
Fibrosis stage
0 (no fibrosis)182 (63.4%)
1 (perisinusoidal or periportal)47 (16.4%)
2 (perisinusoidal and portal/periportal)35 (12.2%)
3 (bridging fibrosis)22 (7.7%)
4 (cirrhosis)1 (0.3%)

In 35/470 (7.4%) none of the criteria to propose a liver biopsy was present. In the other patients at least one of the criteria was met. Liver biopsy was performed in 287/435 of those patients (66.0%, or 61.1% of the overall cohort of 470 patients). Liver biopsy was performed during bariatric surgery in 122/287 (42.5%). Mean biopsy length was 16.2 ± 7.8 mm (range 10-21) and the mean number of portal tracts was 8.5 ± 4.7 (range 5-11). The distribution according to the grade of steatosis, the NAS, and the stage of fibrosis are shown in Table 1.

The allele frequency of PNPLA3 variant (rs738409) was 23%. This estimate is consistent with previously reported allele frequencies [9, 11]. The allele frequency of the APOC3 variant (rs2854117) was 27%, which is also consistent with previously reported allele frequencies [17, 18]. HWE was present for the selected SNP of PNPLA3 (P = 0.86; data not shown) and the selected SNP of APOC3 (P = 0.40; data not shown).

The clinical parameters of the study subjects, stratified by three genotype classes of PNPLA3 and APOC3 are shown in Table 2. Dividing the study subjects into two groups (the variant-allele carriers versus WT homozygotes) did not change the results (data not shown).

Table 2. Clinical features of the study population stratified by PNPLA3 and APOC3 genotypes
 PNPLA3 variant rs738409P valueAPOC3 variant rs2854117P value
CCCGGGCCCTTT
  1. BMI, body mass index; VAT, visceral adipose tissue; TG, triglycerides; HOMA-IR, insulin resistance calculated with homeostasis model assessment; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma glutamyl transpeptidase; NAS, non-alcoholic steatohepatitis activity score; NASH, non-alcoholic steatohepatitis; NS, non significant.

  2. Genotypes were coded as 1 (WT), 2 (heterozygotes) or 3 (homozygous mutants) with 1 = CC, 2 = CT, 3 = TT for rs2854117 and 1 = CC, 2 = CG, 3 = GG for rs738409 and 1 = CC, 2 = CT, 3 = TT.

  3. Data are expressed as mean ± SD (range) for normally distributed variables or as median (interquartile range) when distribution of variable is skewed. The absolute numbers and percentages of cases are given for the different scores of steatosis grade, lobular inflammation, ballooning, NAS, NASH diagnosis and fibrosis. A P-value of < 0.05 (*) is considered statistically significant.

N280/470 (59.6%)166/470 (35.3%)24/470 (5.1%) 260/470 (55.3%)176/470 (37.4%)34/470 (7.2%) 
Female/male204/76109/5718/6 187/73118/5826/8 
Age (years)44 ± 1344 ± 1244 ± 13NS44 ± 1344 ± 1344 ± 14NS
Anthropometry and visceral fat
BMI (kg/m2)37.7 (33.8–41.0)37.6 (33.3–41.8)37.6 (34.4–40.7)NS37.5 (34.0–41.6)37.9 (33.6–40.8)35.7 (32.2–39.8)NS
VAT (cm2)178 (135–247)185 (138–259)182 (134–250)NS178 (135–247)185 (138–259)182 (134–250)NS
Lipid metabolism
Fasting TG (mg/dl)126 (96–191)130 (102–176)129 (75–145)NS123 (96–180)133 (98–179)141 (96–183)NS
Glucose metabolism
Fasting glucose (mg/dl)82 (76–91)82 (74–90)82 (78–86)NS82 (76–91)82 (74–90)82 (78–86)NS
Fasting insulin (mIU/l)13.2 (9.4–18.9)14.0 (9.1–18.8)13.7 (8.0–20.6)NS13.2 (9.4–18.9)14.0 (9.1–18.8)13.7 (8.0–20.6)NS
HOMA-IR2.71 (1.75–3.92)2.80 (1.87–4.22)2.93 (1.86–4.69)NS2.73 (1.85–4.23)2.83 (1.86–3.89)2.74 (1.58–4.19)NS
Parameters of liver disease
AST (U/L)26 (22–33)28 (23–38)35 (25–46)<0.001*28 (22–35)26 (23–36)28 (23–34)NS
ALT (U/L)36 (28–46)39 (30–56.5)53 (30–76)<0.001*37 (29–49)37 (29–49)37 (27–52)NS
GGT (U/L)29 (22–43)29 (21–45)31 (22–40)NS29 (22–43)29 (21–45)31 (22–40)NS
Liver biopsy performed16410617 16310618 
NAS (median-range)3 (0–8)4 (0–8)6 (2–8)<0.001*3 (0–8)3 (0–8)4 (0–8)NS
Steatosis grade   <0.001*   NS
0 (<5%)56 (34.1%)23 (21.7%)0 (−) 47 (28.8%)29 (27.4%)3 (16.7%) 
1 (5–33%)57 (34.8%)32 (30.2%)5 (29.4%) 52 (31.9%)37 (34.9%)5 (27.8%) 
2 (33–66%)39 (23.8%)21 (19.8%)4 (23.5%) 30 (18.4%)28 (26.4%)6 (33.3%) 
3 (>66%)12 (7.3%)30 (28.3%)8 (47.1%) 34 (20.9%)12 (11.3%)4 (22.2%) 
Lobular inflammation   <0.001*   NS
0 (no foci)72 (43.9%)31 (29.2%)0 (−) 57 (35.0%)41 (38.7%)5 (27.8%) 
1 (<2 foci per 200× field)59 (36.0%)40 (37.7%)9 (52.9%) 65 (39.9%)36 (34.0%)7 (38.9%) 
2 (2–4 foci per 200× field)20 (12.2%)27 (25.5%)4 (23.5%) 24 (14.7%)24 (22.6%)3 (16.7%) 
3 (>4 foci per 200× field)13 (7.9%)8 (7.5%)4 (23.5%) 17 (10.4%)5 (4.7%)3 (16.7%) 
Ballooning   0.002*   NS
0 (none)65 (39.6%)28 (26.4%)1 (5.9%) 49 (30.1%)39 (36.8%)6 (33.3%) 
1 (few balloon cells)58 (35.4%)35 (33.0%)6 (35.3%) 65 (39.9%)30 (28.3%)4 (22.2%) 
2 (many cells/prominent ballooning)41 (25.0%)43 (40.6%)10 (58.8%) 49 (30.1%)37 (34.9%)8 (44.4%) 
NASH diagnosis   <0.001*   NS
No NASH98 (59.8%)37 (34.9%)1 (5.9%) 75 (46.0%)54 (50.9%)7 (38.9%) 
NASH66 (40.2%)69 (65.1%)16 (94.1%) 88 (54.0%)52 (49.1%)11 (61.1%) 
Fibrosis stage   NS   NS
0 (no fibrosis)110 (67.1%)62 (58.5%)10 (58.8%) 108 (66.3%)61 (57.5%)13 (72.2%) 
1 (perisinusoidal or periportal)27 (16.5%)17 (16.0%)3 (17.6%) 23 (14.1%)24 (22.6%)0 (−) 
2 (perisinusoidal and portal/periportal)17 (10.4%)17 (16.0%)1 (5.9%) 17 (10.4%)13 (12.3%)5 (27.8%) 
3 (bridging fibrosis)9 (5.5%)10 (9.4%)3 (17.6%) 14 (8.6%)8 (7.5%)0 (−) 
4 (cirrhosis)1 (0.6%)0 (−)0 (−) 1 (0.6%)0 (−)0 (−) 

Association with anthropometry and visceral fat

Anthropometric measurements (BMI, waist, and WHR) did not differ between the PNPLA3 rs738409 and the APOC3 rs2854117 genotypes (Table 2). No significant differences were observed in the amount of visceral or subcutaneous adipose tissue. Correction for age, BMI, and use of lipid lowering medications did not alter the obtained results.

Association with lipid metabolism

There were no significant differences between the three genotypes of PNPLA3 rs738409 and APOC3 rs2854117 in plasma concentrations of TG, total cholesterol, HDL, or LDL (Table 2). Correction for age, BMI, and use of lipid lowering medications did not alter the obtained results.

Association with glucose metabolism

Different parameters of glucose metabolism were investigated, but no association was found between the fasting levels of plasma glucose, insulin, C-peptide, and HOMA-IR of the three different genotypes of both PNPLA3 rs738409 and APOC3 rs2854117 (Table 2). Correction for age, BMI, and use of lipid lowering medications did not alter the obtained results.

Association with parameters of liver disease

There was a significant difference between the genotype classes of PNPLA3 rs738409 in plasma concentrations of ALT and AST, but not for ALP and GGT (Table 2). Correction for age, BMI, and use of lipid lowering medications did not alter the obtained results. We also assessed the influence of the PNPLA3 variant rs738409 on the severity of histologically determined liver damage in overweight and obese patients. In our subcohort with liver biopsy, there was a significant association between rs738409 and steatosis severity. The age and sex adjusted OR for this SNP indicated that heterozygotes and homozygous mutant allele carriers have a much higher risk of having steatosis in more than 66% of liver parenchyma than WT allele carriers (ORage–sex adj=7.41, 95%CI 3.235-16.697, nominal P = 2.17 × 10−6). Also other important histological features such as lobular inflammation and ballooning did differ among the groups. The age and sex adjusted OR for this SNP indicated that heterozygotes and homozygous mutant allele carriers have a higher risk of having lobular inflammation (<2 foci per ×200 field, ORage–sex adj = 1.86, 95%CI 1.046-3.321, nominal P = 0.035), more severe lobular inflammation (2-4 foci per ×200 field, ORage–sex adj = 3.50, 95%CI 1.712-7.150, nominal P = 0.001), and prominent ballooning than WT allele carriers (ORage–sex adj = 2.78, 95%CI 1.514-5.119, nominal P = 0.001). Figure 1 shows that also the steatosis grade (Figure 1A), lobular inflammation (Figure 1B), ballooning (Figure 1C), and NASH diagnosis (Figure 1D), reflecting NASH severity show a significant association with the rs738409 genotype. The severity of liver fibrosis was not related to the PNPLA3 rs738409 genotype (Table 2).

image

Figure 1. PNPLA3 rs738409 genotype association with histological attributes: (A) Association with steatosis grade (P < 0.001). (B) Association with lobular inflammation (P < 0.001). (C) Association with ballooning (P = 0.002). (D) Association with diagnosis of NASH (P < 0.001).

Download figure to PowerPoint

There were no significant differences between the genotype classes of APOC3 rs2854117 in plasma concentrations of the liver tests ALT, AST, and GGT (Table 2). Correction for age, BMI, and use of lipid lowering medications did not alter the obtained results. One of the main aims of the study was to assess the influence of the APOC3 rs2854117 variant on the severity of histologically determined liver damage in overweight and obese patients. In our subcohort with liver biopsy (n = 267/470 or 61.1%), there was no significant association between APOC3 rs2854117 and steatosis severity. Other important histological features such as lobular inflammation and ballooning did not differ among the groups. Also the severity of liver fibrosis was not related to rs2854117 (Table 2).

Association with gene expression data in liver biopsies

In a subcohort of 52 patients, gene expression of APOC3 in liver was determined. Patients were randomly selected from the original cohort. Subject characteristics are similar to the characteristics of the overall group (data not shown). We found no correlation of gene expression data with liver steatosis (P = 0.155), ballooning (P = 0.784), lobular inflammation (P = 0.221), and fibrosis (P = 0.413).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

A variant in PNPLA3 was initially associated with fatty liver and hepatic inflammation in the first GWAS on NAFLD [6] and further robustly replicated in independent candidate gene studies [7]. However, the physiological role and biological function of PNPLA3 in the liver is still unclear. Recently, a potential role for PNPLA3 in the hydrolysis of glycerolipids has been suggested; the I148M substitution in the variant of PNPLA3 can cause a loss of function [37]. Several studies showed an association of the PNPLA3 variant with increased liver enzymes, in particular ALT [8, 9]. Metabolic comorbidities that are frequently associated with the pathogenesis of NAFLD, such as insulin resistance, obesity, and lipid abnormalities are not affected by the PNPLA3 variant rs738409. Sookoian et al. [38] were the first to demonstrate that a PNPLA3 variant was also associated with histological disease severity, a finding that was replicated by others [10]. In our cross-sectional study of a well characterized (consecutively included) patient cohort of Caucasian overweight and obese subjects, we confirm the association between PNPLA3 gene variant rs738409 with liver steatosis and liver tests ALT and AST. No association could be found between the PNPLA3 gene variant and anthropometric and metabolic parameters, which confirm the results of Speliotes et al. [12]. Also the known association with histological parameters of disease activity (inflammation and ballooning) was confirmed. In contrast with earlier findings, we found no association with fibrosis severity [11, 12]. Due to our unselected population of obese patients we have a relatively small percentage of patients with fibrosis. This is in contrast with other studies [10, 12] with more severe liver affection.

Studies aiming at stipulating parameters to identify patients with NASH and advanced fibrosis may suffer from methodological issues. In most studies, patients are highly selected, as they are referred to specialized hepatology clinics because of elevated transaminases or, in case of bariatric surgery series, only represent one extreme of the spectrum. This results in patient series with relatively high prevalence of more advanced disease. In contrast to previous studies, patients in our cohort presented to the obesity clinic for a problem of overweight and had no a priori suspicion of liver disease. They were hence not retrospectively included because of the presence of liver disease, but assessed at baseline in order to detect any sign of liver affection. All patients underwent a series of comprehensive tests aiming at detecting any sign of liver affection. If there was any indication of NAFLD, a liver biopsy was proposed. This approach, although restricted to a population with a specific risk profile, results in a less selected cohort with low prevalence of more advanced disease. As it is clearly associated with steatohepatitis severity but not fibrosis, PNPLA3 is probably not directly related to fibrogenesis, and thus likely does not represent a higher risk for fibrogenesis as such. Its association with fibrosis in some studies might be secondary to its association with inflammation, which drives fibrogenesis. The role of PNPLA3 in NASH natural history is probably restricted to its role in the susceptibility to develop more severe necroinflammatory changes in association with steatosis (which probably indirectly affects fibrogenesis).

A recent meta-analysis [7] provided unequivocal evidence of PNPLA3 (rs738409) as a strong modifier of the natural history of NAFLD in different populations around the world. It has therefore been suggested that the introduction of PNPLA3 into clinical practice may improve the diagnostic protocols and personalize therapeutic strategies in patients with chronic liver diseases [19]. Some hepatic-steatosis-associated variants, like PNPLA3, are not strongly associated with any metabolic co-morbidities (such as lipid abnormalities, insulin resistance, and obesity), indicating that hepatic steatosis is likely to be influenced by different metabolic pathways [39].

In contrast with PNPLA3, APOC3 was not discovered in a GWAS. Petersen et al. [13] assumed a role for APOC3 in steatosis and steatohepatitis, since APOC3 is a regulator of hepatic and plasma TG. They reported a higher hepatic triglyceride content (as measured by H-MRS) in the APOC3 variant-allele carriers compared to the WT homozygotes. In contrast with this study, we did not observe any significant difference between heterozygote, homozygote variant, or homozygote WT subjects in lipid profile, especially plasma TG, and insulin resistance (as defined by HOMA-IR). Possible reasons for the differences between our results and those of Petersen et al. [13] could be the ethnic and gender differences between the populations: our study was performed in a predominantly female (70.4%), European population versus a male Asian population in Petersen's series.

Our results confirm the recent data of Kozlitina et al. [17]. In this multiethnic (African Americans, European Americans and Hispanics) study, two APOC3 variants were genotyped in a population of middle aged men and women. No significant difference in hepatic fat (as measured by H-MRS) content was found between carriers and non-carriers. There was also no association with insulin resistance calculated with HOMA-IR. Also the study conducted in an obese Italian population [18] did not identify any significant association between APOC3 polymorphisms and transaminase levels (as marker of fatty liver disease), serum lipids, and insulin resistance. This is in accordance with a study of Caron et al. [16] suggesting that plasma APOC3 protein levels in overweight subjects are predominantly under control of glucose but not insulin.

In all these studies, the degree of liver fat or NAFLD was, however, not identified by histology. To better identify or define NAFLD, liver histology is still the golden standard for the diagnosis of NAFLD and for the assessment of its severity in terms of disease activity and fibrosis stage [40]. Prevalence studies rarely include histology, with the exclusion of bariatric surgery series, autopsy series or living donor series, all subject to important selection bias. In other studies, diagnosis is based on ultrasound, liver biochemistry, or H-MRS. The latter is an accurate tool to detect the amount of liver fat, but does not diagnose inflammation or fibrosis. The other parameters are even more inaccurate to diagnose NASH or fibrosis. The results of our analysis are consistent with the recent results of Valenti et al. [19], who also reported that they did not find an association between APOC3 polymorphisms and histologically proven NAFLD in Caucasians. A strength of this single centre study, in contrast to the multi-centre study of Valenti et al. [19], is that our population is a homogeneous group of patients selected by the same criteria, with all liver biopsies scored by the same pathologist.

In our study only one of the two previously reported sequence variants in APOC3 (rs2854116 and RS2854117) was measured. However, because they are in strong linkage disequilibrium, this probably has no impact on the results. Other studies who did an attempt to reproduce the results of Petersen et al. [13] genotyped the two SNPs, but could not identify any significant association between APOC3 polymorphisms and fatty liver disease, lipids, and insulin resistance either [17]. Our results are also in accordance with the very recently published study of Hyysalo et al. [20] in which genetic variants in PNPLA3 but not APOC3 contribute to the variance in liver fat content (measured by H-MRS) due to NAFLD, and the data reported by Santoro et al. [21] in a cohort of obese children and adolescents in which the APOC3 rs2854116 SNP was not associated with fatty liver (measured by MRI). The strength of our study is the availability of liver histology in a large number of patients, as golden standard for evaluation of liver damage, and the availability of PNPLA3 genotype as a positive control. Furthermore, we could not detect an association between gene expression of APOC3 in liver with steatosis. Although this analyses was performed in a subcohort of 52 patients this seems to confirm that there is no association between APOC3 and fatty liver.

In conclusion, in this large and consecutively recruited Belgian obese population we observed a significant association between the rs738409 genetic variant in the PNPLA3 gene, but not the rs2854117 genetic variant of APOC3, and the presence and severity of necroinflammatory changes (defined by liver biopsy) independently of metabolic factors.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This work is part of the project “Hepatic and adipose tissue and functions in the metabolic syndrome” (HEPADIP), which is supported by the European Commission as an Integrated Project under the 6th Framework Program (Contract LSHM-CT-2005-018734). S.B. is a postdoctoral fellow of the “Fund for Scientific Research Flanders” (FWO-Vlaanderen). D.Z. holds a pre-doctoral specialization scholarship from the “Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen)”. This study is supported by European Commission as an Integrated Project under the 6th Framework Program (Contract LSHM-CT-2005-018734).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Prentice AM. The emerging epidemic of obesity in developing countries. Int J Epidemiol 2007;35:93-99.
  • 2
    Angulo P. Nonalcoholic fatty liver disease. N Engl J Med 2002;346:1221-1231.
  • 3
    Brunt EM, Tiniakos DG. Pathological features of NASH. Front Biosci 2005;10:1475-1484.
  • 4
    Tiniakos DG, Vos MB, Brunt EM. Nonalcoholic fatty liver disease: pathology and pathogenesis. Annu Rev Pathol 2010;5:145-171.
  • 5
    Chitturi S, Wong VW, Farrell G. Nonalcoholic fatty liver disease in Asia: firmly entrenched and rapidly gaining ground. J Gastroenterol Hepatol 2011;26:163-172.
  • 6
    Romeo S, Kozlitina J, Xing C, et al. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet 2008;40:1461-1465.
  • 7
    Sookoian S, Pirola CJ. Meta-analysis of the influence of I148M variant of patatin-like phospholipase domain containing 3 gene (PNPLA3) on the susceptibility and histological severity of nonalcoholic fatty liver disease. Hepatology 2011;53:1883-1894.
  • 8
    Kotronen A, Johansson LE, Johansson LM, et al. A common variant in PNPLA3, which encodes adiponutrin, is associated with liver fat content in humans. Diabetologia 2009;52:1056-1060.
  • 9
    Kollerits B, Coassin S, Kiechl S, et al. A common variant in the adiponutrin gene influences liver enzyme values. J Med Genet 2010;47:116-119.
  • 10
    Valenti L, Al-Serri A, Daly AK, et al. Homozygosity for the patatin-like phospholipase-3/adiponutrin I148M polymorphism influences liver fibrosis in patients with nonalcoholic fatty liver disease. Hepatology 2010;51:1209-1217.
  • 11
    Rotman Y, Koh C, Zmuda JM, Kleiner DE, Liang TJ; NASH CRN. The association of genetic variability in patatin-like phospholipase domain-containing protein 3 (PNPLA3) with histological severity of nonalcoholic fatty liver disease. Hepatology 2010;52:894-903.
  • 12
    Speliotes EK, Butler JL, Palmer CD, et al. PNPLA3 variants specifically confer increased risk for histologic nonalcoholic fatty liver disease but not metabolic disease. Hepatology 2010;52:904-912.
  • 13
    Petersen KF, Dufour S, Hariri A, et al. Apolipoprotein C3 gene variants in nonalcoholic fatty liver disease. N Engl J Med 2010;362:1082-1089.
  • 14
    Zannis VI, Cole FS, Jackson CL, Kurnit DM, Karathanasis SK. Distribution of apolipoprotein A-I, C-II, C-III, and E mRNA in fetal human tissues. Time-dependent induction of apolipoprotein E mRNA by cultures of human monocyte-macrophages. Biochemistry 1985;24:4450-4455.
  • 15
    Ito Y, Azrolan N, O'Connell A, Walsh A, Breslow JL. Hypertriglyceridemia as a result of human apo CIII gene expression in transgenic mice. Science 1990;249:790-793.
  • 16
    Caron S, Verrijken A, Mertens I, et al. Transcriptional activation of apolipoprotein CIII expression by glucose may contribute to diabetic dyslipidemia. Arterioscler Thromb Vasc Biol 2011;31:513-519.
  • 17
    Kozlitina J, Boerwinkle E, Cohen JC, Hobbs HH. Dissociation between APOC3 variants, hepatic triglyceride content and insulin resistance. Hepatology 2011;53:467-474.
  • 18
    Sentinelli F, Romeo S, Maglio C, et al. Lack of effect of apolipoprotein C3 polymorphisms on indices of liver steatosis, lipid profile and insulin resistance in obese Southern Europeans. Lip Health Dis 2011;10:93.
  • 19
    Valenti L, Nobili V, Al-Serri A, et al. The APOC3 T-455C and C-482T promoter region polymorphisms are not associated with the severity of liver damage independently of PNPLA3 I148M genotype in patients with nonalcoholic fatty liver. J Hepatol 2011;55:1409-1414.
  • 20
    Hyysalo J, Stojkovic I, Kotronen A, et al. Genetic variation in PNPLA3 but not APOC3 influences liver fat in NAFLD. J Gastroenterol Hepatol 2012;27:951-956.
  • 21
    Santoro N, Zhang CK, Zhao H, et al. Variant in the glucokinase regulatory protein (GCKR) gene is associated with fatty liver in obese children and adolescents. Hepatology 2012;55:781-789.
  • 22
    Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502.
  • 23
    Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and β-cell function from fasting glucose and insulin concentrations in man. Diabetologia 1985;28:412-419.
  • 24
    van der Kooy K, Seidell JC. Techniques for the measurement of visceral fat: a practical guide. Int J Obes Relat Metab Disord 1993;17:187-196.
  • 25
    Taylor KJ, Sullivan D, Simeone J, Rosenfield AT. Scintigraphy, ultrasound and CT scanning of the liver. Yale J Biol Med 1977;50:437-455.
  • 26
    Merkel C, Bolognesi M, Bellon S, Bianco S, Honisch B, Lampe H. Aminopyrine breath test in the prognostic evaluation of patients with cirrhosis. Gut 1992;33:836-842.
  • 27
    Farrell GC, George J, de la M, Hall P, McCullough AJ, editors. Fatty Liver Disease: NASH and Related Disorders. 1st ed. Massachusetts: Blackwell Publishing; 2005. pp 1-12.
  • 28
    Cosgrove DO, Meire H, Dewbury K, editors. Abdominal and General Ultrasound. 3rd ed. New York: Churchill Livingstone; 1994. pp 227-242.
  • 29
    Saverymuttu SH, Joseph AE, Maxwell JD. Ultrasound scanning in the detection of hepatic fibrosis and steatosis. Br Med J (Clin Res Ed) 1986;292:13-15.
  • 30
    Joy D, Thava VR, Scott BB. Diagnosis of fatty liver disease: is biopsy necessary? Eur J Gastroenterol Hepatol 2003;15:539-543.
  • 31
    Brunt EM, Kleiner DE, Wilson LA, Belt P, Neuschwander-Tetri BA; NASH Clinical Research Network (CRN). Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology 2011;53:810-820.
  • 32
    Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005;41:1313-1321.
  • 33
    Miller SA, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1998;16:1215.
  • 34
    Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156-159.
  • 35
    Ott J. Documentation to LINKAGE UTILITY Programs, http://linkage.rockefeller.edu/ott/linkutil.htm, accessed on April 20, 2011.
  • 36
    Gauderman WJ, Morrison JM. QUANTO 1.1: a computer program for power and sample size calculations for genetic-epidemiology studies, http://hydra.usc.edu/gxe, accessed on April 20, 2011.
  • 37
    Huang Y, Cohen JC, Hobbs HH. Expression and characterization of a PNPLA3 protein isoform (I148M) associated with nonalcoholic fatty liver disease. J Biol Chem 2011;286:37085-37093.
  • 38
    Sookoian S, Castaño GO, Burgueño AL, Gianotti TF, Rosselli MS, Pirola CJ. A nonsynonymous gene variant in the adiponutrin gene is associated with nonalcoholic fatty liver disease severity. J Lipid Res 2009;50:2111-2116.
  • 39
    Speliotes EK, Yerges-Armstrong LM, Wu J, et al. Genome-wide association analysis identifies variants associated with nonalcoholic fatty liver disease that have distinct effects on metabolic traits. PLoS Genetics 2011;7:e1001324.
  • 40
    Brunt EM. Histopathology of non-alcoholic fatty liver disease. Clin Liver Dis 2009;13:533-544.