PNPLA3 rs738409 and TM6SF2 rs58542926 gene variants affect renal disease and function in nonalcoholic fatty liver disease

This work was funded, in part, by the Piedmont Region Funds Comitato Interministeriale per la Programmazione Economica 2008. Potential conflict of interest: Nothing to report. To the Editor: We read with interest the article by Dongiovanni et al.1 reporting on the impact of the rs58542926 C>T variant of the transmembrane 6 superfamily member 2 gene (TM6SF2) on liver histology and cardiovascular disease (CVD) risk in NAFLD. Elucidating mechanisms connecting NAFLD to CVD would have important preventive and therapeutic implications in these patients. Chronic kidney disease (CKD) is a frequent, underappreciated condition and an established risk factor for CVD.2 In a recent meta‐analysis of observational studies, the presence and severity of NAFLD at baseline were associated with an increased incidence and stage of CKD, independently of traditional risk factors.3 CKD might be a mediator or a marker of an increased CVD risk in NAFLD. Thus, we explored the impact of the two polymorphisms in PNPLA3 and TM6SF2 genes, both modulating liver disease severity and CVD risk in NAFLD, on renal function in a cohort of 202 nonobese nondiabetic subjects (61 with biopsy‐proven noncirrhotic NAFLD; Table 1). CKD, estimated glomerular filtration rate (eGFR), and microalbuminuria were defined according to current Kidney Disease: Improving Global Outcomes (KDIGO) guidelines (see legend to Table 1).2 Table 1 - Main Characteristics of Patients With NAFLD and Controls NAFLD a Controls PNPLA3 CC (n = 30) PNPLA3 CG/GG (n = 31) P NAFLD PNPLA3 CC (n = 60) PNPLA3 CG/GG (n = 81) P Controls Age, years 49 ± 3 48 ± 4 0.892 49 ± 3 39 ± 4 0.889 Sex, %males 63 64 0.713 68 70 0.815 BMI, kg/m2 25.7 ± 1.5 25.9 ± 1.4 0.785 25.8 ± 1.7 26.2 ± 1.3 0.314 Systolic BP, mmHg 130 ± 5 131 ± 6 0.926 130 ± 6 131 ± 8 0.813 Diastolic BP, mmHg 86 ± 6b 88 ± 6b 0.262 81 ± 5 82 ± 6 0.887 Waist, cm 90 ± 4 92 ± 4 0.218 91 ± 4 90 ± 4 0.235 Tg, mg/dL 100 ± 32 102 ± 34 0.389 105 ± 41 107 ± 42 0.964 LDL‐C, mg/dL 111 ± 8 119 ± 7 0.562 129 ± 9 137 ± 9 0.317 HDL‐C, mg/dL 51 ± 3b 48 ± 2b 0.139 59 ± 2 60 ± 2 0.911 Total C, mg/dL 178 ± 10 186 ± 10 0.569 199 ± 12 193 ± 13 0.587 Glucose, mg/dL 97 ± 3 99 ± 4 0.598 97 ± 5 98 ± 6 0.713 Insulin, µU/mL 13.3 ± 2.6b 17.4 ± 3.7b 0.318 7.6 ± 1.6 7.5 ± 1.7 0.713 HOMA‐IR 3.40 ± 1.88b 4.68 ± 2.6b 0.118 1.68 ± 1.11 1.76 ± 0.98 0.369 AST, U/L 42 ± 3 59 ± 4 0.001 15 ± 3 16 ± 4 0.815 ALT, U/L 71 ± 5 89 ± 6 0.025 16 ± 4 17 ± 3 0.613 C‐reactive Protein, ng/mL 1.6 ± 0.2 2.8 ± 0.2b 0.006 1.1 ± 0.3 1.8 ± 0.2 0.042 Metabolic syndrome, % 37b 44b 0.412 12 11 0.835 Histological steatosis, % hepatocytesa 16 ± 10 32 ± 7 0.022 — — — Necroinflammatory gradea 1.7 ± 0.3 2.2 ± 0.4 0.028 — — — Fibrosis stagea 1.0 ± 0.4 1.9 ± 0.4 0.038 — — — NASH, %a 33 64 0.039 — — — TM6SF2 CC, % 82 83 0.771 85 84 0.812 CT/TT, % 18 17 0.816 15 16 0.913 eGFR, mL/min/1.73 m2) 79 ± 11 66 ± 8c 0.001 96 ± 10 69 ± 9 0.041 Microalbuminuria, % 0 26b 0.009 0 10 0.032 CKD, % 0 29b 0.005 0 10 0.032 Data are presented as mean ± SEM. Differences between groups were analyzed by ANOVA for normal variables; otherwise, Mann‐Whitney's test was used for nonparametric variables. Normality was evaluated by Shapiro‐Wilk's test. Fisher's or chi square test were used to compare categorical variables, as appropriate. Differences were considered statistically significant at P < 0.05. Analysis of different parameters and of genetic polymorphisms was made using Spearman's correlation test. Genetic polymorphisms were modeled as an additive effect, that is, quantitative predictor variables reflecting the number of risk alleles (0, 1, or 2). When a relation was found on univariate analysis, multiple linear regression and logistic regression analyses were used to estimate relationships between different variables, after log transformation of skewed data. Microalbuminuria was defined by an albumin‐creatinine ratio of 30‐300 µg/mg on fresh morning urine sample.a40 NAFLD patients with liver biopsy.bP < 0.05 versus controls.cP < 0.05 versus controls with PNPLA3 CC genotype.Abbreviations: BP, blood pressure; C, cholesterol; CKD, chronic kidney disease, defined by persistent (>3 months) albuminuria and/or eGFR <60 mL/min/1.73 m2; eGFR, estimated glomerular filtration rate, calculated with the Chronic Kidney Disease Epidemiology Collaboration (CKD‐EPI) equation; HOMA‐IR, homeostasis model assessment of insulin resistance; Met Sy, metabolic syndrome as defined according to the joint statement of the American Heart Association, International Diabetes Federation, and National Heart Lung and Blood Institute; PNPLA3, patatin‐like phospholipase domain‐containing 3; TM6SF2, transmembrane 6 superfamily member 2; Tg, triglycerides. We found that the presence of PNPLA3 G allele was associated with lower eGFR and with a higher prevalence of microalbuminuria and CKD (Table 1). Conversely, the TM6SF2 T allele was associated with higher eGFR and with a lower prevalence of albuminuria and CKD (P < 0.05). On multiple regression analysis, PNPLA3 polymorphism remained independently associated with eGFR (β = −0.368; P = 0.029) and with a higher risk of microalbuminuria (odds ratio [OR]: 3.52; 95% confidence interval [CI]: 1.19‐7.81; P = 0.028) and of CKD (OR, 3.87; 95% CI: 1.18‐7.30; P = 0.0021). Whether CKD, a correctable risk factor, mediates the effects of these polymorphisms on CVD risk in NAFLD may have relevant screening and therapeutic implications for these patients and warrants confirmation in large follow‐up studies.