Association of UCP3 Polymorphisms with Nonalcoholic Steatohepatitis and Metabolic Syndrome in Nonalcoholic Fatty Liver Disease Brazilian Patients

Abstract

Background:

We investigated the possible association of uncoupling protein 3 gene (UCP3) single nucleotide polymorphisms (SNPs) with nonalcoholic steatohepatitis (NASH) and metabolic syndrome (MetS) in nonalcoholic fatty liver disease (NAFLD) Brazilian patients.

Methods:

UCP3 SNPs rs1726745, rs3781907, and rs11235972 were genotyped in 158 biopsy-proven NAFLD Brazilian patients. Statistics was performed with JMP, R, and SHEsis softwares.

Results:

The TT genotype of rs1726745 was associated with less occurrence of MetS (P = 0.006) and with lower body mass index (BMI) in the entire NAFLD sample (P = 0.01) and in the NASH group (P = 0.02). The rs1726745-T was associated with lower values of AST (P = 0.001), ALT (P = 0.0002), triglycerides (P = 0.01), and total cholesterol (P = 0.02) in the entire NAFLD sample. Between groups, there were lower values of aminotransferases strictly in individuals with NASH (AST, P = 0.002; ALT, P = 0.0007) and with MetS (AST, P = 0.002; ALT, P = 0.001). The rs3781907-G was associated with lower GGT elevation values in the entire NAFLD sample (P = 0.002), in the NASH group (P = 0.004), and with MetS group (P = 0.003) and with protection for advanced fibrosis (P = 0.01). The rs11235972-A was associated with lower GGT values in the entire NAFLD sample (P = 0.006) and in the NASH group (P = 0.01) and with MetS group (P = 0.005), with fibrosis absence (P = 0.01) and protection for advanced fibrosis (P = 0.01). The TAA haplotype was protective for NASH (P = 0.002), and TGG haplotype was protective for MetS (P = 0.01).

Conclusion:

UCP3 gene variants were associated with protection against NASH and MetS, in addition to lower values of liver enzymes, lipid profile, BMI and, lesser fibrosis severity in the studied population.

Introduction

Nonalcoholic fatty liver disease (NAFLD) is characterized by the deposition of triglycerides (>5%) in hepatocytes in patients without significant alcohol consumption.¹ The spectrum of the disease encompasses simple steatosis, nonalcoholic steatohepatitis (NASH), and a risk of progression to cirrhosis and hepatocellular carcinoma.² NAFLD has shown epidemic behavior, making it the most common chronic liver disease worldwide.³

The pathophysiology underlying NAFLD is complex and incompletely understood.⁴ Despite its high prevalence, only a small minority of affected patients develops inflammation and subsequently fibrosis and chronic liver disease, while most of them only exhibit simple steatosis.⁵ The “multiple hit” hypothesis considers several insults acting together on genetically predisposed subjects to induce NAFLD, including insulin resistance (IR), hormones secreted from the adipose tissue, nutritional factors, gut microbiota, and genetic and epigenetic factors.^4

–7

NAFLD is strongly associated with components of metabolic syndrome (MetS),⁸ with IR being the critical association condition.^9,10 In the general population, NAFLD is estimated to occur in 6.3–33%,¹¹ reaching 40–80% of individuals with type 2 diabetes mellitus (T2DM) and 30–90% of obese individuals.¹² In view of the evident association between NAFLD and metabolic dysregulation, a panel of international experts recently proposed a new definition criteria and terminology for the disease as metabolic dysfunction-associated fatty liver disease (MAFLD).¹³ However, NAFLD and MAFLD definitions do not identify exactly the same individuals, and therefore, the two terms are not fully interchangeable,¹⁴ especially in lean NAFLD without coexisting metabolic dysregulation, in which the influence of NAFLD-related genetic variants may be decisive.¹⁵

During the last 10–15 years, genetic studies, including genome-wide association studies (GWAS) and candidate gene studies, have provided new data on the importance of single nucleotide polymorphisms (SNPs) in susceptibility and/or progression of NAFLD.^16,17 It is possible that SNPs in genes involved in excessive fatty acid oxidation would predispose NASH. In contrast, these SNPs could determine the inadequate mitochondrial overload during times of excessive free fatty acid (FFA) supply.¹⁸

Uncoupling protein 3 (UCP3) is a mitochondrial anion carrier located on chromosome 11q13, which is selectively expressed in human skeletal muscle.^16,19 UCP3 belongs to a family of mitochondrial transporters known for uncoupling oxidative phosphorylation through proton leakage from the inner mitochondrial membrane and has been implicated in (a) modulation of oxidative status and reduction in reactive oxygen species (ROS) production; (b) body energy homeostasis; (c) regulation of lipid substrates; and (d) protection against IR and T2DM.^20
–22

The possible adaptation of uncoupling proteins to increased fatty acid oxidation could confer a theoretical benefit in the prevention of hepatic steatosis and, through the flow of electrons in the respiratory chain, might prevent the over-reduction of respiratory complexes and excessive mitochondrial ROS formation, protecting cells from lipoperoxidation and diminishing necro-inflammation and apoptosis.^23,24

Several SNPs in the UCP3 gene (UCP3) have been variably associated with obesity-related phenotypes and variations in body mass index (BMI),²⁵ atherogenic lipid profiles, and a decreased risk of T2DM,²⁶ with contradictory associations with a higher risk of T2DM²⁷ and prediabetes.²⁸ However, only a few studies have evaluated the association of UCP3 SNPs with NAFLD.

A Chinese study²⁹ described a higher prevalence of the GG genotype of UCP3 SNP rs11235972 in the NAFLD Chinese children group, and a Spanish study³⁰ reported an association between the CT genotype of UCP3 SNP -55 C/T (rs1800849) with higher frequency of moderate-intense steatosis and inflammation on liver biopsy in NAFLD individuals. Therefore, the present study aimed to evaluate the possible association of UCP3 SNPs rs1726745, rs3781907, and rs11235972 with NASH and MetS in NAFLD Brazilian individuals.

Methods

Clinical design and patient's selection

This was a cross-sectional study in NAFLD individuals obtained from a database of the Hepatology Outpatient Clinic at the Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HC-FMUSP), São Paulo, Brazil. The material for the SNP analysis was obtained from a DNA sample database stored in the Laboratório de Gastroenterologia Clínica e Experimental (LIM-07) do Departamento de Gastroenterologia e Hepatologia do HC-FMUSP, from individuals who were followed at the Hepatology Outpatient Clinic and who agreed—freely and spontaneously—that their DNA could be stored for future research. The study was approved by the Ethics Committee (2.160.492) of the HC-FMUSP and was conducted in compliance with the Helsinki Declaration.

The diagnosis of NAFLD was made by liver biopsy, which was performed only when the individuals had clinical and laboratory indications. The inclusion criteria were age between 18 and 75 years and biopsy-proven NAFLD. The exclusion criteria were the presence of any other chronic liver disease (viral hepatitis, autoimmune hepatitis, hemochromatosis, alpha-1-antitrypsin, Wilson's disease, celiac disease, schistosomiasis), significant alcohol intake (>100 g ethanol/week), and/or previous use of steatogenic medication (corticosteroids, amiodarone, methotrexate, tamoxifen).

A total of 258 patients were initially evaluated. In the absence of a DNA sample (n = 58) and/or absence of histological criteria for NAFLD (n = 32), a sample of 158 patients was obtained. After UCP3 polymorphism genotyping, patients were evaluated according to the presence of nonalcoholic fatty liver (NAFL) or NASH and according to the absence or presence of MetS.

Clinical variables and biochemical analysis

On the day of the liver biopsy or within a period of up to 3 months thereafter, demographic (age, sex), anthropometric (weight, height, BMI), clinical and laboratory data were obtained. The criteria for MetS were according to the National Cholesterol Education Program/Adult Treatment Panel III (NCEP ATP III).³¹

The biochemical tests were performed after 12 hrs of fasting: aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma glutamyl transferase (GGT), FBG, plasma insulin, total cholesterol and fractions, triglycerides, serum copper, ceruloplasmin, iron profile, hepatic antibodies (antinuclear antibody, anti-smooth muscle antibody), alpha-1-antitrypsin, celiac disease antibodies, and serologies for viral hepatitis B and C. For the evaluation of IR, Homeostatic Model Assessment of Insulin Resistance (HOMA-IR) (fasting glucose (mg/dL)/18 × fasting insulin (μU/mL)/22.5) was used. A HOMA-IR value greater than or equal to 2.5 was used as a marker of IR.³¹

DNA extraction and UCP3 SNP genotyping

Genomic DNA was extracted from the blood of each individual by the salting-out method, and the quantification of the DNA was determined by spectrophotometry (NanoDrop 1000 Spectrophotometer; Thermo Fisher Scientific, Waltham, MA). After extraction, the DNA was stored at −20°C.

Three UCP3 SNPs rs1726745, rs3781907, and rs11235972 were genotyped by real-time polymerase chain reaction using the TaqMan Genotyping Master Mix and TaqMan SNP Genotyping Assay (Applied Biosystems, Foster City, CA). The Assay identifications were: C__8751342_10 (rs1726745), C__25800726_10 (rs3781907), and C__25992010_10 (rs11235972).

Histological analysis

Fragments of hepatic tissue previously fixed in 4% formalin solution were processed and stained with hematoxylin–eosin, Masson's trichrome, and Picrosirius. Liver samples were blindly evaluated by an experienced liver pathologist. The histological criteria were evaluated according to the histological scoring system for NAFLD and NAFLD Activity Score (NAS), defined by the NASH Clinical Research Network Pathology Committee (CRN).³²

Patients with simple steatosis were categorized as NAFL individuals. Fibrosis was evaluated accordingly to absence vs presence of fibrosis, significative fibrosis (F ≥ 2), and advanced fibrosis (F3–F4).

Statistical analysis

Results were expressed as mean ± SD except where stated otherwise. Continuous variables were log-transformed for the analyses when the normality of the distribution was rejected by the Shapiro–Wilk W test. Fisher's chi-squared test, Student's t test (or Mann–Whitney test), and ANCOVA were used for comparisons between groups. The Hardy–Weinberg Equilibrium (HWE) test was performed in all polymorphisms. Associations with NASH, MetS, and fibrosis were assessed by regression models. Logistic regression analyses were used for cross-sectional analyses.

Odds ratios (OR) with its 95% confidence intervals (CI) were computed for the minor allele of each SNP in a codominant, dominant, or recessive model. A value of P ≤ 0.02 was considered as significant for genotype-related comparisons. Statistical analyses were performed using JMP (SAS Institute, Inc., Cary, NC), R (www.R-project.org), and SHEsis (http://analysis.bio-x.cn/myAnalysis.php) softwares.

Results

Characteristics of the study population

The demographic, biochemical, and clinical characteristics of 158 NAFLD patients are listed in Table 1. The mean age was similar between the NAFL and NASH groups, as well as the distribution between genders, with female sex being most prevalent in both groups. NASH patients showed a higher frequency of MetS (P = 0.02), higher BMI (P = 0.02), and higher concentrations of AST (P = 0.002).

Table 1.

Demographic, Biochemical, and Clinical Characteristics of Patients with Nonalcoholic Steatohepatitis/Nonalcoholic Fatty Liver

Characteristics	NASH n = 138	NAFL n = 20	P value
Age (years)	56 ± 9.5	53.8 ± 11.9	0.43
Men/Women (%)	27.5/72.5	30/70	0.79
Dyslipidemia (%)	74	65	0.43
Type 2 diabetes mellitus (%)	63	45	0.15
High blood pressure (%)	70	72	0.84
Metabolic syndrome (%)	85	63	0.02^*
BMI (kg/m²)	32.1 ± 5.3	28.6 ± 5.6	0.02^*
Fasting blood glucose (mg/dL)	117 ± 41	115 ± 45	0.83
Insulin (μU/mL)	19.8 ± 13.4	19.9 ± 29.5	0.08
HOMA-IR	5.8 ± 4.6	6.9 ± 10.0	0.64
Total cholesterol (mg/dL)	195 ± 45	199 ± 43	0.64
LDL (mg/dL)	115 ± 39	111 ± 37	0.63
HDL (mg/dL)	47 ± 14	49 ± 9	0.40
Triglycerides (mg/dL)	174 ± 106	190 ± 112	0.56
ALT (U/L)	46.5 ± 31.4	42.7 ± 23.5	0.54
AST (U/L)	35.5 ± 23.4	26.8 ± 9.8	0.002^*
GGT (U/L)	82.8 ± 86.0	76.5 ± 912	0.77

The results are expressed as the mean ± SD.

P ≤ 0.05.

ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; GGT, gamma glutamyl transferase; HDL, high-density lipoprotein; HOMA-IR, homeostatic model of assessment of insulin resistance; LDL, low-density lipoprotein; NAFL, nonalcoholic fatty liver; NASH, nonalcoholic steatohepatitis.

The overall prevalence of MetS was 82.6% (n = 128), predominating in the NASH group (85% vs. 63%). The prevalence of NASH was higher in patients with MetS (90.6% vs. 74%).

Single site associations

The genotype frequency distributions of the three gene variants were consistent with the HWE (Table 2). The TT genotype of rs1726745 was associated with less occurrence of MetS (OR = 0.18; 95% CI = 0.05–0.61; P = 0.006) (Table 3). No association was found between UCP3 SNP genotypes and NAFL or NASH groups.

Table 2.

Genotypic and Allelic Distributions of UCP3 Polymorphisms

SNPs	Genotypes % (n)			Allele frequency % (n)		HWE P
rs1726745^#	CC	CT	TT	C	T
(n = 156)	31.4 (49)	44.9 (70)	23.7 (37)	53.8 (168)	46.2 (144)	0.22
rs3781907	AA	AG	GG	A	G
(n = 158)	59.5 (94)	31.6 (50)	8.9 (14)	75.3 (238)	24.7 (78)	0.06
rs11235972^#	GG	GA	AA	G	A
(n = 156)	71.8 (112)	25 (39)	3.2 (5)	84.2 (263)	15.7 (49)	0.49

genotyping was not obtained in two individuals.

Genotype representation: major/minor alleles.

HWE, Hardy–Weinberg Equilibrium; SNPs, single nucleotide polymorphisms; UCP3, uncoupling protein 3 gene.

Table 3.

Genotype Frequencies of UCP3 Polymorphisms According to the Presence of Metabolic Syndrome

UCP3 SNPs 5′->3′	With MetS	Without MetS	OR (95% CI)	P value
rs11235972			-^**	0.99
GG	0.706	0.815
GA	0.254	0.185
AA	0.040	0.000
MAF	0.167	0.092
rs3781907			0.51 (0.05–3.88)	0.52
AA	0.609	0.556
AG	0.313	0.333
GG	0.078	0.111
MAF	0.234	0.277
rs1726745			0.18 (0.05–0.61)	0.006^*
CC	0.323	0.231
CT	0.472	0.346
TT	0.205	0.423
MAF	0.441	0.596

OR (odds ratio) for the minor allele in a recessive model obtained in logistic regression analysis adjusted for sex, age, type 2 diabetes mellitus, and dyslipidemia. ^*P ≤ 0.02.

OR [2428148 (9.9 × 1015 – -)]: due to the AA genotype absence among Without MetS individuals, it was not possible to calculate precisely the OR.

MAF, minor allele frequency; MetS, metabolic syndrome.

According to the clinical and biochemical variables

For rs1726745, T-allele carriers were found to have significantly lower values of AST, ALT, triglycerides, and total cholesterol (Table 4). In comparison between groups, there were lower aminotransferase values strictly in NASH (vs. NAFL) and MetS (vs. without MetS) subjects and lower triglyceride values in NASH subjects (vs. NAFL). In homozygosis, the T-allele carriers had lower BMI in the general sample (CX = 31.53 ± 0.56 vs. TT = 29.21 ± 0.84 kg/m²; mean ± standard error of the mean, SEM; P = 0.01) and in NASH individuals (NASH group: CX:31.86 ± 0.61 vs. TT 29.65 ± 0.95; mean ± SEM; P = 0.02) versus NAFL group (CX = 28.52 ± 1.56 vs. TT 27.21 ± 2.05; mean ± SEM; P = 0.65).

Table 4.

Associations Between Uncoupling Protein 3 Gene Single Nucleotide Polymorphisms with Biochemical Variables

Variables	rs1726745, mean ± SEM		P value
Variables	CC (n = 49)	XT (CT + TT) (n = 107)	P value
AST (U/L)	43.6 ± 3.5 (49)	30.7 ± 2.5 (107)	0.001^*
NAFL (n)	25.3 ± 12.9 (5)	23.55 ± 6.4 (14)	0.67
NASH (n)	45.0 ± 3.6 (44)	31.6 ± 2.6 (93)	0.002^*
Without MetS (n)	52.5 ± 11.3 (8)	33.1 ± 5.9 (21)	0.20
With MetS (n)	42.4 ± 3.7 (41)	30.1 ± 2.8 (86)	0.002^*
ALT (U/L)	61.3 ± 4.3	41.3 ± 3.3	0.0002^*
NAFL	57.1 ± 16.9	36.5 ± 8.4	0.22
NASH	61.6 ± 4.7	42.0 ± 3.5	0.0007^*
Without MetS	88.7 ± 14.6	43.4 ± 7.6	0.05
With MetS	58.0 ± 4.8	40.6 ± 3.6	0.001^*
Triglycerides (mg/dL)	216.1 ± 16.9	162.7 ± 12.1	0.01^*
NAFL	289.4 ± 62.8	171.2 ± 31.0	0.11
NASH	210.8 ± 17.5	161.9 ± 12.8	0.02^*
Without MetS	190.2 ± 54.6	122.5 ± 28.3	0.14
With MetS	221.1 ± 18	172.4 ± 13.6	0.04
Total Cholesterol (mg/dL)	208.1 ± 6.9	188.5 ± 5.0	0.02^*
NAFL	235.1 ± 25.7	188.0 ± 12.7	0.12
NASH	206.2 ± 7.1	188.7 ± 5.3	0.05
Without MetS	191.7 ± 22.3	173.7 ± 11.6	0.37
With MetS	210.7 ± 7.4	192.1 ± 5.5	0.04
GGT (U/L)	69.4 ± 0.13	52.9 ± 0.10	0.08
NAFL	56.2 ± 0.29	45.2 ± 0.17	0.51
NASH	73.8 ± 0.15	56.3 ± 0.11	0.13
Without MetS	67.7 ± 0.44	46.7 ± 0.23	0.43
With MetS	70.1 ± 0.14	54.5 ± 0.11	0.13
	rs3781907, mean ± SEM
	AA (n = 94)	XG (AG + GG) (n = 64)
AST (U/L)	31.6 ± 0.07 (94)	29.0 ± 0.08 (64)	0.39
NAFL (n)	25.9 ± 0.13 (11)	23.7 ± 0.13 (9)	0.64
NASH (n)	33.9 ± 0.07 (83)	32.5 ± 0.09 (55)	0.72
Without MetS (n)	34.5 ± 0.18 (16)	31.3 ± 0.19 (14)	0.69
With MetS (n)	31.04 ± 0.07 (78)	28.5 ± 0.09 (50)	0.43
ALT (U/L)	40.8 ± 0.07	37.1 ± 0.09	0.41
NAFL	35.4 ± 0.15	32.8 ± 0.15	0.73
NASH	42.7 ± 0.09	39.8 ± 0.11	0.60
Without MetS	41.0 ± 0.20	43.9 ± 0.22	0.81
With MetS	40.5 ± 0.08	35.7 ± 0.10	0.30
Triglycerides (mg/dL)	160.5 ± 0.05	156.3 ± 0.06	0.75
NAFL	163.2 ± 0.11	177.0 ± 0.11	0.60
NASH	158.7 ± 0.06	146.3 ± 0.08	0.40
Without MetS	137.7 ± 0.14	128.5 ± 0.15	0.73
With MetS	165.8 ± 0.06	164.1 ± 0.07	0.90
Total cholesterol (mg/dL)	186.7 ± 0.03	190.7 ± 0.03	0.60
NAFL	191.5 ± 0.05	190.5 ± 0.05	0.94
NASH	185.3 ± 0.03	190.6 ± 0.04	0.55
Without MetS	170.3 ± 0.07	179.5 ± 0.07	0.60
With MetS	190.1 ± 0.03	193.7 ± 0.04	0.67
GGT (U/L)	95.2 ± 9.5	57.9 ± 11.4	0.002^*
NAFL	69.0 ± 26.9	57.5 ± 31.5	0.74
NASH	98.3 ± 10.0	58.3 ± 12.0	0.004^*
Without MetS	65.8 ± 25.0	57.8 ± 26.8	0.52
With MetS	100.8 ± 10.5	58.4 ± 12.8	0.003^*
	rs11235972, mean ± SEM
	GG (n = 112)	XA (GA + AA) (n = 44)
AST (U/L)	31.4 ± 0.06 (112)	29.0 ± 0.10 (44)	0.47
NAFL (n)	24.3 ± 0.11 (99)	26.0 ± 0.15 (37)	0.71
NASH (n)	34.4 ± 0.07 (13)	31.4 ± 0.12 (7)	0.48
Without MetS (n)	33.2 ± 0.15 (23)	35.2 ± 0.33 (7)	0.87
With MetS (n)	30.9 ± 0.07 (89)	28.5 ± 0.10 (37)	0.49
ALT (U/L)	39.8 ± 0.06	39.2 ± 0.11	0.90
NAFL	33.4 ± 0.13	36.1 ± 0.18	0.72
NASH	43.4 ± 0.08	41.5 ± 0.14	0.89
Without MetS	42.3 ± 0.17	50.9 ± 0.38	0.66
With MetS	39.1 ± 0.08	38.2 ± 0.12	0.86
Triglycerides (mg/dL)	160.5 ± 0.05	156.3 ± 0.08	0.77
NAFL	162.7 ± 0.10	184.3 ± 0.013	0.43
NASH	159.0 ± 0.06	141.8 ± 0.10	0.30
Without MetS	134.1 ± 0.12	132.8 ± 0.3	0.97
With MetS	168.3 ± 0.06	159.4 ± 0.08	0.57
Total Cholesterol (mg/dL)	189.5 ± 0.02	185.9 ± 0.04	0.66
NAFL	193.2 ± 0.05	187.6 ± 0.06	0.71
NASH	188.2 ± 0.03	184.7 ± 0.05	0.73
Without MetS	173.8 ± 0.05	181.0 ± 0.13	0.78
With MetS	192.7 ± 0.03	186.7 ± 0.04	0.43
GGT (U/L)	93.0 ± 9.7	61.2 ± 14.2	0.006^*
NAFL	87.3 ± 25	59.5 ± 33.0	0.38
NASH	93.8 ± 10.3	61.7 ± 15.4	0.01^*
Without MetS	62.3 ± 21.1	47.0 ± 48.1	0.48
With MetS	95.7 ± 10.0	59.2 ± 14.8	0.005^*

The results are expressed as the mean ± standard error of the mean (SEM), with comparisons made by ANCOVA, adjusted for sex, age, body mass index, and type 2 diabetes.

P ≤ 0.02.

For rs3781907 (G-allele) and rs11235972 (A-allele), there was an association with lower GGT elevation values, especially in NASH (vs. NAFL) and MetS (vs. without MetS) individuals (Table 4).

According to the histological variables

The presence of A allele of rs11235972 was associated with fibrosis absence (F0 vs F1–F4; OR = 0.34; 95% CI = 0.14–0.80; P = 0.01; Table 5) and protection against advanced fibrosis (F0–F2 vs. F3–F4; OR = 0.17; 95% CI = 0.03–0.56; P = 0.01). The G allele of rs3781907 also indicated to be protective against advanced fibrosis (F0–F2 vs. F3–F4; OR = 0.25; 95% CI = 0.08–0.69; P = 0.01). No association was found between rs1726745, rs3781907, and rs11235972 genotypes and the histological criteria for NASH (steatosis, lobular inflammation, ballooning), NAFLD Activity Score, and significative fibrosis (F ≥ 2) (Supplementary Tables S1–S3).

Table 5.

Genotype and Allele Frequencies of UCP3 Polymorphisms According to Presence of Fibrosis

UCP3 SNPs 5′- > 3′	F1–F4 (n = 115) % (n)	F0 (n = 43) % (n)	OR (95% CI)	P value
rs11235972
Genotype frequency			0.34 (0.14–0.80)	0.01^*
GG	75.9 (86)	60.5 (26)
GA	20.5 (23)	37.2 (16)
AA	3.6 (4)	2.3 (1)
Allele frequency			0.61 (0.32–1.15)	0.12
G	86.2 (99)	79.1 (34)
A	13.8 (16)	20.9 (9)
rs3781907
Genotype frequency			0.44 (0.19–0.98)	0.05
AA	64.0 (74)	48.9 (21)
AG	28.1 (32)	39.5 (17)
GG	7.9 (9)	11.6 (5)
Allele frequency			0.61 (0.35–1.06)	0.08
A	78.1 (90)	68.7 (30)
G	21.9 (25)	31.3 (13)
rs1726745
Genotype frequency			0.85 (0.34–2.03)	0.72
CC	31.6 (36)	29.3 (13)
CT	42.0 (48)	52.2 (22)
TT	25.4 (29)	19.5 (8)
Allele frequency			1.07 (0.65–1.78)	0.78
C	53.6 (62)	54.4 (23)
T	46.4 (53)	45.6 (20)

OR (odds ratio) for the minor allele in a dominant model obtained in logistic regression analyses adjusted for sex, age, dyslipidemia, metabolic syndrome, and HOMA.

P ≤ 0.02.

Haplotype analyses

Pairwise measures of LD between the analyzed loci were quantified by the correlation coefficient r² (Fig. 1). According to the presence of NASH, TAA (rs1726745, rs3781907, rs11235972) haplotype presented a protective effect against NASH (OR = 0.01; 95% CI = 0.00–0.12; P = 0.002; Table 6) and, according to the presence of MetS, TGG haplotype exerted a protective effect against MetS (OR = 0.22; 95% CI = 0.07–0.69; P = 0.01). None of the studied haplotypes were associated with fibrosis.

FIG. 1.

Linkage disequilibrium (r² metric) measurements for the UCP3 SNPs. UCP3, uncoupling protein 3 gene; SNP, single nucleotide polymorphism.

Table 6.

Haplotype Analysis in UCP3 Polymorphisms According to Presence of Nonalcoholic Steatohepatitis

Haplotype	NASH	NAFL	OR (95% CI)	P value
CAG	0.33	0.24	1.64 (0.75–3.59)	0.37
CGG	0.07	0.08	0.91 (0.26–3.23)	0.87
CGA	0.14	0.10	1.40 (0.47–4.14)	0.77
TAG	0.41	0.39	1.14 (0.57–2.25)	0.83
TGG	0.03	0.10	0.30 (0.09–1.03)	0.10
TAA	0.00	0.04	0.01 (0.00–0.12)	0.002^*

Haplotypes represent the alleles of rs1726745, rs3781907, and rs11235972, respectively. Odds ratios (OR) and 95% CI for NASH for each haplotype. P value adjusted by FDR using Benjamini–Hochberg.

^*P ≤ 0.02.

Discussion

In the present study, we investigated the possible association of UCP3 polymorphisms with NASH and Mets in adult NAFLD individuals. To our knowledge, this was the first study to evaluate this association in the Brazilian population, with the relevant differential of histological diagnosis of NAFLD.

Currently, a global prevalence of NAFLD (diagnosed by imaging) is estimated at 25.24% and a global prevalence of NASH (histological diagnosis) of 59.1%, ranging from 60.64% to 69.25% in patients with NAFLD with biopsy indication (elevation of liver enzymes, clinical signs of liver disease or retrospective biopsy evaluations of tertiary care center) and reduced prevalence (6.67%) in patients without biopsy indication.³³ Consistent with the profile of NAFLD patients from a tertiary center, we identified a high occurrence of NASH (87.3%) in our sample.

A strong association between NAFLD and MetS components was observed in our study, reinforcing a possible bidirectional association with reciprocal causality between the two conditions.^10,34 About 44% of genes associated with NAFLD are shared by other MetS components (almost 14% with T2DM, obesity, and hypertension combined); however, NAFLD and MetS depend not only on genetic factors but also on environmental factors and the interaction between the two.³⁵

In the individual evaluation of the rs1726745, rs3781907, and rs11235972 UCP3 polymorphisms according to presence of NAFL/NASH, no significant association was identified. However, in the haplotype analysis, the TAA (rs1726745, rs3781907, rs11235972) haplotype presented as protective for NASH.

SNPs (biallelic) sometimes capture few information about genetic variability in certain regions of the genome.^36,37 Haplotypes (multiallelic) have a greater possibility of detecting phenotypic changes in relation to isolated SNPs, providing a more informative and effective analysis.^36,37 This result may also be related to the NAFLD multifactorial pathogenesis and its polygenic behavior,^38,39 whereas the genetic component of the disease is not as simple as the Mendelian pattern of unique genetic inheritance.⁷ A single Chinese study analyzed the NAFLD risk in pediatric patients with UCP3 haplotypes (rs1800849, rs11235972, rs11235972, rs3781907, rs1726745); however, no significant association was identified.²⁹

Regarding MetS, our study identified the TT genotype of rs1726745 (individual analysis) and the TGG (rs1726745, rs3781907, rs11235972) haplotype as protectors for MetS. These results are in agreement with Groop,⁴⁰ who considered, among genes contributing to MetS and regulation of lipolysis, the UCP3 as a particularly attractive candidate gene. Conversely, Meirhaeghe et al.⁴¹ and de Luis et al.⁴² did not identify a significant association between UCP3 SNP rs1800849 and MetS as an entity. We consider that a possible reduction in the MetS overall risk through the individual interaction of the UCP3 SNPs with MetS components must be taken into account.⁴³

Several UCP3 polymorphisms have been variably associated with obesity-related phenotypes⁴² and changes in BMI.⁴³ The TT genotype of UCP3 SNP -55 C/T (rs1800849) was associated with higher BMI in Asians.⁴⁴ A significant association between homozygotes for the lowest allele of UCP3 SNPs rs647126 (allele A), rs1685356 (allele A), and rs2075577 (allele G) and higher BMI in Dutch was reported.⁴⁵ Conversely, a Japanese study described an association between the TT genotype of UCP3 SNP -55 C/T with significantly lower BMI and considered that, in homozygous form, the SNP could contribute to the reduction in the prevalence of obesity by increasing UCP3 mRNA levels and fat oxidation, thus reducing BMI.⁴⁶

In the present study, the TT genotype of rs1726745 correlated with lower BMI values in the general sample and in the NASH group (vs. NAFL), supporting that UCP3 seems to be involved in obesity, given its possible participation in energy regulation and lipid metabolism.^21,47,48 The result variability between UCP3 SNPs and BMI changes seems to be importantly related to ethnicity, due to differences in lifestyle and body weight distribution, as well as differences in SNP genotype frequencies among ethnicities.^44,45

Another interesting finding in our study was the association of the T allele of rs1726745 with lower values of total cholesterol and triglycerides in the general sample and lower values of triglycerides in the NASH group, corroborating the possible role of UCP3 in lipid metabolism.^49,50 The association of UCP3 SNPs to a better lipid profile, through higher HDL values⁴⁶ and/or lower cholesterol and LDL values,⁴² has been described in literature. It is conjectured that the risk allele presence of UCP3 SNP could produce a more beneficial lipid state in skeletal muscle and adipose tissue, by modifying the fatty acid consumption rate.⁴²

Controversially, a Finland study described an association of UCP3 SNPs rs3781907 (allele G), rs1726745 (allele G), rs11235972 (allele A), and rs1800849 (allele T) with higher values of total cholesterol and LDL.²⁷ The disagreement results between UCP3 SNPs and lipid profile may be related to variations in allelic frequencies and penetrance in different populations, particularly in different ethnic groups.⁴⁴

According to Sanyal et al.,⁵¹ the liver enzyme elevation (AST, ALT, and GGT) is a liver damage sign and can be considered as useful measure in NAFLD assessment, but their values do not adequately correlate with histological severity.⁵² Our study identified a significant association of the T allele of rs1726745 with lower values of AST and ALT in the general sample and in the NASH (vs. NAFL) and MetS (vs. without MetS) groups, without a correlation with histological variables.

Hypothetically, the T allele (rs1726745) correlation with a better metabolic profile (cholesterol total, triglycerides, and BMI lower values and protection against MetS), especially in NASH patients, could contribute to FFAs/adipokine supply reduction, thereby protecting against lipotoxicity, mitochondrial dysfunction, ROS production, inflammation, and cell death.^50,53 However, the interpretation of aminotransferase changes in UCP3 SNPs and NAFLD remains uncertain.

Another liver enzyme frequently elevated in NAFLD is GGT⁵⁴ and it has been independently associated with several other pathological conditions, including T2DM, MetS, and CVD.⁵⁵ GGT is involved in glutathione metabolism, playing a critical role in the antioxidant defense, detoxification, and inflammation.⁵⁶ In oxidative stress, serum levels of oxidized glutathione increase, inducing hepatic GGT, which enables the oxidized glutathione conversion into reduced glutathione.⁵⁷ Thus, GGT is considered to play an important role in the antioxidant defense system at the cellular level, being a valuable marker of oxidative stress in NAFLD.^57,58

Some studies suggest that GGT is a significant predictor of NAFLD,^59
–61 and more specifically, that high levels of GGT are associated with a more severe NAFLD histological spectrum (presence of NASH and fibrosis),^62
–64 and that the GGT level reduction could predict histological improvement in NAFLD patients after bariatric surgery.⁶⁵ Supporting this correlation, a Turkey study demonstrated that increased GGT levels (cut off of 96.5 U/L, with 83% sensitivity and 69% specificity) was a risk factor for advanced fibrosis in NAFLD (OR = 1.0, 95% CI = 0.98–1.01; P = 0.032).⁶⁴

In the present study, both groups (NASH and NAFL) had GGT elevations. Interestingly, in a subanalysis, our study identified an individual risk allele association of rs3781907 (allele G) and rs11235972 (allele A) with lower levels of GGT elevation in the entire NAFLD sample and in the NASH (vs. NAFL) and MetS (vs. without MetS) groups. In addition, the rs3781907 (allele G) was associated with protection for advanced fibrosis, and rs11235972 (allele A) was associated with fibrosis absence and protection for advanced fibrosis.

Thus, it was possible to observe a correlation between lower levels of GGT elevation with less histological severity (fibrosis) associated with the risk allele presence of UCP3 SNPs rs3781907 and rs11235972, possibly related to a potential influence of UCP3 SNPs in reducing ROS production and oxidative stress, and, consequently, apoptosis, fibrogenesis, and GGT levels. In NAFLD, oxidative stress, worsening of the inflammatory state, and endothelial dysfunction are factors implicated in CVD pathogenesis⁶⁶; however, it is rash to infer a relationship between the lowest GGT levels and cardiovascular risk in our sample.

The UCP3 SNPs protective effect in our sample is highlighted by the interesting statistical significance found strictly in the NASH and with MetS groups for certain variables (AST, ALT, GGT, triglycerides, and BMI), inferring that the minor allele presence promoted a greater protection in groups that, usually, would present worse levels of liver enzymes, lipid profile, and BMI.

Our results must be interpreted within the limitations of cross-sectional studies and the heterogeneity between studies regarding allele frequencies and penetrance in different ethnicities, in addition to variations in age, lifestyle behaviors, medication usage, diet, and sample sizes.⁴⁴ The NAFLD histological diagnosis as an inclusion criteria enabled a detailed analysis of the histological variables in relation to the UCP3 SNPs, in the meantime, restricted the sample size. The absence of a control group was another limitation of the present study.

In conclusion, our study suggests that UCP3 polymorphisms rs1726745, rs3781907, and rs11235972 are associated with protection against NASH and MetS, in addition to lower values of liver enzymes, lipid profile, BMI, and lesser fibrosis severity in NAFLD Brazilian patients. Additional studies are necessary to further validate these findings in larger samples and in different populations.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

References

Perumpail

, Khan

, Yoo

, et al. Clinical epidemiology and disease burden of nonalcoholic fatty liver disease. World J Gastroenterol, 2017; 23:8263–8276.

Sharma

, Mitnala

, Vishnubhotla

, et al. The riddle of nonalcoholic fatty liver disease: Progression from nonalcoholic fatty liver to nonalcoholic steatohepatitis. J Clin Exp Hepatol, 2015; 5:147–158.

, Zhang

, Zhan

. Nonalcoholic fatty liver disease cirrhosis: A review of its epidemiology, risk factors, clinical presentation, diagnosis, management, and prognosis. Can J Gastroenterol Hepatol, 2018; 2018:2784537.

Tilg

, Adolph

, Moschen

. Multiple parallel hits hypothesis in nonalcoholic fatty liver disease: Revisited after a decade. Hepatology, 2020; 73:833–842.

Buzzetti

, Pinzani

, Tsochatzis

. The multiple-hit pathogenesis of non-alcoholic fatty liver disease (NAFLD). Metabolism, 2016; 65:1038–1048.

Lee

, Phyo

, Loo

, et al. Validation of genetic variants associated with metabolic dysfunction-associated fatty liver disease in an ethnic Chinese population. World J Hepatol, 2020; 12:1228–1238.

Hernaez

. Genetic factors associated with the presence and progression of nonalcoholic fatty liver disease: A narrative review. Gastroenterol Hepatol, 2012; 35:32–41.

Coroian Aurelia

. Is the non-alcoholic fatty liver disease part of metabolic syndrome?. J Diabetes Metab, 2015; 6:4.

Pardhe

, Shakya

, Bhetwal

, et al. Metabolic syndrome and biochemical changes among non-alcoholic fatty liver disease patients attending a tertiary care hospital of Nepal. BMC Gastroenterol, 2018; 18:109.

10.

Lonardo

, Ballestri

, Marchesini

, et al. Nonalcoholic fatty liver disease: A precursor of the metabolic syndrome. Dig Liver Dis, 2015; 47:181–190.

11.

Chalasani

, Younossi

, Lavine

, et al. The diagnosis and management of non-alcoholic fatty liver disease: Practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology, 2012; 55:2005–2023.

12.

Brunt

, Wong

, Nobili

, et al. Nonalcoholic fatty liver disease. Nat Rev Dis Primers, 2015; 1:15080.

13.

Eslam

, Newsome

, Sarin

, et al. A new definition for metabolic dysfunction-associated fatty liver disease: An international expert consensus statement. J Hepatol, 2020; 73:202–209.

14.

Targher

, Byrne

. From nonalcoholic fatty liver disease to metabolic dysfunction-associated fatty liver disease: Is it time for a change of terminology?. Hepatoma Res, 2020; 6:64.

15.

Lin

, Huang

, Wang

, et al. Comparison of MAFLD and NAFLD diagnostic criteria in real world. Liver Int, 2020; 40:2082–2089.

16.

Macaluso

, Maida

, Petta

. Genetic background in nonalcoholic fatty liver disease: A comprehensive review. World J Gastroenterol, 2015; 21:11088–11111.

17.

Vespasiani-Gentilucci

, Gallo

, Dell'Unto

, et al. Promoting genetics in non-alcoholic fatty liver disease: Combined risk score through polymorphisms and clinical variables. World J Gastroenterol, 2018; 24:4835–4845.

18.

Di Rosa

, Malaguarnera

. Genetic variants in candidate genes influencing NAFLD progression. J Mol Med (Berl), 2012; 90:105–118.

19.

Dato

, Soerensen

, Montesanto

, et al. UCP3 polymorphisms, hand grip performance and survival at old age: Association analysis in two Danish middle aged and elderly cohorts. Mech Ageing Dev, 2012; 133:530–537.

20.

Cioffi

, Senese

, de Lange

, et al. Uncoupling proteins: A complex journey to function discovery. Biofactors, 2009; 35:417–428.

21.

Busiello

, Savarese

, Lombardi

. Mitochondrial uncoupling proteins and energy metabolism. Front Physiol, 2015; 6:36.

22.

Wang

, Gong

, Wu

. Development of gene polymorphisms in meditators of nonalcoholic fatty liver disease. Biomed Rep, 2017; 7:95–104.

23.

Cortez-Pinto

, Machado

. Uncoupling proteins and non-alcoholic fatty liver disease. J Hepatol, 2009; 50:857–860.

24.

Oliveira

, Stefano

. Genetic polymorphisms and oxidative stress in non-alcoholic steatohepatitis (NASH): A mini review. Clin Res Hepatol Gastroenterol, 2015; 39 Suppl 1:S35–S40.

25.

Liu

, Liu

, Long

, et al. Linkage and association analyses of the UCP3 gene with obesity phenotypes in Caucasian families. Physiol Genomics, 2005; 22:197–203.

26.

Meirhaeghe

, Amouyel

, Helbecque

, et al. An uncoupling protein 3 gene polymorphism associated with a lower risk of developing Type II diabetes and with atherogenic lipid profile in a French cohort. Diabetologia, 2000; 43:1424–1428.

27.

Salopuro

, Pulkkinen

, Lindström

, et al. Variation in the UCP2 and UCP3 genes associates with abdominal obesity and serum lipids: The Finnish Diabetes Prevention Study. BMC Med Genet, 2009; 10:94.

28.

, Chen

, et al. UCP2 and UCP3 variants and gene-environment interaction associated with prediabetes and T2DM in a rural population: A case control study in China. BMC Med Genet, 2018; 19:43.

29.

, Liang

, Wang

, et al. Association between UCP3 gene polymorphisms and nonalcoholic fatty liver disease in Chinese children. World J Gastroenterol, 2013; 19:5897–5903.

30.

Aller

, De Luis

, Izaola

, et al. Role of -55CT polymorphism of UCP3 gene on non alcoholic fatty liver disease and insulin resistance in patients with obesity. Nutr Hosp, 2010; 25:572–576.

31.

Expert Panel on Detection and Treatment of High Blood Cholesterol in Adults. Executive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, And Treatment of High Blood Cholesterol In Adults (Adult Treatment Panel III). JAMA, 2001; 285:2486–2497.

32.

Kleiner

, Brunt

, Van Natta

, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology, 2005; 41:1313–1321.

33.

Younossi

, Koenig

, Abdelatif

, et al. Global epidemiology of nonalcoholic fatty liver disease-Meta-analytic assessment of prevalence, incidence, and outcomes. Hepatology, 2016; 64:73–84.

34.

Zhang

, Zhang

, et al. Identification of reciprocal causality between non-alcoholic fatty liver disease and metabolic syndrome by a simplified Bayesian network in a Chinese population. BMJ Open, 2015; 5:e008204.

35.

Sookoian

, Pirola

. Nonalcoholic fatty liver disease and metabolic syndrome: Shared genetic basis of pathogenesis. Hepatology, 2016; 64:1417–1420.

36.

Ken-Dror

, Humphries

, Drenos

. The use of haplotypes in the identification of interaction between SNPs. Hum Hered, 2013; 1:44–51.

37.

Zhang

, Calabrese

, Nordborg

, et al. Haplotype block structure and its applications to association studies: Power and study designs. Am J Hum Genet, 2002; 71:1386–1394.

38.

Sookoian

, Pirola

. Genetic predisposition in nonalcoholic fatty liver disease. Clin Mol Hepatol, 2017; 23:1–12.

39.

Lonardo

, Nascimbeni

, Maurantonio

, et al. Nonalcoholic fatty liver disease: Evolving paradigms. World J Gastroenterol, 2017; 23:6571–6592.

40.

Groop

. Genetics of the metabolic syndrome. Br J Nutr, 2000; 83 Suppl 1:S39–S48.

41.

Meirhaeghe

, Cottel

, Amouyel

, et al. Lack of association between certain candidate gene polymorphisms and the metabolic syndrome. Mol Genet Metab, 2005; 86:293–299.

42.

de Luis

, Aller

, Izaola

, et al. Relationship of -55C/T polymorphism of uncoupling protein 3 (UCP3) gene with metabolic syndrome by ATP III classification. J Clin Lab Anal, 2012; 26:272–278.

43.

Jia

, Zhang

, Ge

, et al. The polymorphisms of UCP2 and UCP3 genes associated with fat metabolism, obesity and diabetes. Obes Rev, 2009; 10:519–526.

44.

Brondani

, Assmann

, de Souza

, et al. Meta-analysis reveals the association of common variants in the uncoupling protein (UCP) 1–3 genes with body mass index variability. PLoS One, 2014; 9:e96411.

45.

van Abeelen

, de Krom

, Hendriks

, et al. Variations in the uncoupling protein-3 gene are associated with specific obesity phenotypes. Eur J Endocrinol, 2008; 158:669–676.

46.

Hamada

, Kotani

, Fujiwara

, et al. The common -55 C/T polymorphism in the promoter region of the uncoupling protein 3 gene reduces prevalence of obesity and elevates serum high-density lipoprotein cholesterol levels in the general Japanese population. Metabolism, 2008; 57:410–415.

47.

Schrauwen

, Hesselink

. The role of uncoupling protein 3 in fatty acid metabolism: Protection against lipotoxicity?. Proc Nutr Soc, 2004; 63:287–292.

48.

Bézaire

, Seifert

, Harper

. Uncoupling protein-3: Clues in an ongoing mitochondrial mystery. FASEB J, 2007; 21:312–324.

49.

Costford

, Seifert

, Bezaire

, et al. The energetic implications of uncoupling protein-3 in skeletal muscle. Appl Physiol Nutr Metab, 2007; 32:884–894.

50.

Pohl

, Rupprecht

, Macher

, et al. Important Trends in UCP3 Investigation. Front Physiol, 2019; 10:470.

51.

Sanyal

, Mukherjee

, Raychaudhuri

, et al. Profile of liver enzymes in non-alcoholic fatty liver disease in patients with impaired glucose tolerance and newly detected untreated type 2 diabetes. Indian J Endocrinol Metab, 2015; 19:597–601.

52.

Mofrad

, Contos

, Haque

, et al. Clinical and histologic spectrum of nonalcoholic fatty liver disease associated with normal ALT values. Hepatology, 2003; 37:1286–1292.

53.

Zhao

, Jiang

, Zhang

, et al. Mitochondrial electron transport chain, ROS generation and uncoupling (Review). Int J Mol Med, 2019; 44:3–15.

54.

Obika

, Noguchi

. Diagnosis and evaluation of nonalcoholic fatty liver disease. Exp Diabetes Res, 2012; 2012:145754.

55.

Shaikh TZ

, Shah

SZA

, et al. Metabolic syndrome; gamma glutamyltransferase in patients. Professional Med J, 2015; 22:414–419.

56.

Tao

, Li

, Zhu

, et al. Association between gamma-glutamyl transferase and metabolic syndrome: A cross-sectional study of an adult population in Beijing. Int J Environ Res Public Health, 2013; 10:5523–5540.

57.

Irie

, Sohda

, Anan

, et al. Reduced Glutathione suppresses Oxidative Stress in Nonalcoholic Fatty Liver Disease. Euroasian J Hepatogastroenterol, 2016; 6:13–18.

58.

Irie

, Sohda

, Iwata

, et al. Levels of the Oxidative Stress Marker γ-Glutamyltranspeptidase at Different Stages of Nonalcoholic Fatty Liver Disease. J Int Med Res, 2012; 40:924–933.

59.

Banderas

, Escobedo

, Gonzalez

, et al. γ-Glutamyl transferase: A marker of nonalcoholic fatty liver disease in patients with the metabolic syndrome. Eur J Gastroenterol Hepatol, 2012; 24:805–810.

60.

Fujii

, Doi

, Ko

, et al. Frequently abnormal serum gamma-glutamyl transferase activity is associated with future development of fatty liver: A retrospective cohort study. BMC Gastroenterol, 2020; 20:217.

61.

Hossain

, Rahman Shah

, Rahman

, et al. Gamma glutamyl transferase is an independent determinant for the association of insulin resistance with nonalcoholic fatty liver disease in Bangladeshi adults: Association of GGT and HOMA-IR with NAFLD. Diabetes Metab Syndr, 2016; 10(1 Suppl 1):S25–S29.

62.

Petta

, Macaluso

, Barcellona

, et al. Serum gamma-glutamyl transferase levels, insulin resistance and liver fibrosis in patients with chronic liver diseases. PLoS One, 2012; 7:e51165.

63.

Chan

, Ida

, Cheah

, et al. Progression of liver disease in non-alcoholic fatty liver disease: A prospective clinicopathological follow-up study. J Dig Dis, 2014; 15:545–552.

64.

Tahan

, Canbakan

, Balci

, et al. Serum gamma-glutamyltranspeptidase distinguishes non-alcoholic fatty liver disease at high risk. Hepatogastroenterology, 2008; 55:1433–1438.

65.

Dixon

, Bhathal

, O'Brien

. Weight loss and non-alcoholic fatty liver disease: Falls in gamma-glutamyl transferase concentrations are associated with histologic improvement. Obes Surg, 2006; 16:1278–1286.

66.

Sao

, Aronow

. Association of non-alcoholic fatty liver disease with cardiovascular disease and subclinical atherosclerosis. Arch Med Sci, 2018; 14:1233–1244.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB