Association between apolipoprotein E gene polymorphism and nonalcoholic fatty liver disease in Southern China: A case‐control study

Abstract Background Apolipoprotein E (ApoE) polymorphisms have been reported to be associated with nonalcoholic fatty liver disease (NAFLD), but the conclusions of studies are inconsistent in different regions. The present study aims to investigate the role of ApoE genotypes on NAFLD in southern China. Methods A total of 1064 subjects including 372 NAFLD patients and 692 controls who attended Meizhou People's Hospital located in southern China from March 1, 2016 to April 30, 2020 were enrolled in this study. The ApoE genotypes were detected and the laboratory parameters were examined. Results Significant differences were observed between NAFLD patients and controls in the prevalence of ε3/ε3 (p < 0.001) and ε3/ε4 (p = 0.004). NAFLD patients presented higher frequency of ε4 allele than controls (p = 0.013). Logistic regression analysis suggested that ε3/ε3 was an independent risk factor (OR: 1.435, 95% CI: 1.084–1.891, p = 0.010), while ε3/ε4 was an independent protective factor (OR: 0.578, 95% CI: 0.404–0.828, p = 0.003) for development of NAFLD. In addition, allele ε4 showed a protective effect on NAFLD with an adjusted OR of 0.588 (95% CI: 0.420–0.824, p = 0.002). Conclusion Our results suggested that ApoE genotype was associated with the development of NAFLD in the population of southern China. Individuals carrying ε3/ε3 were at higher risk of NAFLD, while those carrying ε3/ε4 were at lower risk of NAFLD.


| INTRODUC TI ON
Nonalcoholic fatty liver disease (NAFLD) refers to the exception of excessive drinking and other clear liver damage factors caused by fat deposition in liver cells, including simple liver steatosis, nonalcoholic steatohepatitis (NASH), and cirrhosis. NAFLD is known to cause liver disability, and has been reported to associate with the high incidence of metabolic syndrome (MetS), type 2 diabetes mellitus (T2DM), arteriosclerotic cardiovascular disease, and colorectal tumors. 1,2 Studies suggest that the incidence of NAFLD is approximately 25%, but varied in different regions and population. 3,4 In the United States, about 25% of adults develop NAFLD, and one-quarter of them may result in NASH, which increases the risk of liver cirrhosis and liver cancer. 4,5 In China, the prevalence of NAFLD ranges from 19% to 40% depending on regions. 6 In southern China, the incidence is estimated to be about 28.83%, slightly higher than the global level. 7 The variation of the NAFLD incidences is probably due to the sample size, modalities used for diagnosis, and diversity of lifestyles and dietary habits in different regions. Recent studies reveal that NAFLD was associated with lipid disorders. [8][9][10] The disruption of the lipid metabolism balance in the liver leads to lipid accumulation, which in turn causes liver toxicity and NAFLD. 11 Dyslipidemia is characterized by elevated levels of free fatty acids, low-density lipoprotein-cholesterol (LDL-c) and triglycerides (TG), or reduced levels of high-density lipoprotein-cholesterol (HDLc). Dyslipidemia increases the deposition of fat in the liver, resulting in inflammation, lipotoxicity, and liver damage, which consequently contributes to the occurrence of NAFLD. 12,13 Apolipoprotein E (ApoE) is a plasma protein that plays an important role in dyslipidemia. The ApoE gene has 3 common alleles (ε2, ε3, ε4) which produce three homozygous (ε2/ε2, ε3/ε3, ε4/ε4) and three hybrid zygote (ε2/ε3, ε2/ε4, ε3/ε4) genotypes, as well as three isomers including ApoE2 (ε2/ε2, ε2/ε3), ApoE3 (ε3/ε3, ε2/ε4), and ApoE4 (ε3/ε4, ε4/ ε4). 14 Studies have shown that ApoE protein participates in the uptake of serum lipid by cells, and affects the metabolism of cholesterol and TG. 15,16 Previous studies have suggested that ε4 allele is a genetic risk factor for the development of liver disease, and individuals carrying ε4 are susceptible to NAFLD. 16,17 ApoE genotypes are correlated with serum LDL-c levels, manifesting as an increased level of LDL-c in person with ε4 allele and a decreased level in person with ε2 allele. 18 Although some studies have investigated the relationship between ApoE genotypes and the incidence of NAFLD, the conclusions varied intensively. 16   Association guideline for the diagnosis and treatment of NAFLD. 19 Six hundred and ninety-two subjects with ultrasonographically normal livers during this period were recruited and served as controls.

| Methods and demographic information
Patients with the following conditions were excluded from this study: (i) previous viral hepatitis, drug-induced liver injury, hepatolenticular degeneration, autoimmune liver disease, or liver cancer; (ii) patients with history of alcoholism (alcohol consumption for men is > 30 g/d and for women > 20 g/d) or drug abuse; (iii) incomplete clinical data.
A total of 372 NAFLD patients and 692 controls were finally included in this case-control study (see selection diagram in Figure 1). Patients voluntarily participated in the study and signed the written informed consent.
The demographic information was collected by a general information questionnaire (including general demographic data, smoking history, drinking history, medical history, and medication history) conducted by doctors with unified training. The diagnosis of fatty liver was completed by a specialist using a B ultrasonic detector in the hospital and two physicians conformed the ultrasonic results.

| Laboratory parameters measurement
After fasting for 8-12 h, 5 ml of venous blood was collected from the patient. The serum lipid profiles and blood glucose were examined by AU5400 analyzer (Beckman Coulter). The concentration of HbA1c was examined by Premier Hb9210 HbA1c Analytical Column (Trinity Biotech). The liver enzymes, including aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), and glutamyltransferase (GGT), were measured using commercially available assays by AU5821 analyzer (Beckman Coulter).

| ApoE genotyping
The DNA was extracted from the venous blood by Blood Genomic DNA Extraction Kit (Tiangen) and quantified by NanoDrop 2000 spectrophotometer (ThermoFisher). The ApoE genotyping was conducted using a commercially available kit (Sinochips Bioscience Co., Ltd). Briefly, samples were amplified on the Thermo Cycler (Life Technologies) with the following program: 50°C for 2 min, predenaturation at 95°C for 15 min, 45 cycles of denaturing at 94 °C for 30 s, and annealing at 65°C for 45 s. The amplified products were analyzed by the fully automated GeneChip detection system. 20

| Statistical analysis
The Hardy-Weinberg equilibrium was verified for all of the recruited patients. Data were analyzed by SPSS Statistics version 20.0 software (IBM). Continuous data were presented as mean ± SD and compared using Student's t test or Mann-Whitney U-test. Categorical variables were presented as number (%) and compared using chi-square (χ 2 ) test or Fisher's exact test. The logistic regression analysis was used to analyze the correlation between ApoE genotypes and risk of NAFLD. Adjusted odds (OR) were calculated to show the predictive value. All statistical tests were two-sided and a p < 0.05 was considered significant.

| Characteristics and laboratory features of study subjects
The characteristics and laboratory features of NAFLD patients and controls are presented in Table 1. A total of 372 NAFLD patients and 692 controls were included in the present study. The mean age was similar between groups, but the gender and DM patient distribution were different. The levels of TG, total cholesterol (TC), LDL-c, apolipoprotein B (Apo B), and GGT were significantly higher in NAFLD compared with those in controls (all p < 0.05). The NAFLD patients had lower level of HDL-c than controls (p < 0.01). In addition, the NAFLD patients had higher levels of ALT and AST than the controls (p < 0.01).

| Effect of ApoE on lipid profile
We next investigated the effect of ApoE genotype on serum lipid profile. Previous studies have shown the opposite effects of ε2 and ε4 allele on lipid metabolism, so the ε2/ε4 phenotype was excluded from the analysis. 14 The subjects were divided into three subgroups based on their phenotype, namely ApoE2 (ε2/ε2 + ε2/ε3), ApoE3 (ε3/ε3), and ApoE4 (ε4/ε3 + ε4/ε4). The comparisons of serum lipidlipoprotein levels and liver function index between NAFLD patients and controls in subgroups were shown in Table 3. It was observed that TG levels were significantly higher in NAFLD patients than those in controls in all subgroups (all p < 0.05). In the subgroup of ApoE3, NAFLD patients exhibited significantly lower HDL-c and higher LDL-c than controls. In the subgroup of ApoE2 and ApoE3, NAFLD patients had higher ApoB than controls.
We also analyzed the liver enzyme between NAFLD patients and controls in different subgroups. It was observed that NAFLD patients presented significantly higher level of ALT (p < 0.05) in all of subgroups, higher level of AST in the subgroup of ApoE2 (p < 0.05), and higher level of GGT in the subgroup of ApoE4 (p < 0.01). In addition, the ratio of ALT/AST was higher than the controls in the subgroup of ApoE3 and ApoE4 (p < 0.001).
F I G U R E 1 Flow diagram of selection of cases and controls from the cohort

| Risk factors for NAFLD
Logistic regression analysis was applied to determine the predicting value of ApoE genotype and allele for NAFLD. As shown in Table 4, it was indicated that ε3/ε3 was a risk factor for NAFLD (adjusted OR: 1.435, 95% CI: 1.084-1.891, p = 0.010), while ε3/ε4 was a protec-  Note: p value: comparisons were made between NAFLD patients and controls using chi-square (χ 2 ) test.

TA B L E 2 Genotype distributions and allele frequencies in NAFLD patients and controls
polymorphisms and NAFLD are summarized in Table 5. In different regions, the association between ApoE gene polymorphism and NAFLD was different. In the present study, we found ApoE ε3/ε3 genotype acted as an independent risk factor of NAFLD, and ε3/ε4 was played as a protective factor in the development of NAFLD for population in southern China.

| DISCUSS ION
Nonalcoholic fatty liver disease is widely considered as the most common cause of chronic liver diseases, and has become an important global public health problem. 5 Recent studies have revealed that NAFLD was associated with lipid disorders and disruption of lipid metabolism balance in the liver. 1 ApoE is a key protein that affects lipid metabolism, and its polymorphism has been reported to be associated with incidence of NAFLD in various populations. In the present study, we investigated the lipid profiles and ApoE genotype distribution in NAFLD patients and controls from southern China. We found that ApoE influenced the serum lipid profile in NAFLD patients, and ε3/ε3 served as independent risk factor, while ε3/ε4 and allele ε4 serve as protective factor of NAFLD.
Apolipoprotein E is a multifunctional protein whose synthesis, secretion, and metabolism are mainly completed in the liver. ApoE plays a key role in the metabolism of TC and TG. 18 ApoE-deficient mice fed with high-fat diet intensively increased TG accumulation in the liver, 21 while introduction of the ε3 transgene reversed this process. 22 It has been reported that ApoE protein participated in the degradation of LDL, HDL, and other lipids by binding to its receptors, and this process depended on the affinity between ApoE and receptors. 23 Study showed that ApoE protein of ε2 allele preserved low affinity with LDL receptor, and ApoE protein of ε4 allele had higher affinity with LDL receptors. As a result, ε2 allele carriers prone to have lower level of TC and LDL-c, while ε4 allele carriers have high level of TC and LDL-c. 24 Consistently, we observed that the serum levels of TC and LDL-c in ε2 NAFLD patients were lower than those in ε4 carrier.  and NAFLD were shown in Table 5. In consistent with previous study, we found that in southern China the ApoE ε3/ε3 was an independent risk factor of NAFLD. However, ε3/ε4 appeared to be a protective factor of NAFLD, which was different from outcomes in other studies. A potential reason for the discrepancy may be attributed to different ethnic, the lifestyle habits, and clinical characteristics. 28 Otherwise, there was not any significant association between ε2/ε4 and ε2/ε3 with NAFLD in the present study. It was also observed that ε4 allele held a significantly protective effect on the development of NAFLD, which was in accord with findings from the previous studies. 29 The mechanisms regarding how ApoE influences the etiopathogenesis of the NAFLD were not very clear, but evidences suggested that this process may be involved in lipid metabolism.
Studies suggested that ApoE is involved in the regulation of hepatic TG-rich VLDL secretion. 30,31 Kypreos et al. 32 found the ApoE4 was essential for promoting hepatic TG-rich VLDL secretion. It was indicative that ε4 allele may act as a protective factor of NAFLD in the regulation of TG-rich VLDL rather than the cholesterolemia. Unlike ε4, ε3 preferentially targeted the HDL rather than VLDL, and alter the lipolytic processing in the circulation. 33 In this regard, individuals carrying ε4/ε3 had a lower level of VLDL-apoB than those with ε3/ε3. 30 It also had a significant association between serum ApoE and lipid peroxide levels in ε4/ ε3 phenotype patients. 34 This was the first study to investigate the relationship be-

| CON CLUS ION
In conclusion, the present study investigated the relationship between ApoE polymorphisms and NAFLD in southern China. Our data suggested that ε3/ε3 might serve as an independent risk factor, while ε3/ε4 and allele ε4 as a protective factor of NAFLD. Our findings may provide practical information for the prevention and treatment of NAFLD in this region.

ACK N OWLED G EM ENTS
The author would like to thank other colleagues who were not listed in the authorship of Center for Precision Medicine, Meizhou People's Hospital (Huangtang Hospital) for their helpful comments on the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used and/or analysis during the current study available from the corresponding author on reasonable request.

NAFLD risk
Note: ↑represents an increased allele frequency in NAFLD patients; ↓represents a decreased allele frequency in NAFLD patients.