Chronic HBV Infection Associates with Improvement of Lipid Profile and Steatosis-Related Impairment in Nonalcoholic Fatty Liver Disease: A Lipidomic Analysis


 BackgroundChronic hepatitis B virus (HBV) infection exerts an impact on lipid metabolism, but its interaction with dysmetabolism-based nonalcoholic fatty liver disease (NAFLD) remains uncertain. The present study performs lipidomic and pathological comparison in NAFLD patients with or without chronic HBV infection so as to highlight its effects on lipid metabolism and metabolic liver disease.MethodsBiopsy-proven Chinese NAFLD patients with (NAFLD-HBV group, n=21) or without chronic HBV infection (NAFLD group, n=41) were enrolled in the case-control study. Their serum lipid profile was subjected to individual investigation by ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS). Steatosis, activity, and fibrosis (SAF) scoring revealed the NAFLD-specific pathological characteristics. The effects of chronic HBV infection on demographic and clinical manifestations, serum lipidomics, and pathological characteristics were statistically assessed.ResultsChronic HBV infection associated with global alteration of serum lipid profile in NAFLD patients. Upregulation of phosphatidylcholine (PCs), choline plasmalogen (PC-Os) and downregulation of free fatty acids (FFAs), lysophosphatidylcholine (LPCs) dominated the HBV-related lipidomic characteristics. Compared to those of NAFLD group, the levels of serum hepatoxic lipids (FFAs: FFA16:0, FFA16: 1, FFA18:1, FFA18:2) were significantly lowered in the NAFLD-HBV group. These low-level FFAs demonstrated correlation to statistical improvements in aspartate aminotransferase activity (FFA16:0, r = 0.34, P = 0.003; FFA16:1, r = 0.37, P < 0.001; FFA18:1, r = 0.33, P < 0.001; FFA18:2, r = 0.42, P < 0.001), hepatocyte steatosis (FFA16: 1, r = 0.39, P = 0.01; FFA18:1, r = 0.39, P = 0.015; FFA18:2, r = 0.33, P = 0.036), and ballooning (FFA16:0, r = 0.31, P = 0.037; FFA16:1, r = 0.45, P < 0.001; FFA18:1, r = 0.37, P = 0.005; FFA18:2, r=0.30, P = 0.013).ConclusionChronic HBV infection may present improvements in the serum lipidomics and steatosis-related pathological characteristics of NAFLD.Trial registrationTrial registration number: ChiCTR-DDT-13003983. Registered 13 May, 2013, http://www.chictr.org.cn/enIndex.aspx?proj=5584.


Introduction
Chronic hepatitis B virus (HBV) re ects a worldwide chronic liver disease that affects over 250 million people, especially in the Chinese population [1,2]. By the studies published to date, there is an inverse association of viral indices (e.g., HBsAg, HBV-DNA) and incidence of hyperlipidemia (e.g., hypertriglyceridemia, hypercholesterolemia), fatty liver and metabolic syndrome (MetS) [3,4]. Thus, chronic HBV infection is proposed to exert a bene cial impact on lipid metabolism, probably on the basis of HBVhost interaction [5][6][7][8].
Serving as one of the components of MetS, NAFLD has been well described to demonstrate an intimate association with abnormalities in systemic lipid metabolism. The individuals with NAFLD exhibit a strong positive association with hyperlipidemia [12]. While patients with hypertriglyceridemia are independently predisposed to the risk of NAFLD [13]. Given the reciprocal causation of NAFLD and hyperlipidemia, they are supposed to be affected by the concurrent HBV infection. However, the role of chronic HBV infection in NAFLD and related lipometabolic disorders remains uncertain until now.
We, therefore, recruited the biopsy-proven Chinese NAFLD patients with or without chronic HBV infection.
Ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) was then employed to investigate their difference in serum lipidomics. Integrated analysis of demographic and clinical manifestations, lipid pro les, and hepatic pathological characteristics further revealed the effect of chronic HBV infection on lipid metabolism, hepatic steatosis and related impairments of NAFLD patients.

Materials And Methods
Patients A total of 62 NAFLD patients with (NAFLD-HBV group, n = 21) or without chronic HBV infection (NAFLD group, n = 41) were enrolled from the inpatients of Xinhua Hospital, Shanghai during May 2012 to May 2014. Subjects with ongoing or recent alcohol abuse (alcohol intake > 20 g/day for male, > 10 g/day for female), anti-HCV IgG/IgM positive, autoimmune hepatitis, drug-induced liver injury, primary biliary cholangeitis, Wilson's disease and other causes of liver steatosis were excluded. Each participant of the study was exposed to pathological evaluation by liver biopsy. Patients with hepatocyte steatosis (> 5%) were diagnosed to be NAFLD, and those with seropositivity for hepatitis B surface antigen (HBsAg) for at least six months were de ned as chronic HBV infection (Guidelines for the Prevention, Care and Treatment of Persons with Chronic Hepatitis B Infection, 2015) [14]. This study was approved by the Research Ethics Committee of Xinhua Hospital, and informed consent was obtained from each patient.
Clinical and laboratory assessment Demographical characteristics including age, gender, height, weight, waist-to-hip ratio and body mass index (BMI) were obtained from medical record. Blood samples were collected from each patient after 12h fasting. Alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transferase (γ-GT) and alkaline phosphatase (ALP) test were performed by a multichannel automatic analyzer (Bayer ADVIA 1650, Moss, Norway). Fasting plasma glucose (FPG), triacylglycerol (TG) andtotal cholesterol (TC) were measured using Wako Bioproducts (Wako Pure Chemical Industries, Richmond, VA, USA). HBV infection assays were routinely performed to measure hepatitis B surface antigen (HBsAg), hepatitis B e antigen (HBeAg) and antibody against HBeAg (anti-HBe) by Enzyme-linked immunosorbent assay (Abbott Laboratories, North Chicago, IL, USA). The HBV DNA levels were detected by real-time polymerase chain reaction assay with a detection limit of 100 IU/mL.

UPLC-MS/MS
The serum lipidomics were analyzed as described previously [17]. In brief, lipids were extracted from the collected serum samples, and analyzed by UPLC (Waters, Milford, USA) combined with a triple TOF 5600 mass spectrometer (AB SCIEX, USA) platform together with thirteen quality control (QC) samples. Lipids separation was performed using a UPLC ACQUITY C8 BEN column (2.1 × 100 mm; 1.7 µm; Waters, Milford, USA). The mobile phases consisted of (A) 60% acetonitrile in water, 10 mmol/L ammonium acetate, and (B) 90% isopropanol in acetonitrile, 10 mmol/L ammonium acetate. Gradient elution was carried out at a ow rate of 0.26 mL/min with the gradient conditions as follows: 0-1. with corresponding internal standards, the detected lipids data in QCs were evaluated based on their relative standard deviation (RSD), and only those with RSD below 30% were subjected to further analysis.

Statistical analysis
Statistical analyses were performed using SPSS Statistics software version 23.0 and R software version 4.0.2. The continuous variables were presented as mean ± SD or median (interquartile range), and were subjected to comparison using unpaired Student's t test. Multivariate analysis including principal component analysis (PCA) was performed using R 4.0.2, and orthogonal partial least squaresdiscriminant analysis (OPLS-DA) was performed using SIMCA 14.1 (MKS Umetrics, Malmö, Sweden). The differential serum lipids with both multivariate and univariate signi cance (OPLS-DA VIP > 1.0 and P < 0.05) were ltered on the basis of variable importance in the projection (VIP), S-plot and P value (unpaired Student's t-test). Spearman's correlation was used to exploring the correlativity between serum lipid pro le and hepatic pathological parameters. Statistical signi cance was de ned as a two-side P value < 0.05.

Results
Chronic HBV infection exhibited bene cial effects on BMI, hepatocyte steatosis and related impairments In the present study, demographic, clinical and pathological indices were compared between NAFLD patients with (NAFLD-HBV group) or without chronic HBV infection (NAFLD group). Interestingly, the NAFLD-HBV group presented the BMI much lower than that of NAFLD group (NAFLD group vs NAFLD-HBV group: 27.4 ± 3.2 vs 25.7 ± 2.6, P = 0.023) ( Table 1). Signi cant amelioration of NAFLD-speci c pathological characteristics, including hepatocyte steatosis (NAFLD group vs NAFLD-HBV group: 2.32 ± 0.65 vs 1.33 ± 0.49, P < 0.001) and ballooning (NAFLD group vs NAFLD-HBV group: 1.84 ± 0.37 vs 1.13 ± 0.52, P < 0.001), was documented in the NAFLD-HBV instead of NAFLD group (Table 1). In contrast to the comparability in most biochemical indices, there was a statistically decreased ALT activity in the NAFLD-  (Table 1). Thus, the bene cial impact of chronic HBV infection, especially on the lipid metabolism and related hepatic injury, was indicated in the NAFLD patients. The continuous variables were presented as mean ± SD or median (interquartile range). HDL, high density lipoprotein; LDL, low density lipoprotein. * P < 0.05, ** P < 0.01.

Serum lipidomics differentiated NAFLD patients with or without chronic HBV infection
Multivariate analyses were employed in our study to take an overview of the serum lipidomics between NAFLD and NAFLD-CHB groups. Dramatically, 3D PCA score plot of serum lipidomics distinctly differentiated the NAFLD patients with or without chronic HBV infection (Fig. 1A). Similar group discrimination was also obtained by the OPLS-DA score plot (Fig. 1B).
To reveal the role of chronic HBV infection in lipid metabolism, a total of 239 serum lipids was exposed to UPLC-MS/MS in both NAFLD and NAFLD-HBV groups. In result, 64 lipids among these ones (26.78%) were ltered to be statistically different by unpaired Student's t-test. Detailedly, the pro le of differential serum triacylglycerol (TGs) (Fig. 2).    Fig. 3). These characteristics convinced the dominating role of FFAs, LPCs, PCs and PC-Os in differential lipid pro le. Low-level FFAs upon chronic HBV infection associated with improvements in hepatocyte steatosis and related impairments Correlation between differential serum lipids (FFAs, LPCs, PCs and PC-Os) and biochemical indices (TBIL, DBIL, ALP, γ-GT, ALT and AST), together with NAFLD-speci c pathological characteristics (hepatocyte steatosis, ballooning, lobular in ammation, brosis), was subjected to assessment in this study.

Discussion
Nowadays, accumulating proofs shed light on the fact that chronic HBV infection deeply involves in the metabolic pro les and, subsequently, affects multiple components of MetS [3-8, 18, 19]. In contrast to the uninfected control subjects, those with chronic HBV infection demonstrate a decreased prevalence of hypertriglyceridemia and lowered level of serum TG [3,6]. Whereas hypertriglyceridemia inversely associates with the viral load in HBeAg seropositives [19]. There is also a negative correlation between HBV viral load (HBV-DNA) and serum TG level [6]. After adjusting for demographic and metabolic factors, HBV infection is now recognized to be the independent factor associated with lower risk of NAFLD, mainly attributed to the HBV-related reduction of serum and intrahepatic TG concentration [7,8,19]. Some other MetS components, including hypercholesterolemia and high blood pressure, are likely to be improved in the patients with HBV seropositivity [3,8,20,21]. Furthermore, both Third Korean National Health and Nutrition Examination Survey (KNHANES III) and cross-sectional population study in Hong Kong Chinese uncovered an association of HBsAg positivity and low prevalence of MetS [4,5,7]. X protein expression and adipokines (e.g., adiponectin) secretion may underlie the interaction of HBV and host metabolism [22][23][24].
Clinical trials have recently shown that NAFLD takes place on the basis of chronic HBV infection with an increasing annual prevalence [25], yet the viral impact on lipid metabolism, hepatic steatosis and related impairments remains to be explored. In the present study, we identi ed 26.78% (64/239) differential lipids in the serum lipidomics of NAFLD-HBV group in comparison to that of NAFLD group. The wide-range alterations veri ed by P < 0.05 and/or VIP > 1.0, including FFAs, LPCs, LPEs, LPIs, PCs, PC-Os, PEs, PE-Os, and SMs, con rmed a global metabolic effect of chronic HBV infection in the NAFLD patients. Moreover, both 3D PCA and OPLS-DA score plots for serum lipidomics distinctly differentiated NAFLD patients with or without chronic HBV infection. Thus HBV carriers with or without concurrent NAFLD are suggested to share the HBV-based modulation of systemic lipid metabolism, with some differences in their features.
To take further insight into the HBV-speci c serum lipodomics, we integrated the differential lipids obtained from Student's t -test and S-plot. When compared to those of the NAFLD group, upregulation of PCs, PC-Os and downregulation of FFAs, LPCs resultantly characterized the lipidomics of NAFLD-HBV group with statistical signi cance. Serving as inhibitors of hepatic lipogenesis, PCs induce the alleviation of orotic acid-induced rodent hepatocyte steatosis [26]. On the other hand, PCs take an essential place in the membrane integrity [27,28]. They prevent the membrane leakage to abolish hepatocyte injury and, subsequently, lobular in ammation and liver brosis [29,30]. PC-Os are a class of phospholipids that contain a vinyl ether linkage at the sn-1 position and highly arachidonic acid at the sn-2 position. They have been reported to act as potential protector against oxidative stress [31,32]. Contrastively, NAFLD patients demonstrate high serum level of FFAs, which are described to be cytotoxic and potential in the early diagnosis [33,34]. LPCs, a kind of lipid intermediate elevated in rodent and human nonalcoholic steatohepatitis (NASH), mediate the interaction of saturated fatty acid and insulin resistance [35,36].
Given their hepatic activities, these HBV-impacted differential lipids are conferred to interact with lipid metabolism and related pathological alterations in the liver.
We further investigate the interaction between lipid pro le of PCs, PC-Os, FFAs, LPCs and NAFLD-related biochemical and pathological indices in patients with concurrent chronic HBV infection and NAFLD. In the multivariate model of our study, low-level FFAs of 16:1, 18:1, and 18:2 showed signi cant association with alleviated hepatocyte steatosis. Consistently, FFAs reduction was accompanied by the improvements in steatosis-based hepatic injury (ballooning, down-regulated aminotransferase activities). With their re ux from adipose tissue to the liver, FFAs lead to lipotoxicity that contributes to hepatic steatosis and related impairments [37]. FFAs reduction upon chronic HBV infection abrogates these lipotoxicity-induced abnormalities. Our study had uncovered an interesting phenomenon that was similar to previous reports [38], namely signi cantly lowered hepatocyte steatosis occurred in the NAFLD-HBV group as compared to that of NAFLD group. Taken together, chronic HBV infection is presented to have bene cial impact on the lipid metabolism and steatosis-related liver impairments, mainly on the basis of lipidomic improvements.
There are some limitations in this study. First, sampling bias could not be excluded for the limited number  All authors have approved the contents of this manuscript and agreed on the manuscript be published.

Availability of data and materials
The data will be available on request.

Competing interests
The authors declare that they have no competing interests.  S-plot identi es the differential serum lipids between NAFLD patients with or without chronic HBV infection. Differential serum lipids with VIP > 1 are labeled in red.

Figure 3
Page 19/19 Heatmap of the differential serum lipid between NAFLD and NAFLD-HBV groups. Red denotes a relative increase, and blue denotes a relative decrease.