Hepatology International

, Volume 7, Issue 2, pp 516–523

Oxidative stress is closely associated with insulin resistance in genotypes 1 and 3 chronic hepatitis C

Authors

  • Said M. Hashemi
    • Storr Liver Unit, Westmead Millennium InstituteUniversity of Sydney at Westmead Hospital
  • David van der Poorten
    • Storr Liver Unit, Westmead Millennium InstituteUniversity of Sydney at Westmead Hospital
  • Francisco Barrera
    • Storr Liver Unit, Westmead Millennium InstituteUniversity of Sydney at Westmead Hospital
    • Departamento de GastroenterologíaPontificia Universidad Católica
  • Priyanka Bandara
    • Storr Liver Unit, Westmead Millennium InstituteUniversity of Sydney at Westmead Hospital
  • Ora Lux
    • Department of Clinical ChemistryPrince of Wales Hospital
  • James Kench
    • Department of Anatomical PathologyRoyal Prince Alfred Hospital
    • Storr Liver Unit, Westmead Millennium InstituteUniversity of Sydney at Westmead Hospital
Original Article

DOI: 10.1007/s12072-012-9400-5

Cite this article as:
Hashemi, S.M., van der Poorten, D., Barrera, F. et al. Hepatol Int (2013) 7: 516. doi:10.1007/s12072-012-9400-5
  • 84 Views

Abstract

Background

Chronic hepatitis C (CHC) infection is associated with insulin resistance and with oxidative stress, but the relationship between the two has not been thoroughly examined.

Purpose

To evaluate the association between insulin resistance and oxidative stress in CHC patients.

Method

In 115 CHC patients (68 with genotype 1 and 47 with genotype 3), the relationship between the serum concentration of malondialdehyde (MDA), a marker of oxidative stress and insulin resistance as defined by the homeostasis model (HOMA-IR) was examined.

Results

There was no significant difference in MDA levels between genotype 1- and genotype 3-infected subjects (12.882 vs. 12.426 ng/mL, p = 0.2). By univariate analysis, factors associated with HOMA-IR in both genotypes were oxidative stress as measured by MDA (p = 0.002), body mass index (BMI), portal activity, and fibrosis. Genotype-specific differences in HOMA-IR association were steatosis and triglycerides (TG) for genotype 1, and age and glutathione (GSH) for genotype 3. In a stepwise multiple linear regression analysis in both genotypes, MDA was a significant and independent predictor of HOMA-IR (p = 0.04). As expected, BMI and fibrosis were likewise independently correlated to HOMA-IR. In addition, MDA levels were higher (p < 0.001) and GSH levels were lower (p = 0.023) in insulin-resistant subjects compared to their insulin-sensitive counterparts.

Conclusions

It is concluded that in CHC, oxidative stress is an independent predictor of HOMA-IR, irrespective of virus genotype. Further studies on the role of oxidative stress in the development of insulin resistance in CHC are warranted.

Keywords

Hepatitis CInsulin resistanceMalondialdehydeOxidative stress

Abbreviations

CHC

Chronic hepatitis C

MDA

Malondialdehyde

GSH

Glutathione

HOMA-IR

Homeostasis model assessment of insulin resistance

BMI

Body mass index

WHR

Waist/hip ratio

IR

Insulin resistance

HCC

Hepatocellular carcinoma

HCV

Hepatitis C virus

IR

Insulin resistance

ROS

Reactive oxygen species

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

GGT

Gamma-glutamyl transpeptidase

T-chol

Total cholesterol

HDL-C

High density lipoprotein cholesterol

LDL-C

Low density lipoprotein cholesterol

TG

Triglycerides

HIV

Human immunodeficiency virus

HPLC

High performance liquid chromatography

SD

Standard deviation

DNPH

Dinitrophenylhydrazine

UV

Ultraviolet

NS3

Non-structural protein 3

NS5A

Non-structural protein 5A

SOCS-3

Suppressor of cytokine signaling-3

NF-κβ

Nuclear factor-κβ

TNF-α

Tumor necrosis factor-α

Introduction

Oxidative stress is defined as a persistent imbalance between the production of highly reactive molecular species (chiefly oxygen and nitrogen) and antioxidants defenses [1]. When oxidative stress is present in tissues, lipid, protein, and DNA, peroxidation ensues, which in the case of lipids leads to the destruction of polyunsaturated fatty acid constituents of cellular membranes [2]. This process in biological systems may occur under enzymatic control (leading to the generation of lipid-derived inflammatory mediators) or non-enzymatically in association with cellular damage. Oxidative destruction of fatty acids further leads to the production of toxic and reactive aldehyde compounds (metabolites), such as malondialdehyde (MDA) [3]. These highly cytotoxic metabolites produced in relatively large amounts in the presence of oxidative stress can diffuse from their site of origin to attack distant targets and tissues, impairing their function.

Hepatic and systemic oxidative stress is increasingly recognized as an accompaniment of chronic hepatitis C (CHC) infection [410]. Concomitantly, CHC is specifically associated with metabolic dysregulation, including the development of hepatic steatosis and insulin resistance (IR). IR in CHC is clinically relevant as it predicts faster progression to fibrosis and cirrhosis that may culminate in liver failure and hepatocellular carcinoma (HCC) [6, 1012]. Similarly, it has been recognized that IR in CHC predicts a poor response to antiviral therapy [1315].

More recently, the role of oxidative stress as an inflammatory pathway that can induce IR has been suggested [1, 1619]. Because the hepatitis C virus (HCV) can induce oxidative stress and reactive oxygen species (ROS) can exacerbate or even cause IR, we hypothesized that there might be a close relationship between IR and oxidative stress in subjects with CHC. In the present study, we sought to investigate this possibility. To reduce the effect of confounding variables, we restricted our analysis to patients with genotypes 1 and 3 HCV infection. MDA was used to assess systemic oxidative stress in CHC as it is a stable product of lipid peroxidation that has been validated as an indicator of oxidative cellular damage [20, 21].

Methods

Study cohort and data collection

A total of 115 consecutive, prospectively enrolled CHC patients in a longitudinal natural history study formed the cohort. The study protocol was approved by the ethics committee of the University of Sydney and the Sydney West Local Health District, and all participants provided informed written consent. All participants were serologically positive for anti-HCV antibodies (Monolisa anti-HCV; Sanofi Diagnostics Pasteur, Marnes-la-Coquette, France), and HCV RNA positive by polymerase chain reaction (PCR) (Amplicor HCV; Roche Diagnostics, Branchburg, NJ, USA). Exclusion criteria were coinfection with hepatitis B virus (HBV) or human immunodeficiency virus (HIV), autoimmune hepatitis, Wilson’s disease, homozygosity for the genetic hemochromatosis (HFE) allele C282Y, α1-antitrypsin deficiency or cholestatic liver disease. Patients who were taking insulin or oral hypoglycemic medications for previously diagnosed diabetes mellitus were excluded.

Demographic variables recorded included age, gender, country of birth, and ethnicity. Patients’ height, weight, and waist and hip circumference were measured. Body mass index (BMI, calculated as weight [kg]/(height [m])2) and waist/hip ratio (WHR) were used as anthropometric measures. The assessment of lifestyle and dietary factors was by direct interview. Patients with more than 40 g/day alcohol consumption were excluded.

Biochemical and histological analyses

Venous blood was drawn at the time of liver biopsy following a 12-h overnight fast. Liver biochemical tests [alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transpeptidase (GGT), bilirubin, albumin, globulin), plasma glucose, lipid studies [total cholesterol (T-chol), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), TG], iron, copper, ferritin, prothrombin time, and INR were measured using standard automated techniques in the clinical pathology laboratory. Serum insulin was determined by radioimmunoassay (Phadaseph insulin RIA; Pharmacia and Upjohn Diagnostics AB, Uppsala, Sweden). Serum c-peptide was estimated by a competitive immunoassay (IMMULITE; Diagnostic products, Los Angeles, CA, USA). IR was calculated by the homeostasis model (HOMA) method using the formula: IR = fasting insulin (μU/mL) × fasting glucose (mmol/L)/22.5. We defined IR as a HOMA-IR score ≥ 75th percentile (1.64) from a previously defined normal population, as per World Health Organization (WHO) guidelines (WHO defines IR as HOMA-IR score of the highest quartile of the non-diabetic population; this cut-off value discriminates healthy patients from those at risk of diabetes) [2224]. HCV genotyping was performed using a second-generation reverse hybridization line probe assay (Inno-Lipa HCV II; Innogenetics, Zwijndrecht, Belgium). Viral load determination was by Roche Amplicor HCV Monitor version 2.0 kit (Roche Diagnostics, Raritan, NJ, USA) and was categorized as “low” (<850,000 IU/mL) or “high” (>850,000 IU/mL).

To estimate MDA, blood collected in an Ethylenediaminetetraacetic acid (EDTA)-coated tube was centrifuged within 30 min of collection at 4 °C (10 min at 4,000 rpm). The plasma phase was separated and kept frozen at −70 °C till assay. MDA was measured by reverse phase high performance liquid chromatography (HPLC) with UV detection using Waters-Lamda-Max 481 set at UV 310 nm after derivatization with 2,4-dinitrophenylhydrazine (DNPH) as previously published [25]. The linear range for the assay is 0–100 μM and the coefficient of variation 7 %.

For the measurement of GSH, 800 μL of 5 % metaphosophoric acid was added to 200 μL of freshly collected blood and incubated on ice for 10 min before centrifugation (2 min at 14,000 rpm). The acidified supernatant was frozen at −70 °C. Total GSH was assayed by reverse phase HPLC with fluorometric detection, according to the method of Toyo’oka et al. [26]. Vitamins E and C were measured in heparin plasma. Upon collection, blood samples were protected from light and centrifuged within 30 min at 4 °C (10 min at 4,000 rpm). HPLC with electrochemical detector was used for vitamin C determination [27], and HPLC with UV detection was employed for vitamin E determination [28]. Liver biopsies were assessed by an experienced hepatopathologist (JK) blinded to clinical information other than the positive HCV status. Portal/periportal and lobular inflammatory grade and fibrosis stage were graded from 0 to 4, according to Scheuer [29].

Statistical analysis

The statistical software package SPSS (version 16) was used for data analysis. Continuous variables were reported as mean ± standard deviation (SD) and categorical variables as frequency and percentage, unless otherwise stated. Two-sided tests with a significance level of 5 % were employed throughout. The independent sample t test was used for comparison of continuous variables between two groups. Spearman’s rank correlation was used to quantify the associations between continuous or ordered categorical variables. Stepwise multiple linear regression analysis was performed to establish the independent predictors of HOMA-IR. Log transformation of HOMA-IR was done prior to analysis due to the skewed distribution of data. Relative weight was calculated for every independent predictor to estimate the individual effect of the variable in the estimation of HOMA-IR in the multivariate model [30]. An interaction term (genotype multiplied by MDA) was used in the HOMA-IR model to determine if the association between MDA and HOMA-IR was genotype dependent.

Results

The baseline demographic, clinical, and histological characteristics of the patient cohort are summarized in Table 1. Most of the subjects were male (70.4 %) with a mean age of 41 ± 8.9 years (range 16–70) and an average BMI of 26.33 ± 4.72 kg/m2. A total of 68 (59.1 %) patients were infected with HCV genotype 1 and 47 (40.9 %) patients with HCV genotype 3. Hepatic steatosis (grade ≥1) was present in 72 (62.6 %) subjects, with a greater percentage seen in those with genotype 3 disease (G3: 72.3 % vs. G1: 55.9 %, p < 0.05). Of the patients 20 % had advanced fibrosis (F3 or 4). A total of 85 (73.9 %) patients had IR as defined by a HOMA-IR >1.64 [53 (79.9 %) cases in genotype 1 and 32 (68.1 %) cases in genotype 3]. There was no significant difference between genotypes 1 and 3 in BMI (26.76 vs. 25.7 kg/m2, p = 0.23), HOMA-IR (3.27 vs. 2.79, p = 0.21), MDA (12.88 vs. 12.42 ng/mL, p = 0.19), and GSH (1.64 vs. 1.72 mmol/L, p = 0.2). The mean age was significantly different between the genotype 1 and genotype 3 groups (43 vs. 38 years, respectively, p = 0.002). There was no difference between males and females in BMI, HOMA-IR, MDA, and age.
Table 1

Demographic, biochemical, and histopathological features of 115 patients with CHC

Variable

Total (n = 115)

Genotype 1 (n = 68)

Genotype 3 (n = 47)

p value

Male/female

81/34 (70.4 %/29.6 %)

46/22 (67.6 %/32.4 %)

35/12 (76 %/24 %)

0.28

Age (years)

41 (8.9)

43.1 (9.19)

38 (7.73)

0.002

BMI (kg/m2)

26.33 (4.72)

26.76 (4.83)

25.7 (4.55)

0.23

WHR

0.9 (0.07)

0.9 (0.08)

0.9 (0.07)

0.46

Virologic parameters

   

0.5

 Low viral load (≤ 8.5 × 105) IU/mL

42 (37.5 %)

25 (36.8 %)

17 (38.6 %)

 

 High viral load (> 8.5 × 105) IU/mL

70 (62.5 %)

43 (63.2 %)

27 (61.4 %)

 

Treatment condition

   

0.035

 Naïve

97 (84.4 %)

53 (77.9 %)

44 (93.6 %)

 

 IFN-non-responder

18 (15.6 %)

15 (22.1 %)

3 (6.4 %)

 

Plasma glucose (mmol/L)

5.39 (0.97)

5.54 (1.15)

5.18 (0.6)

0.03

Serum cholesterol (mmol/L)

4.43 (0.93)

4.75 (0.86)

3.99 (0.84)

<0.001

Serum triglyceride (mmol/L)

1.15 (0.63)

1.19 (0.65)

1.1 (0.61)

0.43

Ferritin (mg/mL)

274 (235)

269.6 (230.9)

280.7 (243.9)

0.81

HOMA-IR

3.08 (2.03)

3.27 (1.97)

2.79 (2.1)

0.19

ALT (U/L)

119 (94)

105.24 (77.7)

139.43 (113.3)

0.08

GGT (U/L)

67 (67)

64 (51.8)

73.1 (85.4)

0.52

Malondialdehyde (MDA) ng/mL

12.69 (1.86)

12.88 (1.94)

12.42 (1.71)

0.2

GSH (mmol/L)

1.68 (0.34)

1.64 (0.34)

1.72 (0.35)

0.2

Vitamin C (μmol/L)

35.7 (24.4)

38.6 (26.1)

31.1 (21.2)

0.13

Vitamin E (μmol/L)

22.1 (7.7)

23.8 (8.2)

21.1 (6.4)

0.08

Fibrosis stage

   

0.114

 0

15 (13 %)

11 (16.2 %)

4 (8.5 %)

 

 1

39 (33.9 %)

26 (38.2 %)

13 (27.7 %)

 

 2

38 (33 %)

17 (25 %)

21 (44.7 %)

 

 3

16 (13.9 %)

12 (17.6 %)

4 (8.5 %)

 

 4

7 (6.1 %)

2 (2.9 %)

5 (10.6 %)

 

Portal activity

   

0.26

 0

3 (2.6 %)

1 (1.5 %)

2 (4.3 %)

 

 1

21 (18.3 %)

15 (22.1 %)

6 (12.8 %)

 

 2

67 (58.3 %)

40 (58.8 %)

27 (57.4 %)

 

 3

21 (18.3 %)

11 (16.2 %)

10 (21.3 %)

 

 4

3 (2.6 %)

1 (1.5 %)

2 (4.3 %)

 

Steatosis grade

   

0.036

 0

43 (37.4 %)

30 (44.1 %)

13 (27.7 %)

 

 1

52 (45.2 %)

30 (44.1 %)

22 (46.8 %)

 

 2

16 (13.9 %)

5 (7.4 %)

11 (23.4 %)

 

 3

4 (3.5 %)

3 (4.4 %)

1 (2.1 %)

 

Results are reported as frequency (percentage) or mean (SD) as appropriate

HOMA-IR homeostasis model assessment of insulin resistance, BMI body mass index, WHR waist-hip ratio

Factors associated with HOMA-IR on univariate and multivariate analysis

By univariate analysis, the factors positively associated with HOMA-IR in both genotypes were oxidative stress as measured by MDA, BMI, portal activity, and fibrosis (Table 2). There was no association between vitamin C or vitamin E levels and IR. Genotype-specific differences in HOMA-IR association were for steatosis (positive association only in genotype 1), TG (positively associated only in genotype 1), age (positive association only in genotype 3), and GSH (negative association only in genotype 3). Because cirrhosis is known to cause IR, we repeated the analysis in those with fibrosis stage ≤2. The results were unchanged (for MDA in genotype 1: r = 0.491, p < 0.001 and in genotype 3: r = 0.346, p = 0.036), confirming that advanced disease was not the cause of the association between HOMA-IR and MDA levels. To determine those factors independently associated with IR in hepatitis C, we entered variables with p value <0.2 in a stepwise multiple linear regression analysis. MDA levels independently predicted IR in both genotypes, explaining approximately 17.7 and 11.9 % of the change in HOMA-IR for genotypes 1 and 3, respectively, even when controlled for BMI and fibrosis stage. As expected, BMI and fibrosis stage were also strong independent predictors of IR, explaining 23.5 and 24.1 % (BMI) and 9.4 and 17.8 % (liver fibrosis) of HOMA-IR variation in genotypes 1 and 3, respectively.
Table 2

Univariate and multivariate analysis of factors significantly associated with HOMA-IR

Factor

Univariate genotype 1

Univariate genotype 3

Multivariate genotype 1

Multivariate genotype 3

r

p

r

p

β

p

β

p

MDA (ng/ml)

0.454

<0.001

0.44

0.002

0.37

<0.001

0.229

0.04

GSH (mmol/L)

0.05

0.69

−0.402

0.006

0.13

BMI (kg/m2)

0.552

<0.001

0.657

<0.001

0.455

<0.001

0.436

<0.001

TG

0.344

0.004

0.196

0.186

0.062

0.51

Fibrosis stage

0.36

0.003

0.485

0.001

0.241

0.011

0.365

0.02

Steatosis grade

0.392

0.001

0.038

0.8

0.074

Portal activity

0.328

0.006

0.397

0.006

0.254

0.87

Age

0.14

0.256

0.482

0.001

0.71

Vitamin C (μmol/L)

−0.115

0.396

0

0.998

Vitamin E (μmol/L)

−0.048

0.723

−0.009

0.957

HOMA-IR homeostasis model assessment of insulin resistance, GSH glutathione, MDA malondialdehyde, BMI body mass index, TG triglycerides

In multivariate analysis that included all patients (n = 115), the results were unaltered (p < 0.001 for BMI, MDA, and fibrosis). To determine if the association between MDA and IR was genotype dependent, we used an interaction term between genotype and MDA in the multivariate model. This demonstrated that MDA levels were significantly associated with HOMA-IR irrespective of virus genotype (p value for interaction term = 0.226).

To further verify the association between IR and MDA, we sub-divided patients based on HOMA-IR values greater than or less than 1.64 as per published guidelines [23]. In univariate analysis of the insulin-resistant group (n = 85), MDA, steatosis, portal activity, and fibrosis were positively associated with HOMA-IR. This relationship was absent in insulin-sensitive subjects. On multivariate analysis in insulin-resistant subjects, BMI (p < 0.001), MDA (p = 0.006), and fibrosis (p < 0.001) were independent predictors of HOMA-IR, but again, this relationship was absent in insulin-sensitive subjects. Thereafter, we compared MDA and GSH between these two groups. Our results indicated that in insulin-resistant subjects, MDA concentration was higher [mean difference (MD) = 1.606, p < 0.001] and reduced GSH was lower (MD = −1.625, p = 0.023) than in the insulin-sensitive cohort (Table 3).
Table 3

Comparison of univariate predictors of IR in CHC subjects

Factor

HOMA-IR ≥ 1.64 (n = 85)

HOMA-IR < 1.64 (n = 30)

r

p

r

p

MDA, ng/ml

0.288

0.008

0.061

0.75

Steatosis grade

0.259

0.017

−0.148

0.435

Portal activity

0.28

0.01

0.075

0.693

Fibrosis stage

0.381

<0.001

0.053

0.781

HOMA-IR homeostasis model assessment of insulin resistance, MDA malondialdehyde

Factors associated with MDA levels

We finally sought to determine factors independently associated with elevated MDA levels. As shown in Table 4, by univariate analysis in both genotypes, apart from the described association of MDA with HOMA-IR, MDA was positively associated with GGT. In genotype 1 patients, steatosis grade and increasing age were strongly associated with MDA, but these were not important in genotype 3 where MDA was positively associated to BMI, portal activity, and vitamin E levels and negatively associated to GSH. In stepwise multivariate analysis, HOMA-IR was an independent predictor of MDA levels for both genotypes, whereas age (p = 0.001) was exclusively for genotype 1, and GGT (p = 0.005) and vitamin E (p = 0.002) for genotype 3. No significant difference in MDA levels was observed between the two groups (in genotype 1: mean MDA = 12.88 vs. 12.42 ng/mL in genotype 3, p = 0.197).
Table 4

Univariate analysis of factors associated with MDA

Factor

Genotype 1 (n = 67)

Genotype 3 (n = 48)

Multivariate genotype 1

Multivariate genotype 3

r

p

r

p

β

p

β

p

HOMA-IR

0.454

<0.001

0.44

0.002

0.43

<0.001

0.38

0.003

GSH (mmol/L)

−0.027

0.831

−0.419

0.004

0.63

BMI (kg/m2)

0.085

0.492

0.322

0.027

0.65

Fibrosis stage

0.204

0.095

0.247

0.095

0.575

0.679

Steatosis grade

0.282

0.02

−0.055

0.712

0.789

Portal activity

0.121

0.327

0.379

0.009

0.83

Age (years)

0.422

<0.001

0.259

0.078

0.39

0.001

0.75

GGT (U/L)

0.344

0.004

0.612

<0.001

0.82

0.37

0.005

Vitamin C (μmol/L)

0.22

0.1

−0.256

0.127

0.139

0.97

Vitamin E (μmol/L)

0.54

0.082

0.443

0.006

0.766

0.41

0.002

MDA malondialdehyde, HOMA-IR homeostasis model assessment of insulin resistance, GSH glutathione, BMI body mass index

Discussion

In the present study, we demonstrate that markers of oxidative stress are present in the serum of patients with CHC, and that there is no significant difference in the levels of oxidative stress between subjects with genotypes 1 or 3 infection. In addition, we show that MDA levels (a marker of oxidative stress) are higher and GSH levels are lower in insulin-resistant subjects compared to their insulin-sensitive counterparts. MDA levels in serum were a strong, independent predictor for IR in both genotypes 1- and 3-infected CHC subjects. Finally and as previously demonstrated, IR was independently associated with liver fibrosis in both genotypes [31].

In the present study, GSH demonstrated a significant negative correlation with IR and MDA only in genotype 3 CHC. However, interpretation of GSH is confounded by artifact, principally relating to oxidation during sample preparation [32]. This could bias the utility of GSH to predict plasma antioxidant status as previously reported [33]. In contrast, MDA serum levels have been suggested to be a reliable marker of oxidative stress particularly when assessed by HPLC as done herein [34].

While IR has a well-described association with CHC, there is scant published data on the underlying pathophysiology [31, 35, 36]. Oxidative stress may be critically and causally involved in the pathogenesis of primary IR [19], but few studies have examined this relationship in CHC. The present study is the first to confirm an important genotype-independent association between oxidative stress and IR and additionally to provide data to suggest that oxidative stress is present only when IR develops. Vidali et al. [6] reported an association between IR and oxidative stress (as measured by MDA) in a cohort of 107 patients, but only in non-3 genotype subjects (n = 89) and not in those with genotype 3 (n = 18). They suggested that oxidative stress may be responsible for the steatosis seen in patients with non-3 genotype. Mitsuyoshi et al. [16] and Oliveira et al. [37] studied and demonstrated a relationship between oxidative stress and IR in non-3 genotype-infected patients. In the present study, we analyzed patients of genotypes 1 and 3 separately and demonstrated similar associations and results. The association persisted even in those with early disease (fibrosis ≤2), and most importantly, only in those with IR and not in those with hepatitis C but without IR. Our data suggest that more robust inflammatory responses to the virus may induce higher levels of oxidative stress that could lead to both IR, fibrosis progression and a poor response to treatment. Our hypothesis is supported by other reports that HCV can directly induce oxidative stress intracellularly in hepatocytes [38, 39]. HCV core protein expression has been associated with increased ROS, decreased intracellular and/or mitochondrial GSH content, and increased levels of oxidized thioredoxin and lipid peroxidation products [5]. Of the HCV non-structural proteins, NS3 and NS5A may act as key mediators in the induction of oxidative stress and inflammation [40, 41]. Thus, NS5A-induced mitochondrial ROS production has been suggested to activate nuclear factor-κβ (NF-κβ) and to enhance suppressor of cytokine signaling-3 (SOCS-3), which is a well-established mediator of IR [1, 42]. Moreover, oxidative stress mediates signals involving the p38 mitogen-activated protein kinase pathway, resulting in the activation of NF-κβ. This transcription factor plays a key role in modulating the expression of cytokines such as tumor necrosis factor-α and interleukin 6—both established mediators of IR[1, 40, 43].

Recent studies have revealed a strong positive association between IR and MDA concentration in patients with impaired glucose regulation and type 2 diabetes in the absence of diseases such as CHC [12, 33]. Furthermore, higher levels of MDA have been reported in patients with the metabolic syndrome in comparison to healthy controls [44]. Likewise, higher levels of MDA have been reported in patients with nonalcoholic fatty liver disease [45] and in the polycystic ovary syndrome, which is another manifestation of IR [46, 47]. Together, these data suggest that irrespective of etiology, oxidative stress results in relative IR and that this may play a role in the IR observed in patients with CHC [17, 18].

One of the limitations of the present study is that we cannot exclude a non-hepatic source for serum MDA elevations in viral hepatitis. However, previous studies in CHC have demonstrated that MDA adducts are present in liver tissue and are correlated with the extent of fibrosis [4]. Further, the correlation between serum MDA levels and liver fibrosis, GGT, liver steatosis, and portal activity score, suggests that the liver is the principal contributor to oxidative stress in our patients.

In conclusion, the present work indicates that oxidative stress is present in CHC patients as assessed by serum MDA and is an independent predictor of IR in HCV genotypes 1 and 3. Moreover, MDA levels were higher and GSH levels were lower in insulin-resistant subjects compared to insulin-sensitive subjects. This suggests that in CHC infection, oxidative stress and IR are closely related and that the former may be causative for the IR observed in these patients.

Conflict of interest

The authors had financial support from Robert W. Storr Bequest to the Sydney Medical Foundation and National Health and Medical Research Council (NHMRC). The authors have no conflict of interest to disclose.

Copyright information

© Asian Pacific Association for the Study of the Liver 2012