Elevated Serum Levels of Interferon-γ-Inducible Protein-10 in Patients Coinfected with Hepatitis C Virus and HIV

As a consequence of shared routes of transmission, some 16%– 33% of individuals infected with HIV are coinfected with hepatitis C virus (HCV); in intravenous drug users, the coinfection rate has been reported to range from 50% to 90% [1]. Chronic HCV infection progresses slowly in monoinfected individuals, with ∼20% of patients developing cirrhosis after 20 years of infection [2]. In contrast, HCV/HIV coinfection has been associated with a faster progression to liver disease and with the development of cirrhosis and hepatocellular carcinoma [3].

Cytokines, chemokines, and their receptors play an important role in the pathogenesis of both HCV and HIV infection. Previous studies have reported increased serum and intrahepatic levels of the 10-kDa interferon (IFN)-γ-inducible protein (IP-10) in HCV genotype 1-infected individuals [4, 5]. Related studies have found elevated IP-10 levels to be associated with increased liver damage [6], and it has also been shown that serum IP-10 concentrations are higher in nonresponders to HCV therapy than in those who achieve a sustained virological response [7]. Elevated IP-10 levels have also been reported in HIV-monoinfected patients, with nonresponders to highly active antiretroviral therapy (HAART) having higher IP-10 levels than treatment responders [8].

To date, there have been limited reports on cytokine and chemokine expression in HCV/HIV-coinfected individuals. The aim of the present study was to compare serum cytokine and chemokine levels between HCV/HIV-coinfected and HCVmonoinfected patients and to investigate the possible association between cytokine/chemokine expression and liver damage in HCV/HIV coinfection.

Subjects, materials, and methods. Four groups were enrolled: HCV/HIV-coinfected patients (n = 15), HCV-monoinfected patients (n = 18), HIV-monoinfected patients (n = 17), and healthy control subjects (n = 19). HCV infection was defined as the presence of serum HCV antibodies and detectable viral RNA; HIV infection was defined as seropositivity for HIV. Healthy control subjects were negative for both HCV and HIV antibodies. All HCV-positive patients were infected with genotype 1. Ten of the 15 HCV/HIV-coinfected patients and 12 of the 17 HIV-monoinfected patients were receiving antiretroviral treatment. None of the HCV-positive patients had received IFN therapy at the time of the study. All experiments were conducted with the approval of the Institutional Review Board of St. James's Hospital.

HCV and HIV loads were measured using branched, DNAbased Versant assays (Bayer Diagnostics). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured using commercial kits. Liver fibrosis scores were evaluated according to the method of Ishak et al. [9].

Six chemokines (IP-10, macrophage inflammatory protein [MIP]-1α, MIP-1β, RANTES, monocyte chemoattractant protein [MCP]-1, and interleukin [IL]-8) and 6 cytokines (IL-2, IL-4, IL-6, IL-10, IL-12 [p70], and IFN-γ) were measured in the serum of all participants. All levels except for RANTES were measured using a multibead human cytokine LincoPlex kit; RANTES levels were measured using a single-plex RANTES human cytokine LincoPlex kit (both from Linco Research). Multianalyte profiling was performed using a Luminex-100 system and XY platform (Luminex).

All statistical analyses were performed using Statview software (version 5.0.1; SAS Institute). Study cohorts were compared using the Mann-Whitney U test. Fisher's exact test was used for categorized data. Correlations between parameters were estimated using Spearman's correlation coefficient (Ur). P < .05 was considered to indicate statistical significance.

Results. The demographics and laboratory parameters of the 3 groups of infected patients are shown in table 1. In all, 6 serum cytokine (IL-2, IL-4, IL-6, IL-10, IL-12, and IFN-γ) and 6 chemokine (MIP-1α, MIP-1β, RANTES, IP-10, MCP-1, and IL-8) levels were analyzed. IP-10 levels were higher in all 3 infected groups than in healthy control subjects (P < .0001) (figure 1A). HCV/HIV-coinfected patients had higher IP-10 levels than both HCV-monoinfected (P = .015) and HIV-monoinfected (P = .012) patients. There was no significant difference in IP-10 levels between HCV-monoinfected and HIVmonoinfected patients. Correlations between IP-10 levels and (1) HCV load, (2) liver enzyme levels, (3) liver fibrosis scores, and (4) CD4 cell counts in the HCV/HIV-coinfected and HCVmonoinfected patients were investigated. It could be shown that increased ALT and AST levels were positively correlated with IP-10 levels in the HCV-monoinfected group (for ALT levels, r = .668 and P = .006; for AST levels, r = .822 and P = .0007). Patients with elevated ALT levels (>40 IU/L) had higher IP-10 levels than those with normal levels (7–35 IU/L) (P = .021) (figure 1B, subpanel a). An increase in IP-10 levels was also evident in HCV-monoinfected patients who had ALT levels >60 IU/L (P = .001) and 1100 IU/L (P = .003) (data not shown). Furthermore, HCV-monoinfected patients with elevated AST levels (140 IU/L) had higher IP-10 levels than those who had normal AST levels (7–40 IU/L) (P = .001) (figure 1B, subpanel b). This increase in IP-10 levels was also evident in HCV-monoinfected patients with AST levels 160 IU/L (P = .0005) and >100 IU/L (P = .009) (data not shown). In contrast, in HCV/HIV-coinfected patients, there was a strong negative correlation between IP-10 and ALT levels (r = -.609; P = .028) and no association between IP-10 and AST levels. HCV/ HIV-coinfected patients with elevated ALT levels had lower IP- 10 levels than those with normal ALT levels (P = .034) (figure 1B, subpanel c).

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Figure 1.

A, Serum levels of the 10-kDa interferon-γ-inducible protein (IP-10) in healthy control subjects, HIV-monoinfected patients, hepatitis C virus (HCV)-monoinfected patients, and HCV/HIV-coinfected patients (*statistically significant difference vs. healthy control subjects; †statistically significant difference vs. HIV-monoinfected patients; ‡statistically significant difference vs. HCV-monoinfected patients). B, Correlation between IP-10 levels and alanine aminotransferase (ALT) levels in HCV-monoinfected subjects (a), aspartate aminotransferase (AST) levels in HCV-monoinfected subjects (b), ALT levels in HCV/HIV-coinfected patients (c), liver fibrosis scores in HCV-monoinfected patients (d), liver fibrosis scores in HCV/HIVcoinfected patients (e), and CD4 cell counts in HCV/HIV-coinfected patients (f). Results are expressed as box plots; each box represents the 25th to the 75th percentile, the outer bars represent the 5th and 95th percentiles, and the line inside each box represents the median. Groups were compared using the Mann-Whitney U test.

A strong positive correlation between liver fibrosis and IP- 10 levels was evident in HCV-monoinfected patients (r = .605; P = .013). HCV-monoinfected patients with a greater degree of liver fibrosis (score ⩾4) had higher IP-10 levels than those with a liver fibrosis score <4 (P = .006) (figure 1B, subpanel d). This correlation was not evident in HCV/HIV-coinfected patients (figure 1B, subpanel e). A strong negative correlation was found between IP-10 levels and CD4 cell count in HCV/HIV-coinfected patients (P = .035). HCV/HIV-coinfected patients with low CD4 cell counts (<400 cells/µL) appeared to have higher serum IP-10 levels than those with higher CD4 cell counts (figure 1B, subpanel f). This correlation was not observed in HIV-monoinfected patients; similarly, no association was found between IP-10 levels and ALT or AST levels. There was no difference between IP-10 levels and HAART status in either of the groups of HIV-infected patients, and no association was observed between HCV load and IP-10 levels in either of the groups of HCV-infected patients.

We also measured levels of CCR5-binding chemokines (table 2). All 3 groups of infected patients had lower MIP-1α levels than healthy control subjects. HCV-monoinfected patients had MIP-1β levels comparable to that of healthy control subjects. HIV-monoinfected patients had higher RANTES levels than both healthy control subjects (P = .041) and HCV-monoinfected patients (P = .004). HCV/HIV-coinfected patients had higher RANTES levels than HCV-monoinfected patients (P = .027).

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 1.

Summary of patient characteristics.

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 2.

Levels of Th1 cytokines, Th2 cytokines, CCR5-binding chemokines, monocyte chemoattractant protein (MCP)-1, and interleukin (IL)-8.

Three Th1 cytokines (IL-2, IFN-γ, and IL-12) and 3 Th2 cytokines (IL-4, IL-6, and IL-10) were quantified (table 2). There was no obvious trend toward a Th1 or Th2 response in any of the 3 groups of infected patients, compared with healthy control subjects. HCV/HIV-coinfected patients had higher MCP-1 levels thanHIV-monoinfected patients (P = .03) (table 2). All 3 groups of infected patients had lower IL-8 levels than healthy control subjects (P < .05). IL-8 levels were further decreased in HIVmonoinfected patients, compared with those in HCV-monoinfected (P = .027) and HCV/HIV-coinfected (P = .038) patients.

Discussion. Chemokines and cytokines play an important role in the pathogenesis of both HCV and HIV infection. There is increasing evidence that the pathological changes observed in the livers of HCV-infected patients are a direct result of the intrahepatic antiviral immune response [10]. The role played by cytokines and chemokines in HCV/HIV coinfection has not been well defined, and, to our knowledge, this is the first detailed study of cytokine and chemokine expression in HCV/HIV-coinfected patients with HCV genotype 1 infection.

Previous reports have shown that increased peripheral and intrahepatic IP-10 levels are associated with increased liver damage [6]. Our results are consistent with these findings; we found that, in HCV-monoinfected patients, increased serum IP-10 levels are associated with increased ALT and AST levels and with increased liver fibrosis. These results strongly indicate that IP-10 may play an important role in the progression of liver fibrosis in patients with chronic HCV infection. We also found elevated IP-10 levels in HIV-monoinfected patients, compared with those in healthy control subjects. Previous studies have reported elevated IP-10 levels in HIV-monoinfected individuals [8, 11] and have shown that IP-10 promotes HIV replication [12]. These reports have also shown that nonresponders to HAART have higher IP-10 levels than do treatment responders [11]. Thus, IP-10 clearly plays an important role in the pathogenesis of both HCV and HIV infection. To date, there have been no documented reports on the role played by IP-10 in HCV/HIV-coinfected patients. We found that IP-10 levels were significantly higher in HCV/HIV-coinfected patients than in monoinfected and uninfected control groups. Given that elevated IP-10 levels in HCV-monoinfected patients are correlated with increased liver disease, the increased IP-10 levels in HCV/HIV-coinfected patients may partly explain why there is an accelerated progression to liver disease in these patients. However, we did not find a positive correlation between IP-10 levels and either liver enzyme levels or liver fibrosis in our HCV/HIV-coinfected population. There are 2 possible explanations. First, the majority of HCV/HIV-coinfected patients in our study were receiving HAART, and previous studies have shown that HAART can slow the progression of liver disease [13]. Second, HAART has been shown to decrease levels of IP- 10 in HIV-infected patients [11]. Therefore, it is possible that HAART both down-regulates IP-10 levels and slows liver damage.

We found a significant negative correlation between IP-10 levels and CD4 cell counts; HCV/HIV-coinfected patients with low CD4 cell counts (<400 cells/µL) had higher IP-10 levels than those with higher CD4 cell counts. In HCV infection, IP- 10 recruits CD4 lymphocytes to the liver as part of the antiviral immunological response. HCV-infected patients with concurrent HIV infection have a reduced number of CD4 cells. Therefore, it is possible that, because of the reduced number of CD4 cells, IP-10 expression is up-regulated in the liver in an attempt to recruit the required number of CD4 lymphocytes to respond to HCV infection and that, as a result, the level of liver damage is increased. It has recently been reported that IP-10 may have predictive value for the outcome of HCV treatment, with nonresponders to HCV treatment having significantly higher baseline IP-10 levels than do those who respond [7]. It was not possible to pursue this in our study, because our cohort did not contain a balanced number of treatment responders and nonresponders. Individuals coinfected with HCV and HIV have lower rates of response to HCV treatment than do HCV-monoinfected patients [14]. Future studies evaluating IP-10 levels in HCV/HIV-coinfected patients receiving HCV treatment might help determine whether the increased IP-10 levels in HCV/ HIV-coinfected patients contribute to the lower treatment response rates. It should be noted that, because this was a crosssectional study, it was not possible to determine whether elevated IP-10 levels are a cause or a consequence of progressive liver fibrosis. It should also be highlighted that, as with any study measuring serum levels of cytokines and/or chemokines, any additional unknown infection(s) could potentially influence the results. However, we have no evidence to suggest that this was the case in the present study. Indeed, we have at least shown that none of the patients had hepatitis B virus infection.

HCV/HIV-coinfected patients had higher levels of RANTES than HCV-monoinfected patients. RANTES plays an important role in recruitment of T cells to the liver; in addition, elevated intrahepatic RANTES levels have been found to be associated with an increase in liver damage [15]. Given that serum RANTES levels are elevated in HCV/HIV-coinfected patients, this could partly explain the accelerated progression to liver disease; however, further analysis is warranted.

In conclusion, the present study has provided further evidence that liver disease in HCV-infected patients is a direct result of specific immune responses to the virus. Our study has also confirmed previous observations that serum IP-10 levels are associated with increased liver damage in HCV-monoinfected patients and has provided new evidence implicating IP- 10 in the increased liver disease associated with HCV/HIV coinfection. These findings could have important implications for developments in the treatment and management of disease in HCV/HIV-coinfected patients.

• Received January 29, 2007.
• Accepted April 8, 2007.

• © 2007 by the Infectious Diseases Society of America