Inhibition of HCV by the serpin antithrombin III
© Asmal et al.; licensee BioMed Central Ltd. 2012
Received: 28 February 2012
Accepted: 27 September 2012
Published: 2 October 2012
Skip to main content
© Asmal et al.; licensee BioMed Central Ltd. 2012
Received: 28 February 2012
Accepted: 27 September 2012
Published: 2 October 2012
Although there have been dramatic strides made recently in the treatment of chronic hepatitis C virus infection, interferon-α based therapy remains challenging for certain populations, including those with unfavorable IL28B genotypes, psychiatric co-morbidity, HIV co-infection, and decompensated liver disease. We have recently shown that ATIII, a serine protease inhibitor (serpin), has broad antiviral properties.
We now show that ATIII is capable of inhibiting HCV in the OR6 replicon model at micromolar concentrations. At a mechanistic level using gene-expression arrays, we found that ATIII treatment down-regulated multiple host cell signal transduction factors involved in the pathogenesis of cirrhosis and hepatocellular carcinoma, including Jun, Myc and BMP2. Using a protein interactive network analysis we found that changes in gene-expression caused by ATIII were dependent on three nodes previously implicated in HCV disease progression or HCV replication: NFκB, P38 MAPK, and ERK1/2.
Our findings suggest that ATIII stimulates a novel innate antiviral host cell defense different from current treatment options.
Antiviral treatment of hepatitis C virus (HCV) is aimed at persistent eradication of the virus, as measured in sustained virological response (SVR). SVR rates are high with current treatment options, a combination of peg-interferon-apha (IFN-α)–ribavirin and direct-acting antiviral agent (DAA) but HCV patients infected with HIV and/or other co-morbidities might benefit less from the new treatment options .
HCV infection is currently one of the most clinically relevant co-morbidities in the HIV population; it affects 15–30% of the 1 million HIV-positive patients in the US . Furthermore, progression to end-stage liver disease occurs six times faster in co-infected patients [3–6], with decompensated cirrhosis being one of the primary causes of hospitalization and death in this population [4, 6–8]. Treatment of HCV infection in HIV-positive patients has a lower success rate [9–13] and an increased risk of hepatotoxicity from antiretroviral drugs compared to HIV-negative patients [14–16]. Because of an increased risk of life-threatening complications during peg-IFN-α–ribavirin therapy, patients with hepatic decompensation are not typically candidates for this therapy unless easy access to orthotopic liver transplantation is available [17, 18]. Furthermore, since the antiviral activity of IFN-α is mediated, at least in part, through the cytokine network, immunological abnormalities, such as those that often result from HIV infection, reduce IFN-α efficacy. This loss of efficacy has the result of higher treatment failure in HIV/HCV coinfected patients when compared to HCV monoinfected patients [19–22].
The effective treatment of HIV in persons with advanced cirrhosis may be challenging due to cirrhosis-induced alterations in the hepatic metabolism of antiretroviral drugs and the potential for increased risk of drug-induced liver injury. To prevent possible liver toxicity, drug doses may be reduced and certain otherwise preferred drugs may be avoided [23, 24]. Reducing antiretroviral doses and hence plasma concentrations, however, may also lower the barrier to the emergence of drug-resistant HIV. Thus, effective therapy to eliminate HCV is necessary to optimize therapy for HIV.
We are investigating here antithrombin III (ATIII), a member of the serine protease inhibitor protein family (serpins), as its anti-inflammatory and anticoagulant activities were found to decrease liver damage . Serpins participate in the early innate immune response to viral infection  and they simultaneously possess broad-spectrum anti-viral and anti-inflammatory capabilities. In the case of HIV infection, serpins reportedly interfere with viral replication at both the entry and the reverse transcription stages [27–30]. In particular, the serpins alpha-1 anti-trypsin, the secretory leukocyte protease inhibitor and ATIII have significant in vitro ability to inhibit HIV-1, with the latter, ATIII, being the most potent [29–34].
In HCV infections with co-morbidities new drugs with different mechanisms of action other than the DAAs are urgently needed. We hypothesized that the broad immunomodulatory and anti-viral properties of ATIII might extend to other chronic viral infections due to a different mechanism of action, in particular, since a serpin receptor, the LDL receptor-related protein 1 (LRP1), is highly expressed on hepatocytes  and was found to block HCV infection .
Therefore, we undertook an investigation of whether ATIII has the potential to inhibit HCV replication in vitro. We used gene-arrays to probe the molecular mechanisms underlying ATIII’s immunomodulatory and antiviral properties, and uncover the signal transduction pathways that result in inhibition of viral replication.
IFN-α is currently part of the standard therapy for chronic HCV infection, in addition to ribavirin and an NS3-4A protease inhibitor. In certain patient subpopulations, this regimen is not always effective or is poorly tolerated. We have previously reported that the serpin ATIII has potent anti-viral activity against HIV [33, 34]. We sought to determine whether ATIII might also have activity against HCV since serpin receptors are highly expressed on hepatocytes . We employed the OR6 replicon system  expressing full-length genotype 1b virus to assess whether ATIII is capable of inhibiting HCV [38, 39]. Although heparin activation augments the anti-HIV activity of ATIII we used unmodified ATIII because heparin activation also increases the off-target effects of ATIII on thrombin. Unmodified ATIII has a demonstrated favorable toxicity profile and has been used in humans for more than 20 years.
For comparison, we assessed the ability of IFN-α2 monotherapy to inhibit the replicon. We tested doses of 4 and 16 IU IFN-α2, and found 71.4±10.1% and 84.4±8.4% inhibition of HCV, respectively. These results are similar to what has been reported previously . We next sought to determine whether ATIII and IFN-α might have an additive effect on HCV replication. We treated replicon cells with 7, 17 and 58 μM ATIII and with 4 and 16 IU/ml IFN-α2 (Figure 1B/C). We observed an additive effect, as treatment with ATIII significantly decreased HCV replication compared to IFN-α2 monotherapy (P-value of between <0.05 and <0.01). This additive effect was already observed at the lowest dose (7 μM) of ATIII tested (Figure 1). We performed similar experiments using IFN-α5, a different subtype of IFN-α, and confirmed the additive effects of ATIII observed with IFN-α2 (data not shown).
To exclude the possibility that the antiviral effect of ATIII was due to a cytotoxic effect, we assayed for cytotoxicity using Neutral Red and Trypan Blue exclusion staining at the indicated concentrations of drugs. Neither ATIII alone or in combination with IFN-α2 or IFN-α5 showed a cytotoxic effect (data not shown).
To assess the effect of ATIII treatment on host cell gene expression in the absence of HCV protein expression, we treated Huh7.5. non-replicon cells with the highest concentration of ATIII that would be used in the subsequent gene array experiments: 24 U/ml (58 μM), which is 24-fold the physiologic concentration. We found no significant alterations in expression of genes in the array following ATIII treatment of the non-OR6 replicon cells, demonstrating that, at these concentrations and in the absence of HCV replication, ATIII has no significant effect on expression of our transcriptional pathways of interest. Using Trypan Blue exclusion staining we also found no drug-related cytotoxicity at the concentrations used (data not shown).
To assess the effect of HCV replication on hepatocyte physiology we compared the transcriptional profile of HCV replicon cells to that of Huh7.5 non-replicon cells using the Transduction Pathfinder RT2 Profiler PCR Assay. Initially, experiments were performed in the absence of exogenous ATIII. We observed substantial differences in the transcription of multiple genes involved in the innate host cell response between cells expressing and not-expressing HCV (Tab. I). Many of these HCV-induced changes have been previously described elsewhere [42–49] confirming the validity of our system. The gene with the greatest increase in expression was Matrix Metallopeptidase 10 (MMP10) (49-fold, P = 0.006), a key mediator in the Jak-Stat pathway and part of the inflammatory response of the host cell against HCV. Additional effects of HCV on host gene expression include activation of p53 and TGF-β pathways: we found a 3-fold (P = 0.03) increase of Cyclin-dependent kinase inhibitor (CDKN) 1A and a 10-fold (P = 0.005) increase of CDKN2B. Other observations from this study that are consistent with previously described associations with HCV include findings of a 9-fold (P = 0.04) increase of Bone morphogenetic protein 4 (BMP4), part of the hedgehog pathway, and a 4-fold (P = 0.03) increase in Heat Shock Protein (HSP) 90AA2, part of the cellular stress response.
Determination of ATIII treatment effect on gene expression in the context of HCV
Replicon/Non-replicon Gene expression
Replicon/Replicon+7 μM ATIII Gene expression
Bone morphogenetic protein 4
Cyclin-dependent kinase inhibitor 1A
Cyclin-dependent kinase inhibitor 2B
Cytochrome P450, family 19, subfamily A, polypeptide 1
Fatty acid synthase
Heat shock 27kDa protein 1
Heat shock protein 90kDa alpha class A member 2
Interleukin 4 receptor
Matrix metallopeptidase 10
Ornithine decarboxylase 1
Retinol binding protein 1
WNT1 inducible signaling pathway protein 1
At higher concentrations of ATIII, we found only a modest effect on HCV-induced transcriptional changes (data not shown). There was no ATIII dose-dependent effect on expression of any of the genes in Table I. These results suggest that the mechanism by which ATIII inhibits HCV within 48 h may not involve modulation of the genes influenced by HCV infection.
HCV infection often leads to chronic hepatitis, cirrhosis, and occasionally to hepatocellular carcinoma . This progression in liver pathology is associated with increased expression in hepatocytes of the transcription factors JUN and MYC, which may play crucial roles in oncogenesis [51, 52]. In order to investigate the influence of ATIII on pathways important for HCV disease progression we employed the Transduction Pathfinder RT2 Profiler PCR Assay to quantify the expression of 84 key genes belonging to 18 different regulatory pathways in the presence of different concentrations of ATIII.
To investigate whether the therapeutic use of ATIII might have an influence on gene-expression in OR6 replicon cells, we treated these cells with supra-physiologic concentration of ATIII: 2.4-fold (7 μM), 7-fold (17 μM) and 24-fold (58 μM) blood concentrations. We used supra-physiologic doses of ATIII in part because ATIII is known to accumulate in the liver - a fact which may be of therapeutic advantage.
The next largest decrease in gene expression, up to 346-fold (P = 0.009 at 58 μM), was observed for the transcription factor CAAT/enhancer binding protein (CEBPB), a protein regulated by insulin. Another gene downstream of insulin Hexokinase 2 (HK2), was down-regulated up to 14-fold (P = 0.0039). Growth arrest and DNA-damage-inducible protein (GADD45A), a gene in the p53 pathway, was down-regulated 35-fold at 58 μM (P = 0.0023). Bone morphogenetic protein 2 (BMP2), a gene of the Hedgehog pathway, was down-regulated 13-fold (P = 0.03) at 58 μM. B-cell CLL/lymphoma2-like 1 (BCL2L), a transcript belonging to the Jak/Src pathway, exhibited an approximately 10-fold decrease in expression (Figure 2). Down-regulation of these genes was specific to ATIII treated OR6 cells with ongoing HCV replication, and was not observed in the untreated OR6 replicon, nor in the ATIII-treated Huh7.5 controls suggesting that ATIII induces a specific anti-viral gene program.
When IFN-α was administered in combination with ATIII, additional genes were significantly altered, potentially explaining the additive antiviral effect of ATIII when added to IFN-α treatment. The most substantially down-regulated gene was BMP2, belonging to the Hedgehog pathway, which was decreased by 37-fold (P=0.04). JUN and PTGS2 both belonging to the Phospholipase C pathway were 14-fold (P=0.08) and 9-fold (P=0.02) down-regulated. CEBPB of the insulin pathway was 8-fold down-regulated (Figure 3).
We repeated these experiments using IFN-α5, to exclude the possibility that our results may have been idiosyncratic to IFN-α2. We observed the same gene expression pattern with IFN-α5 treatment, with or without ATIII treatment (data not shown).
To gain further insight into the mechanism of action of ATIII in reducing HCV replication, we performed a biologic network analysis of ATIII treated OR6 replicons. This analysis method complements data generated from our gene arrays by facilitating the recognition of hierarchical gene clusters that might intersect with HCV replication. This software-supported interactome analysis is based on a vast library of gene interactions known to regulate certain pathways and allows to identify pathways effected by a drug.
We have demonstrated additive activity of IFN-α and ATIII in inhibiting HCV. We thus next sought to determine whether they might exhibit overlapping effects on the HCV interactome. We compared the effect of IFN-α/low ATIII dose therapy to that of IFN-α alone on HCV induced nodules. Treatment with 4 IU/ml IFN-α alone altered three HCV induced nodules: P38 MAPK, MAPK and NFκB (Figure 4D). The addition of low dose ATIII did not alter the number of nodules affected, but did increase the IRF1 response (Figure 4E) which is known to be responsible for some of the HCV inhibition and might be the reason for the additive effect on IFN-α treatment.
The cornerstone of effective therapy for chronic hepatitis C infection has been IFN-α, a critical mediator in the innate immune response to viral infection. Even with the advent of small molecule direct inhibitors of viral enzymatic activity, IFN-α remains important for attaining sustained virologic response, perhaps because of the need to engage host antiviral programs to completely eradicate viral reservoirs. However, interferon-based therapy is not without its shortcomings, including poor tolerability or poor efficacy in certain patient populations.
We now demonstrate that a novel activator of host innate antiviral responses, ATIII, may give insight into adjunctive therapies for HCV that might augment or even replace IFN-α in cases where there are co-morbidities or genetic factors that preclude the use of IFN-α. There is circumstantial evidence that ATIII may play a role in the pathogenesis of HCV infection: low plasma concentrations of ATIII have previously been correlated with progression to chronic hepatitis and cirrhosis . Furthermore, there is a high density of serpin receptors on hepatocytes, suggesting that serpins may have localized effects on hepatic innate immunity .
We used the OR6 replicon to probe how ATIII might influence HCV pathogenesis. We demonstrated that ATIII inhibited HCV replication at micromolar concentrations. While this inhibition was not as potent as that of either IFN-α or fluvastatin  it was many-fold greater than that of ribavirin .
We next investigated the mechanism of ATIII’s anti-HCV activity. After more than 20 years of research, the mechanism of action of IFN-α in inhibiting HCV has only recently been determined. HCV cell-based expression models, as the one used in this study, were used to demonstrate that IFN-α-induced signal transduction through the Jak/STAT pathway was necessary for HCV inhibition . To elucidate the mechanism(s) through which serpins activate the host defense system, we employed the OR6 replicon system, and analyzed changes in gene-expression patterns of 84 key genes representative of 18 different signal transduction pathways. We found that ATIII treatment down-regulated JUN expression. It has previously been shown that the JNK inhibitor, SP600125, interferes with the oncogenic potential of HCV non-structural protein 3 . In addition, we found that ATIII treatment reduced induction of the transcription factor MYC, expression of which has been associated with progression of liver disease to chronic hepatitis, cirrhosis, and hepatocellular carcinoma [52, 59]. ATIII treatment also reduced CEBPB, a transcription factor of the CCAAT/enhancer-binding protein class. These transcription factor proteins have been shown to be important for HCV inhibition: the CCAAT/enhancer-binding protein-homologous protein (CHOP) is activated by HCV envelope protein and is associated with disease progression . The mechanism of action of this protein class is not fully understood, but CHOP can sensitize cells to apoptosis by down-regulation of BCL-2 expression, depletion of cellular glutathione, and exaggerated production of reactive oxygen species .
Our data also identified a substantial down-regulation of BMP2, a protein that regulates hepcidin . Hepcidin is critical for iron homeostasis and is also a critical host cofactor involved in promoting HCV replication . Chronic HCV infection has been found to be associated with increased serum iron and transferrin saturation as well as hepatic iron accumulation; moreover, hepatic hepcidin mRNA expression is increased in patients with chronic HCV infection . Hepcidin transcription is stimulated by iron overload as well as by inflammation through IL-6 , which is elevated in patients with chronic HCV. The identification of hepcidin as a HCV replication cofactor suggests a molecular basis for the well-known clinical association between chronic HCV infection and dysregulation of iron homeostasis. Moreover, it is possible that the up-regulation of hepcidin transcription by IL-6 potentially creates a positive feedback loop between chronic inflammation and HCV replication. Together, these findings suggest that ATIII therapy may reduce the pathogenic impact of HCV infection.
Thus, our data indicate that ATIII targets several genes that are known to promote both liver disease and HCV replication. ATIII treatment may therefore alter the expression of these genes and act to simultaneously slow both HCV replication and ultimately liver degeneration. ATIII’s effect on gene-expression was also observed when replicon cells were co-treated with low concentrations of IFN-α. It was during this dual drug treatment that gene expression of BMP2, CEBPB, and JUN were most dramatically down-regulated.
Protein interactive network analysis demonstrated that the genes that were altered by ATIII treatment were dependent on three nodes: NFκB, P38 MAPK and the ERKs. All of these nodes have been described previously as having a role in HCV replication and HCV-related liver disease [47, 66–69] confirming ATIII’s potential to limit HCV’s destruction of the liver. These nodules are also affected in ATIII-mediated inhibition of HIV .
Although our replicon model can facilitate identification of substances that affect either viral genomic replication or host cell factors involved in viral genomic replication, it cannot be used to identify substances that alter other stages of the viral life cycle. Therefore, future studies using fully infectious, cell-culture-adapted HCV strains will be required to study other aspects of the HCV life cycle, such as viral entry, uncoating, viron assembly and secretion. Our data identified several genes altered by ATIII that were previously shown to be correlated with HCV disease outcome. This might explain the additive therapeutic effect when ATIII was used in combination with IFN-α. We further found that ATIII’s mechanism of action is most likely multi-faceted, warranting further research into each distinct signaling pathway.
OR6 replicon cells were a gift from Dr. Nobuyuki Kato (Okayama University, Japan) and were propagated in Dulbecco’s Modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) supplemented with 1% penicillin-streptomycin, and 500 μg/ml Geneticin (Invitrogen Corp., Carlsbad, CA). Cells were cultured in a 37°C, 5% CO2-humidified incubator for all experiments. To decrease day-to-day variability in the assay, a large homogenous population of subconfluent cells was passaged so that a similar lot of cells could be used throughout the assay.
Clinical grade human ATIII (Talecris, Durham, NC) had a concentration of 6 U/mg and a purity >98%. For ATIII drug combination experiments, recombinant human IFN-α2 and IFN-α5 (PBL Interferon Source, Piscataway, NJ) was used, which had a concentration of 2.38 x 108 and 2.33 x 108 units/mg, respectively, and a purity of > 98%.
HCV replication inhibition was determined as the percentage of luciferase activity retained by the OR6 replicon after ATIII treatment, compared to a vehicle-treated control . Luciferase activity was measured using the Renilla Luciferase Assay System (Promega, Madison, WI) at 1 sec. intervals using the 1420 Multilabel Counter Victor 3 (PerkinElmer, Waltham, MA). Cell viability was assessed using Neutral Red staining and Tryphan Blue exclusion staining.
OR6 replicon cells were harvested after 48 h of ATIII and/or IFN-α treatment and total RNA was recovered using the RNeasy Kit (Qiagen) with an on-column DNAse digest (Qiagen) according to the manufacturer’s protocol. Approximately 100 ng RNA was used for cDNA synthesis using the SuperArray RT2 First Strand Kit (SABiosciences, Frederick, MD, cat. no. C-03). cDNA was used for the RT2 Profiler PCR Array Human Signal Transduction PathwayFinder (SABiosciences, cat. no. PAHS-014A). The genes that were investigated can be found at http://www.sabiosciences.com/rt_pcr_product/HTML/PAHS-014A.html. Three arrays of three independent experiments were performed for each treatment condition. Relative levels of transcription were determined by using the ΔCt values for each gene obtained by subtracting the mean threshold cycle (Ct) of the GAPDH housekeeping gene from the Ct value of the gene of interest. The average ΔCt value for 3 experiments was calculated, for each gene of interest, and the average normalized transcription was calculated as follows: 2 (−averageΔCt)-1. Fold increases of gene transcription, before and after treatment was calculated by dividing the average normalized transcription of each gene in the test sample by the corresponding control. Statistical significance in up- or down-regulation of transcription was determined by Student T-test (2-sample, equal variance, 2-tailed distribution), comparing the ΔΔCt (ΔΔCt = ΔCt treated - ΔΔCt control). Significant differences were identified when P was less than 0.05.
Functional analysis of interacting proteins was determined using a commercial System Biology package, Ingenuity Pathways Analysis (IPA 8.0) ( http://www.ingenuity.com) following the application protocols.
The statistical significance of differences between groups was determined using the program GraphPad Prism. A P value of <0.05 was considered statistically significant. Statistical analysis was performed using one table T-test or the unpaired T-test. Error bars represent standard error of the mean (S. E.).
We thank Dr. Nobuyuki Kato (Okayama University, Japan) for his kind gift of the OR6 replicon cell-line. We thank Christa Osuna-Gutierrez (Beth Israel Deaconess Medical Center) for critical reading of the manuscript. This study was funded by NIH and the Department of Defense (N01 AI30048, N01 AI30049, AI067854 and AI060354-06). We thank Dr. Carmen Plasencia (Acceleromics, Spain) for help with the systems biology approach.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.