The NTPase/helicase domain of hepatitis C virus nonstructural protein 3 inhibits protein kinase C independently of its NTPase activity

Helicase motif VI is a short arginine-rich motif within the NTPase/helicase domain of the non-structural protein 3 (NS3) of the hepatitis C virus (HCV). We previously demonstrated that it reduces the catalytic activity and intracellular shuttling of protein kinase C (PKC). Thus, NS3-mediated PKC inhibition may be involved in HCV-associated hepatocellular carcinoma (HCC). In this study, we expand on our earlier results, which were obtained in experiments with short fragments of NS3, to show for the first time that the catalytically active, longer C-terminal NTPase/helicase of NS3 acts as a potent PKC inhibitor in vitro. PKC inhibition assays with the NTPase-inactive mutant NS3h-D1316A revealed a mixed type kinetic inhibition pattern. A broad range of 11 PKC isotypes was tested and all of the PKC isotypes were inhibited with IC50-values in the low micromolar range. These findings were confirmed for the wild-type NTPase/helicase domain in a non-radiometric PKC inhibition assay with ATP regeneration to rule out any effect of ATP hydrolysis caused by its NTPase activity. PKCα was inhibited with a micromolar IC50 in this assay, which compares well with our result for NS3h-D1316A (IC50 = 0.7 μM). In summary, these results confirm that catalytically active NS3 NTPase/helicase can act in an analogous manner to shorter NS3 fragments as a pseudosubstrate inhibitor of PKC.


INTRODUCTION
Chronic hepatitis C virus (HCV) infection is a major risk factor for the development of hepatocellular carcinoma (HCC) [1]. In previous studies, we showed that within its NTPase/helicase domain, NS3 contains an arginine-rich sequence motif consisting of residues 1487 to 1500 of HCV polyprotein, also known as helicase motif VI [2]. This motif strongly resembles the pseudosubstrate sequence within the auto-regulatory domain of protein kinase C (PKC) [3,4] and it interacts with the active site of PKC, thereby inhibiting the catalytic activity and function of this enzyme [5,6]. It has also been shown that fragments of HCV-NS3 that contain this motif can inhibit the catalytic activity of cAMP-dependent protein kinase (PKA) and slow down its cytoplasm-tonucleus translocation [2,7]. Similar interactions with protein kinases are well documented for several synthetic peptides that mimic the phosphorylation site sequences of the natural protein substrates or pseudosubstrate auto-regulatory regions of kinases [8,9]. However, our previous studies on the interaction between HCV-NS3 and PKC were only conducted with smaller, catalytically inactive fragments of NS3. This may not have accurately reflected the situation in vivo. Furthermore, other groups have argued that the inhibition of PKA and possibly PKC by HCV-NS3 was rather due to the effect of NS3-4A-catalyzed ATP hydrolysis [10]. Here, we address the question whether the full-length NS3 NTPase/helicase domain can inhibit PKC independently of its NTPase activity. In this study, we substantiate the results of our previous studies [5,6] and overcome one of their substantiate limitations: the use of smaller nonphysiological NS3 fragments. Moreover, we determine the PKC isotype specificity and the mode of inhibition for NS3 NTPase/helicase-mediated inhibition of PKC. We show that all PKC isotypes are inhibited by NS3 with IC 50 values in the low micromolar range independently of the NS3 NTPase activity.
Our findings could be physiologically relevant because members of the PKC family play pivotal roles in cellular proliferation, apoptosis and cell survival, and there is a substantial amount of evidence linking disturbance of PKC to tumorigenesis [11]. It is plausible that NS3-mediated PKC inhibition may be involved in the molecular pathogenesis of HCV infection, particularly in the development of HCV-associated HCC.

Cloning, bacterial expression and purification of NS3h and NS3hD1316A proteins
The point mutation D1316A was introduced into pET21b-NS3HCV (provided by Dr. Choe [12]) via QuikChange mutagenesis (Stratagene, USA). Coding sequences for NS3h-D1316A and wt-NS3h were cloned into pET102/D-TOPO (Invitrogen, USA) to be expressed as N-terminal Thioredoxin and C-terminal Strep-Tag II fusions. The coding sequence for the Strep-Tag II (WSHPQFEK) followed by a stop codon was introduced to the 3' end of the coding sequence via the reverse primer. Recombinant proteins were expressed in E. coli BL21(DE3) as previously described [4], purified from sonicated bacterial pellets using Strep-Tactine sepharose affinity chromatography (IBA GmbH, Germany), and concentrated using Amicon Ultra centrifugal filter units (Millipore, USA).

Radiometric protein kinase C assay
The protein kinase C activity and kinetic parameters were determined as previously described [6].

Expression and purification of NS3h and NS3h-D1316A
To investigate the in vitro interaction between the enzymatically active C-terminal NS3 NTPase/helicase and PKC isotypes, we used the previously well-characterized C-terminal NTPase/helicase (HCV polyprotein amino acids 1193-1658) of HCV-1 [12], expressed as N-terminal thioredoxin and C-terminal Strep-Tag II fusion protein. In addition to the wt-protein (NS3h), we generated an NTPase/helicase-deficient point mutant (NS3h-D1316A) [10,13]. Recombinant proteins were purified by Strep-tag affinity chromatography as described in the Materials and Methods section. , illustrating the positions of the NS3 NTPase/helicase subdomains and the arginine-rich stretch. The ribbon diagram was rendered using POV-Ray [15] based on the coordinates PDB 1HEI [16]. Subdomains 1, 2 and 3 are respectively red, blue and turquoise. C -The purity of the recombinant proteins was verified using SDS-PAGE. A representative 10% SDS PAGE gel of NS3h after staining with Coomassie Brilliant Blue is shown. Fig. 1A and B depict the proteins used in this study. The obtained NS3 polypeptide preparations were confirmed as homogenous using SDS-PAGE (Fig. 1C). The purified NS3h protein exhibited an intrinsic ATPase activity that was strongly stimulated by the presence of poly(U) ( Fig. 2A, B). These characteristics concur with those described in the literature [13,14] and suggest a native conformation for NS3h. As expected, NS3h-D1316A did not exert any NTPase activity at the concentrations tested ( Fig. 2A).

NS3h-D1316A and NS3h are potent inhibitors of PKC in vitro
The influence of NS3 constructs on PKC activity was determined in vitro by quantifying the phosphate incorporation into the PKC substrate histone HIIIS in the absence or presence of increasing concentrations of the recombinant HCV NS3 proteins. Since ATP to ADP conversion potentially interferes with PKC activity measurement in standard radiometric PKC assays by reducing the specific radioactivity of [γ-32 P]-ATP [10] we also tested the NTPase inactive mutant NS3h-D1316A for its capability to inhibit PKC. Fig. 3 shows the inhibition of PKCα by wt-NS3h and the NTPase-inactive mutant NS3h-D1316A in the radiometric PKC assay. The wt-protein (NS3h) inhibits the activity of PKCα with an apparent IC 50 of 0.05 µM in this assay, which is in part due to ATP hydrolysis (see below). The IC 50 for NS3h-D1316A was determined to be 0.7 µM. IC 50 values for the inhibition of all 11 PKC isotypes by NS3h-D1316A were determined. All of the PKC isotypes were potently inhibited with IC 50 values that fall within the low micromolar range (Table 1).  We also used a non-radiometric PKC assay that contained an ATP regeneration system (phosphocreatine and creatine phosphokinase [17]) to keep the ATP concentration constant during the course of the reaction in the presence of NS3h (Fig. 4). For this purpose, a commercial assay (PepTag, Promega) with a fluorescently labeled peptide PKC substrate was used, as previously described by Aoubala et al. [10]. As expected, PKC activity was completely abrogated in the presence of 1.2 µM NS3h when the ATP regeneration system was absent (lane 1). When the full ATP regeneration system was included, the presence of 1.2 µM NS3h still caused a marked reduction in PKC activity (> 50%; lane 4), which compares well with the results for NS3h-D1316A and PKCα in the radiometric assay (IC 50 = 0.7µM). These results demonstrate that NS3h inhibits the enzymatic activity of PKC independently of ATP hydrolysis in a similar range to NS3h-D1316A. NS3h-D1316A acts as a mixed-type inhibitor of rat brain PKC Finally, we aimed to determine the mode of inhibition for NS3 NTPase/helicasemediated inhibition of PKC. We measured the inhibition of rat brain PKC by NS3h-D1316A at multiple concentrations of NS3 and the PKC substrate histone HIIIS. We then subjected the data to graphical analysis according to the methods described by Dixon and Cornish-Bowden [18,19]. The resulting plots (Fig. 5) revealed that NS3h-D1316A acts as a mixed-type inhibitor of rat brain PKC. K i (the dissociation constant for the enzyme-inhibitor complex) was 0.5 µM and K i ' (the dissociation constant for the enzyme-substrate-inhibitor complex) was 1.5 µM. The magnitude of K i is considerably smaller than K i ', which suggests that the observed PKC inhibition is primarily due to specific interactions between NS3h-D1316A and the active sites of rat brain PKC.

DISCUSSION
The aim of this follow-up study was to resolve the question whether the fulllength NTPase/helicase domain of HCV-NS3 can act similarly to its smaller fragments as an inhibitor of PKC in vitro. We also wished to expand on our previous findings with an additional analysis of the PKC isotype specificity of this interaction and the mode of action for NS3 NTPase/helicase-mediated inhibition of PKC. Most importantly, we can confirm that the enzymatic activity of PKC is indeed potently inhibited by the catalytically active C-terminal helicase portion of NS3 in vitro. The observed PKC inhibition is independent of the NTPase activity of NS3, as demonstrated by experiments with an NTPasedeficient point mutant. Additionally, a PKC activity assay containing an ATP-regenerating system was employed to confirm these results for wild-type NS3-NTPase/helicase and consequently rule out the influence of the D1316A point mutation on the accessibility of the arginine-rich stretch. Our data are in line with the results of our earlier studies [3,5,6,20]. In those studies smaller NS3 fragments comprising the arginine-rich stretch [5,6,20], and notably the short peptide HCV-polyprotein (1496-1500) representing only the pseudosubstrate-like sequence [3], were all found to bind and inhibit PKC with IC 50 values in the same order of magnitude. Together with our new finding that NS3h-D1316A acts as a mixed-type inhibitor of rat brain PKC, these novel results strongly suggest that NS3-NTPase/helicase-mediated inhibition of PKC is at least in part due to pseudosubstrate inhibition and follows a complex mechanism similar to the PKC inhibition by the shorter fragments analyzed in our earlier studies conducted [5,20]. Further studies should focus on experimentally elucidating the exact mechanism of the complex interaction between PKC, its substrate and the inhibiting viral protein.
For most PKC isotypes, the IC 50 value is within a similar range as determined earlier for the isolated subdomain 2 of NS3 NTPase/helicase [6]. PKC isotypes γ, δ, η and ζ are even more potently inhibited by NS3h-D1316A than by the isolated subdomain 2 with a roughly tenfold higher IC 50 , suggesting additional interactions between these PKC isotypes, NS3 and/or the substrate histone HIIIS (as discussed in [20] [21]. It is intriguing to hypothesize that HCV may have evolved the capacity to disturb the activity of conventional PKCs (α, βΙ, βΙΙ and γ) as a mechanism to evade the IFN-mediated cellular response. Notably, Chen et al. showed in a recent study that hepatitis B virus (HBV) uses a similar evasion mechanism that involves impairment of IFN-α-induced STAT activation through inhibition of PKCδ by HBV polymerase [22]. There are several examples of viral proteins that interfere with PKC-mediated signal transduction to affect host cell functions [23][24][25].
We have demonstrated for the first time that the longer, catalytically active NS3-NTPase/helicase can act as a potent PKC inhibitor in vitro. All PKC isotypes are inhibited by NS3-NTPase/helicase to a varying degree, including the isotypes α and δ, for which inhibition or downregulation has been linked to carcinogenesis [11,26]. Whether the determined inhibitory effects of NS3 NTPase/helicase on PKC are potent enough to be relevant in vivo will mainly depend on the exact local intracellular concentration of NS3 in infected cells, which is not known and difficult to assess. However, considering that HCV NS3 has been shown to localize to the endoplasmic reticulum (ER) or ER-like compartments when coexpressed with NS4A [27], high subcellular concentrations of NS3 in ERcompartments are conceivable. In future studies, it will be important to investigate the influence of HCV-NS3 on the activity of cellular PKC and related kinases in vivo in relevant cell culture systems harboring the viral protein, to clarify the physiological relevance of our findings.