Selection of functional variants of the NS3-NS4A protease of hepatitis C virus by using chimeric Sindbis viruses

TLDR: Using chimeric viruses to select and characterize active variants of the NS3-NS4A protease revealed that all of the mutated proteases still efficiently processed the chimeric polyprotein in infected cells and also cleaved an HCV substrate in vitro.

Abstract: The major etiological agent of non-A, non-B hepatitis was identified in 1989 and named hepatitis C virus (HCV) (8, 23). Presently, it is estimated that approximately 1% of the human population is infected by HCV (42). Exposure to HCV results in an overt acute disease in only a small percentage of cases, while in most instances the virus establishes a chronic infection which causes liver inflammation and slowly progresses to liver failure and cirrhosis (24). In addition, seroepidemiological surveys have indicated an important role of HCV in the pathogenesis of hepatocellular carcinoma (27). The absence of a protective vaccine and the limited efficacy of alpha interferon treatment (55) have raised considerable interest in developing alternative anti-HCV therapies. The genetic organization of HCV is similar to that of flaviviruses and pestiviruses (9, 37), and therefore HCV was assigned to a separate genus of the family Flaviviridae (43). The HCV genome consists of a single-stranded RNA of about 9.5 kb in length encoding a precursor polyprotein of 3,010 to 3,033 amino acids (8, 9, 26, 50). Individual viral proteins are produced by proteolysis of the precursor: the putative structural proteins (C, E1, E2, and p7) span the amino-terminal third of the precursor and are generated by cleavages probably mediated by the endoplasmic reticulum signal peptidase (21, 44), and the remaining part of the precursor contains the nonstructural proteins (NS2, NS3, NS4A, NS4B, NS5A, and NS5B), which presumably form the virus replication machinery and are released from the nascent precursor by two virus-encoded proteases. A zinc-dependent protease associated with NS2 and the N terminus of NS3 is responsible for the cleavage between NS2 and NS3 (16, 19, 39). A distinct serine protease located in the N-terminal domain of NS3 is responsible for proteolytic cleavages at the NS3/NS4A, NS4A/NS4B, NS4B/NS5A, and NS5A/NS5B junctions (3, 17, 52). Substantial efforts have been devoted to the characterization of the HCV serine protease, which is contained within the amino-terminal 180 amino acids of NS3 (3, 13, 17, 20, 52). Although the NS3 protease domain possesses enzymatic activity, the 54-amino-acid NS4A protein is required for cleavage at the NS3/NS4A and NS4B/NS5A sites and increases cleavage efficiency at the NS4A/NS4B and NS5A/NS5B junctions (2, 14, 33, 51). The central domain of NS4A, encompassing amino acids 21 to 32, was shown to be sufficient for activation of the protease (30, 34, 46, 53). In transfected cells, NS3 and NS4A assemble into a stable heterodimeric complex whose formation requires both the amino-terminal and the central domains of NS4A, as well as about 30 amino acids at the amino terminus of NS3 (4, 14, 30, 34, 45, 51). The determination of the crystal structure of the NS3 protease domain uncomplexed and complexed with central domain of NS4A (29, 36, 56) has confirmed the characteristics of this enzyme predicted by molecular modelling and biochemical studies (12, 40). The enzyme adopts a chymotrypsin-like fold and features a tetrahedrally coordinated metal distal to the active site. The central domain of NS4A forms a β strand which contributes to the formation of an eight-stranded β barrel with the amino-terminal domain of NS3 and plays a significant role in stabilizing NS3. Thus, NS4A is considered an integral structural component of the enzyme. For this reason, we here refer to the HCV serine protease as the NS3-NS4A protease. This protease cleaves the viral polyprotein in a precise temporal order which is probably critical for virus replication: the NS3/NS4A cleavage is the first event and occurs only in cis, and this is followed by cleavage at the NS5A/NS5B, NS4A/NS4B, and NS4B/NS5A sites, which can also occur in trans (2, 14, 33). An additional peculiar feature of the protease domain of NS3 is that it is covalently attached to an RNA helicase possessing ATPase activity (18, 25, 28). This overwhelming amount of data makes the NS3 protease an attractive candidate for developing effective HCV therapies. Indeed, several in vitro assay systems have been developed and are being used for the identification of specific inhibitors. Protease inhibitors have proved to be good therapeutic agents in the case of the human immunodeficiency virus protease. However, the long-term clinical efficacy of these drugs is potentially limited by the existence of inhibitor-resistant protease variants which are found in untreated subjects and emerge both in vivo during treatment and during selection in culture (10, 22, 32). Apparently, the ability of the virus to produce inhibitor-resistant protease variants depends largely on the ability of the protease to tolerate substitutions in critical subsites. Thus, attempts to subvert viral resistance should take this feature in account and concentrate on the search for inhibitors active against a broad range of variants. These attempts greatly benefit from the possibility of using in vitro cell culture systems for the selection and characterization of virus variants with decreased sensitivity to inhibitors. Sequence analysis of several HCV isolates indicates that there are multiple HCV genotypes and subtypes and that even in the same individual the virus exists as quasispecies (6, 47). Accordingly, a number of sequence differences are found in the portion of the genome encoding the NS3-NS4A protease. Only a few variants have been characterized biochemically, and these show similar kinetic parameters. On the other hand, the lack of an efficient in vitro infection system prevents a large-scale comparison of the different protease variants present in various HCV isolates and also precludes testing the sensitivities to inhibitors of the different variants in cell culture. We have recently described the generation of stable Sindbis virus (SBV)-HCV chimeric viruses whose propagation depends on the activity of the serine protease of HCV (15). Here we report the use of these viruses as a genetic system for the identification of functional variants of the NS3-NS4A protease which could be used to identify and characterize protease variants with decreased sensitivity to inhibitors. Since there are no selective inhibitors of the NS3-NS4A protease presently available, we investigated whether variants of the NS3-NS4A protease could be selected in the absence of a specific selective pressure, taking advantage of the high rate of spontaneous mutations of the SBV replication machinery. We selected more-infectious virus mutants by serial passaging on BHK cells, and almost all of them produced an NS3-NS4A protease different from those encoded by the original chimeras. All of these mutant enzymes displayed a measurable activity when assayed in vitro and efficiently processed the chimeric polyprotein in infected cells. These results imply that HCV-SBV chimeric viruses can be used under the appropriate conditions of selective pressure as a surrogate system for the identification and characterization of inhibitor-resistant variants of the NS3-NS4A protease.