Identification of Residues in the Hepatitis C Viru
http://www.100md.com
病菌学杂志 2005年第11期
Department of Pathobiology, University of Washington
Department of Medicine, University of Washington, Seattle, Washington 98195
ABSTRACT
Significant advances have been made in understanding hepatitis C virus (HCV) replication through development of replicon systems. However, neither replicon systems nor standard cell culture systems support significant assembly of HCV capsids, leaving a large gap in our knowledge of HCV virion formation. Recently, we established a cell-free system in which over 60% of full-length HCV core protein synthesized de novo in cell extracts assembles into HCV capsids by biochemical and morphological criteria. Here we used mutational analysis to identify residues in HCV core that are important for capsid assembly in this highly reproducible cell-free system. We found that basic residues present in two clusters within the N-terminal 68 amino acids of HCV core played a critical role, while the uncharged linker domain between them was not. Furthermore, the aspartate at position 111, the region spanning amino acids 82 to 102, and three serines that are thought to be sites of phosphorylation do not appear to be critical for HCV capsid formation in this system. Mutation of prolines important for targeting of core to lipid droplets also failed to alter HCV capsid assembly in the cell-free system. In addition, wild-type HCV core did not rescue assembly-defective mutants. These data constitute the first systematic and quantitative analysis of the roles of specific residues and domains of HCV core in capsid formation.
INTRODUCTION
Hepatitis C virus (HCV) is a major public health concern worldwide. Two percent of the world's population is infected with the virus (54), which is the major cause of non-A, non-B hepatitis and which often leads to cirrhosis of the liver or hepatocellular carcinoma (42). While current treatments have improved, they are poorly tolerated, not completely efficacious, and not available in all settings (31). Development of better treatments and vaccines has been hampered by the lack of virus replication systems. Standard cell culture systems do not support HCV replication, and the chimpanzee is the only animal model capable of being infected and yielding high HCV titers (23). These limitations have been partly overcome by the development of HCV replicon systems, which support autonomous replication of HCV RNA (2, 10, 36, 37). Replicon systems have allowed specific requirements of HCV RNA replication to be studied; however, HCV particles, or even nucleocapsids, are not produced in these systems (9, 27, 53). Consistent with this observation, infectious particles are not released from these cells (27). As a result, the requirements for critical steps in virion formation, including capsid assembly, genome encapsidation, budding, and release, remain largely unknown.
HCV is an enveloped, single-stranded, positive-sense RNA virus in the Flaviviridae family (3). The HCV genome has a single open reading frame that codes for a 3,000-amino-acid polyprotein. The structural proteins, core, E1, and E2, are the N-terminal products in the polyprotein and are released from the polyprotein by host proteinases (34, 43, 57). Once cleaved from the polyprotein, the 173-amino-acid mature core protein assembles into HCV capsids at the cytoplasmic face of the endoplasmic reticulum (ER) (7, 8, 45). Core is known to interact with the HCV envelope glycoprotein E1 at the ER (35), and assembled capsids are thought to acquire their envelopes by budding into the ER (4, 7, 8, 35). However, the specific details of these late events in the viral life cycle have not been elucidated.
Besides forming the capsid that houses the HCV genomic RNA, core is also known to modulate diverse cellular functions. Core is carcinogenic when expressed in transgenic mice (47), has pro- and antiapoptotic functions (29), alters the transcription of other viral promoters (58, 59), and appears to induce steatosis (48) and the formation of lipid droplets (1). Core is known to interact directly with many different cellular proteins that most likely play a role in modulating these diverse functions (42) or may modulate the ability of core to assemble. Most interacting proteins have been identified through yeast two-hybrid approaches. All reported interactions of core with intracellular proteins, including the cytoplasmic domain of tumor necrosis factor (TNF) receptor 1 (65), lymphotoxin ? receptor (15, 41), hnRNP K (26), DDX3 (51), Cap-Rf (64), and PA28 (46), are mediated by the N-terminal two-thirds of core. The observation that no cellular proteins associate with the C terminus of core may reflect an artifact of the yeast two-hybrid assay (many screens have been performed with truncations of core, e.g., 41) or may be because the hydrophobic C terminus is less accessible to interactions with cellular proteins. However, the interaction of the HCV envelope glycoprotein E1 with core has been mapped to the C terminus (35, 39). Core is also known to target to lipid droplets via its C terminus, where it colocalizes with apolipoprotein AII (1). Deleting amino acids 153 to 169 or mutating the prolines at positions 138 and 143 abolishes lipid droplet association (25, 43). While the significance of core trafficking to lipid droplets remains unclear, this targeting event may have implications for replication or pathogenesis (42). Besides interacting with cellular proteins, core is known to self-associate. Amino acids from 82 to 102, referred to as a homotypic interacting domain, mediate core-core interactions in a yeast two-hybrid assay (50). Core is also highly basic and binds RNA (19, 57), and this association is dependent on the basic N terminus (57). Because interactions of core with cellular proteins can be studied in many cultured cell lines, much is known about these interactions, including which domains of core are required. However, due to the lack of systems for studying assembly in a quantitative manner, no systematic mutational analysis of core has been performed to define domains that are important for its ability to assemble into an HCV capsid.
We have recently established a cell-free system (CFS) that supports robust HCV core assembly and forms bona fide HCV capsids, as demonstrated by velocity sedimentation, buoyant density, and transmission electron microscopic analyses (28). Thus, HCV capsids produced in this system are nearly indistinguishable from authentic capsids isolated from the serum of infected patients (28). This system links de novo translation of core to HCV capsid assembly in eukaryotic extracts, recapitulating events beginning with synthesis of HCV core through the completion of capsid assembly. Additional advantages and features of this system are that it is amenable to quantitation, highly reproducible, can be manipulated, and supports synthesis of mutant core constructs. Previously we used this system to define characteristics and requirements of HCV assembly and showed the importance of the N-terminal RNA binding region for capsid assembly (28). Here we use this permissive HCV capsid assembly system to further characterize important residues in the N terminus and to examine domains elsewhere in the core protein that may be involved in capsid assembly. We find that clusters of basic charges are critical for HCV assembly and that certain uncharged regions in both the N terminus and the remainder of the protein are largely dispensable for HCV capsid assembly.
MATERIALS AND METHODS
DNA plasmids. Cell-free plasmid vectors were derived from SP64 vector (Promega) into which the 5' untranslated region (UTR) of Xenopus laevis globin had been inserted at the HindIII site (44). The parental plasmid used for this study contained the HCV core coding region from an HCV 1b isolate as described previously (28). Using the parental plasmid as the template, mutant core constructs were prepared by either site directed mutagenesis (Stratagene), or by two-step-PCR and cloned into the BglII and EcoRI sites of the parental plasmid (data not shown). All coding regions were verified by sequencing.
In vitro transcription and cell-free translation/assembly. In vitro transcription was performed using the SP6 polymerase (New England Biolabs) and the SP64 expression plasmids described above. Resulting transcripts (or transcription buffer for mock transcript, as indicated) were used to program cell-free translation and assembly reactions for 120 min, either using wheat germ extracts at 26°C or using rabbit reticulocyte lysate at 37°C. Reactions were radiolabeled using [35S]methionine (ICN Biochemicals), as described previously (18, 33, 52).
Gradient analysis of cell-free reactions and cellular lysate. Calibration of gradients to determine S value positions has been described previously (32, 33). Ten to 20 microliters of cell-free assembly reactions, diluted into 200 μl final volume containing 0.625% NP-40 detergent, 10 mM Tris-acetate, pH 7.4, 50 mM KAc, 100 mM NaCl, and 4 mM Mg acetate, were layered onto gradients containing sucrose prepared in the same buffer. Velocity sedimentation was performed on step gradients containing 400 μl each of 10%, 20%, 30%, and 40% sucrose with a 200-μl 50% sucrose cushion. Centrifugation of velocity sedimentation gradients was performed at 201,000 x g for 55 min at 4°C in a TLS55 rotor (Beckman Coulter Optima Max centrifuge), and 200-μl fractions were collected serially from the top. Equivalent aliquots of each gradient fraction were trichloroacetic acid (TCA) precipitated and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by autoradiography (AR).
Chymotrypsin digestions. Chymotrypsin digestions were done as previously reported (30). Either fractions 1 and 2 or fractions 6 and 7 from velocity sedimentation gradients were pooled and then added to chymotrypsin reactions. Chymotrypsin reactions were set up on ice and contained 300 μl of sample, 50 mM Tris, pH 8.0, 20 mM CaCl2, and 1 ng/μl chymotrypsin (Bovine Pancreas; Calbiochem) in a volume of 1.2 ml. At the indicated time points, 200 μl of the reaction were removed, quenched with TCA, TCA precipitated, and analyzed by SDS-PAGE and AR. The concentration of chymotrypsin was decreased 10-fold from the previous report (30) to compensate for greatly reduced amount of substrate in our samples.
Electron microscopy. Cell-free reactions were subjected to velocity sedimentation centrifugation, and fractions 6, 7, and 8 were pooled and dialyzed (molecular weight cutoff, 55,000) for 1 h against phosphate-buffered saline (PBS) at room temperature. Dialyzed samples were then settled onto carbon-coated grids (Ted Pella) for 2 to 3 min, stained with 1% uranyl acetate for 30 s, and visualized using a JEOL 1010 transmission electron microscope. An experienced electron microscopist identified and examined reactions and controls in single-blinded fashion in three separate experiments. Histograms were prepared by measuring the diameters of all capsids in three to six fields (containing 120 capsids). Criteria for excluding small particles that were not capsids have been described previously (28).
Quantitation. Autoradiographs were digitized using an AGFA Duoscan T1200 scanner and Adobe Photoshop 5.5 software (Adobe Systems Incorporated). Mean band densities were determined and adjusted for band size and background. Amount of assembly was quantitated as the amount of core present in fractions 6, 7, and 8.
Statistics. P values were obtained using a paired student t test (one-tailed).
RESULTS
Internal deletions within the N-terminal 42 amino acids diminish HCV capsid assembly. We previously established a CFS that supports HCV capsid assembly and demonstrated that approximately 60 to 80% of newly synthesized HCV core polypeptides assemble into 100S complexes that migrate in fractions 6 to 8 on our velocity sedimentation gradients. The 100S complexes correspond to HCV capsids by multiple criteria (28). Additionally, we showed that HCV capsids are heterogeneous in size as is seen in vivo, and the 100S cell-free capsids range from 75S to 120S upon analysis in higher resolution gradients (28). In our previous study, we demonstrated that both forms of full-length HCV core that are found in vivo (C191, which includes the E1 signal sequence, and the mature, signal-cleaved version of core, termed C173) assemble to an equivalent extent. Furthermore, there was no difference in assembly when wheat germ extract (WG) versus rabbit reticulocyte lysate (RRL) was utilized as a source of cellular factors for the CFS (28). Thus, unless otherwise noted, experiments in the current study were performed in the WG CFS using C173 as wild-type (WT) core, with % assembly defined as % of radiolabeled core polypeptides present in gradient fractions 6 to 8.
Additionally, in the previous study, we showed that the highly basic N-terminal 68 amino acids of HCV core contain residues important for HCV capsid assembly, since N-terminal truncations of HCV core altered assembly while truncations in the C-terminal half of core had no effect. We found that deleting the N-terminal 10 amino acids (N10) had no effect on cell-free assembly (79% assembly) but deleting the N-terminal 42 (N42) or 68 (N68) amino acids decreased assembly to 12% and 8%, respectively (28). Qualitatively this was seen in velocity sedimentation profiles in which WT C173 migrated largely in fractions 6 to 8, and N42 and N68 migrated at the top of the gradient in fractions 1 and 2, representing complexes of <10S (28).
Based on these observations, we wanted to further identify specific residues or motifs that are important for capsid assembly. Therefore, we generated a series of internal deletions within the N-terminal 42 amino acids (Fig. 1A). When total translations were analyzed by SDS-PAGE, we found that all the deletion mutants expressed to similar levels as WT C173 and migrated at their expected sizes (Fig. 1B). We then tested their ability to assemble into HCV capsids in the CFS by velocity sedimentation (Fig. 1C), as described previously (28). Because core is known to be extremely lipophilic, the velocity sedimentation gradients contained 0.625% NP-40 to solubilize nonspecific proteins aggregating with core as well as associating membranes, allowing the true sedimentation value of HCV core and assembled HCV capsids to be assessed. Deletion of 32 amino acids (designated 11-42) resulted in a statistically significant decrease to 21% assembly, compared to WT C173, which yielded 60% assembly. This was consistent with results obtained with the previously described N-terminal truncation mutants (28) and confirmed that residues required for assembly are present between amino acids 11 to 42. However, we found that smaller deletions that removed 11 to 22 residues within this region (11-21, 21-42, and 31-42) inhibited assembly to only a modest extent (Fig. 1C), suggesting that this region does not contain a single discrete motif that is critical for assembly.
Basic residues in the N terminus of core are critical for HCV capsid assembly. Given these findings, we hypothesized that residues important for assembly are likely to be dispersed between amino acids 11 to 42 and might function cooperatively. The N terminus contains two regions with a high density of basic amino acids linked by a neutral region of 15 amino acids (Fig. 2A). Therefore, we hypothesized that the two basic regions and possibly the neutral linking region are important for HCV assembly. To test this possibility, we constructed targeted deletions of the two basic clusters (1 and 2) and the neutral linker (L). The 1 construct (amino acids 8 to 25 deleted) removed 7 basic residues, while 2 (amino acids 39 to 64 deleted) removed 10 basic residues, and L (amino acids 27 to 38 deleted) removed no basic residues (Fig. 2A). The deletion mutants expressed to similar levels and migrated on SDS-PAGE at their expected sizes (Fig. 2B). When we analyzed their ability to assemble, we found that deleting either cluster of basic amino acids (1 or 2) diminished the ability of core to assemble to approximately 25%. Surprisingly, deleting the entire neutral region (L) between the basic clusters had no effect on HCV capsid assembly in the CFS (Fig. 2C and D).
Next, we directly assessed the contribution of the basic residues in the N terminus by substituting alanines for basic residues in the N terminus. Fifteen mutants encoding as few as one and as many as 16 basic residue substitutions in one or both basic clusters were constructed (Fig. 3A). We found that the ability of the mutant constructs to assemble in the CFS decreased as more basic residues were mutated (Fig. 3B and C). When only one basic residue was mutated (RK1A and RK1B, Fig. 3) there was little or no effect on assembly (60% assembly for each construct compared to 61% assembly for WT core). However, as the number of basic residues mutated increased, a progressive decrease in the ability of core to assemble was observed (i.e., 48% for RK2A and RK2B; 30% for RK3A and RK3B; 25% for RK4A and RK4B), suggesting that basic charge in this region is critical for capsid assembly. It should be noted that not all mutants expressed as well as WT (Fig. 3B, line graph), but this likely does not account for the lack of assembly because we have previously demonstrated that HCV assembly is minimally dependent on core concentration in the CFS (28). Consistent with this observation, when WT core expression was lowered to less than that of RK4B, were observed (data not shown). Additionally, some basic charge substitution mutants displayed equivalent amounts of translation relative to WT C173 but had significantly reduced levels of assembly (i.e., RK6, RK9, RK13, and RK16 in Fig. 3B). These findings are consistent with basic charge in this region being required for efficient capsid assembly.
Table 1 summarizes all the constructs encoding N-terminal mutations (truncations, deletions, and substitutions) that we have studied in the CFS and is organized by the number of basic residues that are altered. The amount of assembly for each construct is indicated as % of WT core assembly (with WT assembly set at 100%). This comparison reveals that removal or mutation of one or two basic residues was generally well tolerated, while further increasing the number of basic residues progressively diminished HCV capsid assembly. The apparent correlation between the number of basic residues present in the N terminus and the ability of HCV core to assemble suggests that these basic residues are an important determinant of HCV capsid assembly.
HCV core polypeptides encoding mutations form capsids that are stable. We repeatedly observed that core mutants displaying intermediate levels of assembly migrated differently in velocity sedimentation gradients compared to fully assembly-competent or assembly-incompetent constructs. In addition to being present in the assembled fractions, core polypeptides that assembled to intermediate levels were dispersed throughout the gradient (i.e., 1 and 2 in Fig. 2D) rather than solely forming peaks in specific fractions as was the case for assembly-competent (i.e., fractions 6 and 7 for WT C173 in Fig. 2D) or assembly-incompetent mutants (i.e., fractions 1 and 2 for N68 in Fig. 6B). This led us to ask whether constructs that assemble to intermediate levels form unstable complexes that give rise to the more dispersed migration pattern during the velocity sedimentation analysis. We examined stability by isolating capsids (in fractions 6 and 7) formed by C173 or core mutants as well as other complexes (in fractions 3 and 4) formed by core mutants that display a limited level of assembly. We then subjected these complexes to a second round of velocity sedimentation on the same type of gradients. If the capsids and/or complexes are inherently stable, they should migrate in the same fractions upon repeat velocity sedimentation. Capsids of WT C173 and two mutants (2 and RK7) were stable, since they migrated to fractions 6 and 7 on the second velocity sedimentation gradient (Fig. 4). RK7 did appear to spread out somewhat during the second velocity sedimentation analysis but mostly migrated in fractions 6 and 7. In contrast, when smaller complexes (i.e., fractions 3 and 4) of the two mutants 2 and RK7 were reanalyzed by velocity sedimentation, they migrated mostly to the top of the gradient (fraction 1, which contains monomers and complexes of <10S), suggesting that they lack stability. Thus, it appears that some mutant constructs that assemble to intermediate or low levels form at least two types of complexes: stable capsid-like complexes that migrate in the 100S assembled fractions and complexes that migrate in other regions of the gradient that are unstable.
Unassembled core polypeptides are sensitive to chymotrypsin, unlike assembled capsids. To further address whether mutant HCV core constructs assemble into capsids at a low level in the cell-free system, we modified a previously described assay that was used to demonstrate that assembled recombinant HCV core is resistant to chymotrypsin digestion, whereas unassembled recombinant HCV core is relatively susceptible to chymotrypsin digestion (30). If the mutant complexes in fractions 6 and 7 are intact capsids, they should be resistant to chymotrypsin digestion. To test this, we isolated 100S capsid like complexes, as well as unassembled HCV core, from cell-free reactions expressing WT or mutant constructs and subjected them to chymotrypsin digestion. Figure 5 shows that WT core from the assembled fractions 6 and 7 (right panels) was relatively resistant to chymotrypsin, while the very small amount of WT core present in the top two fractions (left panels) was protease sensitive. Both the protease sensitivity of unassembled HCV core and the protease resistance of assembled capsids in the CFS resembled the findings previously reported for unassembled and assembled recombinant HCV core (30). Similarly, despite different input amounts for different constructs, mutant core present in unassembled top fractions was relatively protease sensitive while core from assembled fractions 6 and 7 was relatively protease resistant for all constructs examined, including an assembly-competent mutant construct (RK2B), mutant constructs with basic amino acids mutated in either the first or second cluster (RK4B and RK6, respectively), and a construct containing a large deletion (2) (Fig. 5). Thus, the protease resistance of the small amounts of 100S complexes formed by mutant constructs supports the conclusion that these constructs undergo a limited amount of capsid assembly. Additionally, the examined protease sensitivity of the unassembled fractions suggests that the mutations are unlikely to grossly affect protein folding, although it is impossible to rule out very subtle effects on folding.
Wild-type core does not rescue assembly-defective mutants. In certain systems it is possible to rescue defective mutants by expressing a competent construct in trans. For example, budding-defective human immunodeficiency virus type 1 (HIV-1) constructs that are missing the p6 domain of the HIV-1 capsid protein can be complemented in trans in cells with constructs that encode regions of the p6 domain (40). In the case of HIV-1 capsid assembly, assembly-defective nonmyristoylated capsid precursor GagPol proteins can be rescued by coexpression of myristoylated assembly-competent capsid proteins (60). Conversely, mutant capsid proteins can interfere with the function of WT proteins by acting as dominant-negative mutants, as shown by studies of specific mutations in HIV-1 Gag (14). Together, these data indicate that in the case of HIV-1 capsid formation, capsid proteins produced from different transcripts interact during the assembly process. To test whether WT C173 rescues the assembly of assembly-defective constructs or whether core mutants act in a dominant-negative manner to inhibit assembly of WT C173, we programmed cell-free reactions with equivalent amounts of transcripts encoding WT and mutant constructs (N68, 11-42, or N2). These constructs migrate at different positions on SDS-PAGE from WT C173 (Fig. 6A) and therefore can be distinguished when expressed in the same reaction. When we analyzed their ability to assemble by velocity sedimentation, we found that WT C173 assembled to an equivalent extent in the absence or presence of mutant constructs, suggesting that none of the mutants examined behaved in a dominant-negative manner (compare panels in Fig. 6A to C173 panel in B; graphed in C). Moreover, assembly of the mutant constructs did not increase when coexpressed with WT, demonstrating that WT C173 does not complement the assembly-defective mutants (compare panels in Fig. 6A to three lower panels in B; graphed in D). Additionally, coexpressing WT and mutant constructs did not alter the distribution on velocity sedimentation gradients of the mutant core constructs examined, showing that formation or accumulation of other complexes was unaffected (compare Fig. 6A to B). Equivalent results were found when an assembly-competent mutant (82-102; see Fig. 7) was coexpressed with the mutants in Fig. 6 (data not shown).
Mutations in other regions of core do not alter HCV capsid assembly. Other groups have suggested that regions outside of the N terminus of core are important for HCV capsid assembly. Amino acids 82 to 102 were found to contain a homotypic interaction domain by a yeast two-hybrid assay that measures core-core interactions (50). A morphological study of capsid assembly done in BHK21 cells using a Semliki forest virus expression system implicated an aspartic acid at amino 111 as important for HCV assembly (8), although it was unclear to what extent WT core or the mutant assembled in this system. Additionally, other regions in the central region of core called domain 2 (42) are known to be important for targeting of the HCV core protein (24, 25, 43) but have not been examined with respect to capsid assembly. For example, two proline residues (at amino acids 138 and 143) form a proline knot motif that is important for targeting core to lipid droplets (25). Because core-core interactions and targeting are likely to be important for HCV capsid formation, we wanted to quantitatively examine what effect these mutations, shown in Fig. 7A, had on the ability of core to assemble in the cell-free system. In contrast to the assembly-defective N-terminal mutants shown previously, core constructs encoding 82-102, D111A, and mutations in prolines 138 and 143 assembled as well as WT C173 by velocity sedimentation (Fig. 7B and C). Similar results were obtained when D111A was examined using a cell-free system containing RRL rather than WG (Fig. 7B, compare right and left sides). To confirm these findings we also subjected fractions containing 100S capsids encoding the D111A and 82-102 mutations to transmission electron microsocopy (TEM; Fig. 8A to C). We had previously shown that capsids formed by WT C173 in the CFS closely resemble capsids obtained from patient serum (28). We found that the assembly-competent 82-102 and D111A mutants formed abundant numbers of capsids that resembled capsids produced by WT C173 in the CFS. Measurements revealed a bimodal distribution of diameters for capsids formed by the 82-102 mutant. This distribution closely resembled the distribution of diameters displayed by WT C173 capsids prepared in parallel (Fig. 8D). In contrast, capsids encoding the D111A mutation had larger diameters while maintaining a bimodal distribution of diameters (Fig. 8D). These data suggest that, at least in the CFS, these residues are not essential for HCV capsid assembly, although the aspartic acid at position 111 does appear to influence capsid size.
Phosphorylated serines are not essential for capsid assembly. The fact that cultured cells do not support HCV assembly suggests that this process can be regulated; however, this regulation has not been studied in detail. One common mechanism frequently used to regulate cellular processes and cellular proteins is phosphorylation. HCV core has been shown to be phosphorylated in vitro at serines at amino acids 53, 99, and 116 by protein kinase A and C (58) and are phosphorylated in human liver cell lines (38). These serines are highly conserved across different HCV genotypes (11), suggesting they may play an important role for HCV. Studies show that phosphorylation of core may regulate targeting of core to the nucleus (38), and the ability of HCV core to suppress HBV transcription (58). Whether these phosphorylated serine residues play a role in capsid assembly has not been addressed previously. To examine this, serines at 53, 99, and 116 were sequentially converted to alanines (Fig. 9A). Constructs encoding these mutations translated to an equivalent extent and assembled as well as WT C173 in both WG and RRL cell extracts (Fig. 9B and C). Thus, it appears that serines in core that are thought to undergo phosphorylation are not required for HCV capsid assembly in the cell-free system, even when a mammalian extract is used. Conversion of serine 53 to aspartic acid (S53D) to mimic the phosphorylated state also had no effect (Fig. 9B and C). The effect of conversion of other serines to aspartic acids has not yet been examined.
DISCUSSION
Here we have systematically analyzed regions of the HCV core protein that are important for capsid assembly using a recently described HCV cell-free system that allows for a quantitative analysis of assembly in a cellular context. Because it supports efficient capsid assembly (28), this system differs from almost all reported cellular systems in which HCV core fails to assemble into capsids. Because of the lack of systems for studying HCV capsid assembly, surrogate assays for assembly, such as core-core interactions in yeast two-hybrid assays and nonquantitative assays, have been used to define residues that appeared to be important for HCV assembly. It has been difficult to fully interpret results from these studies, leaving a gap in our understanding of which residues and domains are important for HCV assembly. Here we showed that the N-terminal basic region plays a key role in HCV assembly in the CFS. Interestingly, this region is also known to bind RNA, which is thought to nucleate capsid assembly for a number of viruses (5, 12, 13, 55, 62). Conversely, we found that regions outside the N terminus (i.e., distal to amino acid 82) did not influence HCV capsid assembly per se.
The N terminus of HCV core contains two highly conserved clusters of basic residues separated by a neutral linker region (11). Within the N terminus, the most important residues for assembly appeared to be the basic residues, because deleting the entire neutral linker did not affect HCV assembly. Truncations, deletions, and point mutants encompassing basic residues diminished the ability of core to assemble. Analyzing all of these mutants together revealed that removing more than four basic residues severely impaired the ability to assemble in a cell-free system containing WG extract (Table 1). Our data also suggest that no specific motif for assembly exists because mutating several basic amino acids, regardless of their exact location within the two N-terminal basic clusters, effectively inhibited HCV assembly (see Table 1). These data suggest that the overall charge of the N terminus is important for HCV assembly; however, we cannot rule out that specific residues in this region are more important than others for HCV capsid assembly. Additionally, the contribution of many of the uncharged amino acids has not been addressed. Note that even though its ability to assemble is greatly diminished, the RK16 mutant (encoding 16 basic to neutral amino acid substitutions) maintained some capacity to assemble (see Fig. 3B). This small amount of residual assembly may reflect contributions of basic residues that remain in the N terminus, contributions of basic residues distal to the N-terminal 62 amino acids that we examined, or the contribution of nonbasic residues for assembly.
Our finding that basic charge in the N terminus of HCV core is important for HCV capsid assembly is consistent with studies of assembly of other RNA viruses. While the effect of mutations in capsid proteins on assembly have not been studied systematically for flaviviruses, mutational analyses have been performed for other RNA viruses. Most extensive have been mutational analyses of the retroviral capsid protein, Gag (6). Typically, these studies have demonstrated that basic charge in the Gag nucleocapsid domain (NC) is important for nonspecific association with RNA and for assembly. For example, in the case of HIV-1 Gag, charged residues in NC are required for capsid assembly (16, 17, 56). Consistent with these findings, studies of assembly of purified retroviral capsid proteins in vitro have demonstrated the importance of RNA in nucleating assembly of capsid proteins (12, 13). Studies of alphavirus capsid proteins have also demonstrated the importance of basic residues for RNA encapsidation (61). However, the role of basic charge in promoting capsid formation remains less clear for alphaviruses than for retroviruses. Studies of purified alphavirus capsid proteins in vitro showed that nonspecific association with RNA is important for nucleating assembly of alphavirus core proteins (49, 62, 63). A study of Semliki Forest virus capsid protein mutants expressed in mammalian cells found that complete deletion of the basic charged region in the N terminus eliminated assembly, but large partial deletions were tolerated (20). In contrast, deletions in the Ross River virus N terminus that eliminated most of the basic residues did not impair capsid protein interactions or virus particle production, although viral RNA content was drastically reduced (22). Nevertheless, most studies of RNA virus assembly support a model in which regions of basic charge in capsid proteins interact nonspecifically with RNA, which in turn promotes assembly. Data presented here for HCV capsid assembly fit this model. The exact mechanism by which basic charge acts to promote assembly remains unclear, but the identification of key residues presented here will allow future studies to address this question.
We also investigated whether expressing WT or mutant core constructs together would alter assembly of either construct. Surprisingly, WT C173 failed to rescue any mutants tested, nor did any of the mutant constructs examined act in a dominant-negative manner, suggesting that the core encoded by separate constructs are unable to interact. This is consistent with the possibility that there exist microenvironments where assembly occurs and that these microenvironments contain only one type of core polypeptide. If such microenvironments contain a single mRNA, core polypeptides translated from that particular mRNA might be preferentially incorporated together into capsids and fail to interact with core polypeptides translated from a different mRNA. Such close coordination of translation with assembly could explain the failure of WT C173 to complement core mutants in our experiments and would also be consistent with our previous findings that HCV assembly occurs very rapidly in pulse-chase experiments and is relatively independent of core polypeptide concentration (28). However, this model remains to be tested directly.
Morphological studies of core assembly and two-hybrid assays for core-core interactions have been used in the past to evaluate mutations in HCV core. When a construct encoding a deletion in a domain thought to be involved in homotypic interactions (amino acids 82 to 102) (50) was programmed into our cell-free assembly system, we observed no difference in its ability to assemble relative to WT C173. This region, which is thought to facilitate core dimerization (50), may not be essential for assembly but instead may be important for core to interact with cellular factors that regulate other functions. We also found that converting an aspartic acid to alanine at amino acid 111 (D111A) had no effect on HCV capsid assembly, in contrast to another study that implicated a role for the aspartic acid at 111 in HCV assembly (8). The amount of assembly in the earlier study was not quantified, so the severity of the reported defect is unclear. Interestingly, the D111A mutation in core creates a PSAP motif, which is known to facilitate budding for other viruses through interaction with TSG101 and ESCRT proteins (21). Thus, it is possible that the D111A mutation causes core to interact with cellular machinery such as TSG101 or ESCRT proteins, which may mistarget core and thereby inhibit the ability of core to assemble properly in intact cells, but this hypothesis remains to be tested.
It is likely that interaction of core with other cellular factors plays a role in regulating its ability to assemble. For example, core is known to target to lipid droplets in mammalian cells (1, 24, 25, 43, 53), and this is dependent on proline residues located in the C terminus of core. When prolines at amino acids 138 and 143 are mutated, core no longer targets to lipid droplets, and these core mutants are degraded (25). When core constructs containing these same mutations were expressed in the CFS, they were stable and assembled into capsids to the same extent as WT core. This underscores the need to take into account that although cell-free systems faithfully recapitulate many cellular processes, the CFS described here may not reproduce some of the regulatory events that occur in human liver cells, the natural target of HCV. This dissociation of HCV capsid assembly in the CFS from negative (or positive) regulatory processes is useful because it allows assembly to be studied independently. However, this caveat must be kept in mind when interpreting the assembly data. Note that more complex cell-free systems can be devised to reconstitute regulatory events associated with assembly. Indeed, we have found that by adding liver cell lysates to cell-free reactions, WT HCV capsid assembly is partially inhibited, suggesting that in the presence of specific mammalian cellular factors, HCV assembly can be down-regulated (28). While the critical residues for HCV assembly appear to reside at the N terminus, it is possible that other domains in the core protein (i.e., the C terminus and the E1 signal sequence) play important roles in regulating HCV assembly in vivo. It will be interesting to determine whether mutations in these regions alter the ability of mammalian cell lysates to inhibit HCV capsid assembly in the CFS.
ACKNOWLEDGMENTS
This work was supported by a pilot project grant to J.R.L. from Puget Sound Partners (#26145) and an NIH training grant award to K.C.K. (NIH T32 CA09229).
RRL was a gift from V. Lingappa at the University of California at San Francisco. We thank L. Caldwell at the Fred Hutchinson Cancer Research Center for assistance with electron microscopy; L. Walker for technical assistance; and J. Dooher, V. Lingappa, M. Linial, M. Newman, M. Orr, J. Overbaugh, and L. Walker for helpful discussions.
REFERENCES
Barba, G., F. Harper, T. Harada, M. Kohara, S. Goulinet, Y. Matsuura, G. Eder, Z. Schaff, M. J. Chapman, T. Miyamura, and C. Brechot. 1997. Hepatitis C virus core protein shows a cytoplasmic localization and associates to cellular lipid storage droplets. Proc. Natl. Acad. Sci. USA 94:1200-1205.
Bartenschlager, R., and V. Lohmann. 2001. Novel cell culture systems for the hepatitis C virus. Antiviral Res. 52:1-17.
Bartenschlager, R., and V. Lohmann. 2000. Replication of hepatitis C virus. J. Gen. Virol. 81(Pt. 7):1631-1648.
Baumert, T. F., S. Ito, D. T. Wong, and T. J. Liang. 1998. Hepatitis C virus structural proteins assemble into viruslike particles in insect cells. J. Virol. 72:3827-3836.
Beckett, D., H. N. Wu, and O. C. Uhlenbeck. 1988. Roles of operator and non-operator RNA sequences in bacteriophage R17 capsid assembly. J. Mol. Biol. 204:939-947.
Berkowitz, R., J. Fisher, and S. P. Goff. 1996. RNA packaging. Curr. Top. Microbiol. Immunol. 214:177-218.
Blanchard, E., D. Brand, S. Trassard, A. Goudeau, and P. Roingeard. 2002. Hepatitis C virus-like particle morphogenesis. J. Virol. 76:4073-4079.
Blanchard, E., C. Hourioux, D. Brand, M. Ait-Goughoulte, A. Moreau, S. Trassard, P. Y. Sizaret, F. Dubois, and P. Roingeard. 2003. Hepatitis C virus-like particle budding: role of the core protein and importance of its Asp111. J. Virol. 77:10131-10138.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, J. Marcotrigiano, and C. M. Rice. 2003. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J. Virol. 77:3181-3190.
Bukh, J., R. H. Purcell, and R. H. Miller. 1994. Sequence analysis of the core gene of 14 hepatitis C virus genotypes. Proc. Natl. Acad. Sci. USA 91:8239-8243.
Campbell, S., and A. Rein. 1999. In vitro assembly properties of human immunodeficiency virus type 1 Gag protein lacking the p6 domain. J. Virol. 73:2270-2279.
Campbell, S., and V. M. Vogt. 1995. Self-assembly in vitro of purified CA-NC proteins from Rous sarcoma virus and human immunodeficiency virus type 1. J. Virol. 69:6487-6497.
Chazal, N., B. Gay, C. Carriere, J. Tournier, and P. Boulanger. 1995. Human immunodeficiency virus type 1 MA deletion mutants expressed in baculovirus-infected cells: cis and trans effects on the Gag precursor assembly pathway. J. Virol. 69:365-375.
Chen, C. M., L. R. You, L. H. Hwang, and Y. H. Lee. 1997. Direct interaction of hepatitis C virus core protein with the cellular lymphotoxin-beta receptor modulates the signal pathway of the lymphotoxin-beta receptor. J. Virol. 71:9417-9426.
Cimarelli, A., S. Sandin, S. Hoglund, and J. Luban. 2000. Basic residues in human immunodeficiency virus type 1 nucleocapsid promote virion assembly via interaction with RNA. J. Virol. 74:3046-3057.
Dawson, L., and X. F. Yu. 1998. The role of nucleocapsid of HIV-1 in virus assembly. Virology 251:141-157.
Erickson, A. H., and G. Blobel. 1983. Cell-free translation of messenger RNA in a wheat germ system. Methods Enzymol. 96:38-50.
Fan, Z., Q. R. Yang, J. S. Twu, and A. H. Sherker. 1999. Specific in vitro association between the hepatitis C viral genome and core protein. J. Med. Virol. 59:131-134.
Forsell, K., M. Suomalainen, and H. Garoff. 1995. Structure-function relation of the NH2-terminal domain of the Semliki Forest virus capsid protein. J. Virol. 69:1556-1563.
Freed, E. O. 2002. Viral late domains. J. Virol. 76:4679-4687.
Frolov, I., E. Frolova, and S. Schlesinger. 1997. Sindbis virus replicons and Sindbis virus: assembly of chimeras and of particles deficient in virus RNA. J. Virol. 71:2819-2829.
Gale, M., Jr., and M. R. Beard. 2001. Molecular clones of hepatitis C virus: applications to animal models. Ilar J. 42:139-151.
Hope, R. G., and J. McLauchlan. 2000. Sequence motifs required for lipid droplet association and protein stability are unique to the hepatitis C virus core protein. J. Gen. Virol. 81:1913-1925.
Hope, R. G., D. J. Murphy, and J. McLauchlan. 2002. The domains required to direct core proteins of hepatitis C virus and GB virus-B to lipid droplets share common features with plant oleosin proteins. J. Biol. Chem. 277:4261-4270.
Hsieh, T. Y., M. Matsumoto, H. C. Chou, R. Schneider, S. B. Hwang, A. S. Lee, and M. M. Lai. 1998. Hepatitis C virus core protein interacts with heterogeneous nuclear ribonucleoprotein K. J. Biol. Chem. 273:17651-17659.
Ikeda, M., M. Yi, K. Li, and S. M. Lemon. 2002. Selectable subgenomic and genome-length dicistronic RNAs derived from an infectious molecular clone of the HCV-N strain of hepatitis C virus replicate efficiently in cultured Huh7 cells. J. Virol. 76:2997-3006.
Klein, K. C., S. J. Polyak, and J. R. Lingappa. 2004. Unique features of hepatitis C virus capsid formation revealed by de novo cell-free assembly. J. Virol. 78:9257-9269.
Kountouras, J., C. Zavos, and D. Chatzopoulos. 2003. Apoptosis in hepatitis C. J. Viral Hepat. 10:335-342.
Kunkel, M., and S. J. Watowich. 2002. Conformational changes accompanying self-assembly of the hepatitis C virus core protein. Virology 294:239-245.
Lindsay, K. L. 2002. Introduction to therapy of hepatitis C. Hepatology 36:S114—S120.
Lingappa, J. R., R. L. Hill, M. L. Wong, and R. S. Hegde. 1997. A multistep, ATP-dependent pathway for assembly of human immunodeficiency virus capsids in a cell-free system. J. Cell Biol. 136:567-581.
Lingappa, J. R., R. L. Martin, M. L. Wong, D. Ganem, W. J. Welch, and V. R. Lingappa. 1994. A eukaryotic cytosolic chaperonin is associated with a high molecular weight intermediate in the assembly of hepatitis B virus capsid, a multimeric particle. J. Cell Biol. 125:99-111.
Liu, Q., C. Tackney, R. A. Bhat, A. M. Prince, and P. Zhang. 1997. Regulated processing of hepatitis C virus core protein is linked to subcellular localization. J. Virol. 71:657-662.
Lo, S. Y., M. J. Selby, and J. H. Ou. 1996. Interaction between hepatitis C virus core protein and E1 envelope protein. J. Virol. 70:5177-5182.
Lohmann, V., F. Korner, A. Dobierzewska, and R. Bartenschlager. 2001. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J. Virol. 75:1437-1449.
Lohmann, V., F. Korner, J. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
Lu, W., and J. H. Ou. 2002. Phosphorylation of hepatitis C virus core protein by protein kinase A and protein kinase C. Virology 300:20-30.
Ma, H. C., C. H. Ke, T. Y. Hsieh, and S. Y. Lo. 2002. The first hydrophobic domain of the hepatitis C virus E1 protein is important for interaction with the capsid protein. J. Gen. Virol. 83:3085-3092.
Martin-Serrano, J., and P. D. Bieniasz. 2003. A bipartite late-budding domain in human immunodeficiency virus type 1. J. Virol. 77:12373-12377.
Matsumoto, M., T. Y. Hsieh, N. Zhu, T. VanArsdale, S. B. Hwang, K. S. Jeng, A. E. Gorbalenya, S. Y. Lo, J. H. Ou, C. F. Ware, and M. M. Lai. 1997. Hepatitis C virus core protein interacts with the cytoplasmic tail of lymphotoxin-beta receptor. J. Virol. 71:1301-1309.
McLauchlan, J. 2000. Properties of the hepatitis C virus core protein: a structural protein that modulates cellular processes. J. Viral Hepat. 7:2-14.
McLauchlan, J., M. K. Lemberg, G. Hope, and B. Martoglio. 2002. Intramembrane proteolysis promotes trafficking of hepatitis C virus core protein to lipid droplets. EMBO J. 21:3980-3988.
Melton, D. A., P. A. Krieg, M. R. Rebagliati, T. Maniatis, K. Zinn, and M. R. Green. 1984. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res. 12:7035-7056.
Mizuno, M., G. Yamada, T. Tanaka, K. Shimotohno, M. Takatani, and T. Tsuji. 1995. Virion-like structures in HeLa G cells transfected with the full-length sequence of the hepatitis C virus genome. Gastroenterology 109:1933-1940.
Moriishi, K., T. Okabayashi, K. Nakai, K. Moriya, K. Koike, S. Murata, T. Chiba, K. Tanaka, R. Suzuki, T. Suzuki, T. Miyamura, and Y. Matsuura. 2003. Proteasome activator PA28gamma-dependent nuclear retention and degradation of hepatitis C virus core protein. J. Virol. 77:10237-10249.
Moriya, K., H. Fujie, Y. Shintani, H. Yotsuyanagi, T. Tsutsumi, K. Ishibashi, Y. Matsuura, S. Kimura, T. Miyamura, and K. Koike. 1998. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nat. Med. 4:1065-1067.
Moriya, K., H. Yotsuyanagi, Y. Shintani, H. Fujie, K. Ishibashi, Y. Matsuura, T. Miyamura, and K. Koike. 1997. Hepatitis C virus core protein induces hepatic steatosis in transgenic mice. J. Gen. Virol. 78(Pt 7):1527-1531.
Mukhopadhyay, S., P. R. Chipman, E. M. Hong, R. J. Kuhn, and M. G. Rossmann. 2002. In vitro-assembled alphavirus core-like particles maintain a structure similar to that of nucleocapsid cores in mature virus. J. Virol. 76:11128-11132.
Nolandt, O., V. Kern, H. Muller, E. Pfaff, L. Theilmann, R. Welker, and H. G. Krausslich. 1997. Analysis of hepatitis C virus core protein interaction domains. J. Gen. Virol. 78:1331-1340.
Owsianka, A. M., and A. H. Patel. 1999. Hepatitis C virus core protein interacts with a human DEAD box protein DDX3. Virology 257:330-340.
Perara, E., and V. R. Lingappa. 1985. A former amino terminal signal sequence engineered to an internal location directs translocation of both flanking protein domains. J. Cell Biol. 101:2292-2301.
Pietschmann, T., V. Lohmann, A. Kaul, N. Krieger, G. Rinck, G. Rutter, D. Strand, and R. Bartenschlager. 2002. Persistent and transient replication of full-length hepatitis C virus genomes in cell culture. J. Virol. 76:4008-4021.
Poynard, T., V. Ratziu, Y. Benhamou, P. Opolon, P. Cacoub, and P. Bedossa. 2000. Natural history of HCV infection. Baillieres Best Pract. Res. Clin. Gastroenterol. 14:211-228.
Sacher, R., R. French, and P. Ahlquist. 1988. Hybrid brome mosaic virus RNAs express and are packaged in tobacco mosaic virus coat protein in vivo. Virology 167:15-24.
Sandefur, S., R. M. Smith, V. Varthakavi, and P. Spearman. 2000. Mapping and characterization of the N-terminal I domain of human immunodeficiency virus type 1 Pr55(Gag). J. Virol. 74:7238-7249.
Santolini, E., G. Migliaccio, and N. La Monica. 1994. Biosynthesis and biochemical properties of the hepatitis C virus core protein. J. Virol. 68:3631-3641.
Shih, C. M., C. M. Chen, S. Y. Chen, and Y. H. Lee. 1995. Modulation of the trans-suppression activity of hepatitis C virus core protein by phosphorylation. J. Virol. 69:1160-1171.
Shih, C. M., S. J. Lo, T. Miyamura, S. Y. Chen, and Y. H. Lee. 1993. Suppression of hepatitis B virus expression and replication by hepatitis C virus core protein in HuH-7 cells. J. Virol. 67:5823-5832.
Smith, A. J., N. Srinivasakumar, M. L. Hammarskjold, and D. Rekosh. 1993. Requirements for incorporation of Pr160gag-pol from human immunodeficiency virus type 1 into virus-like particles. J. Virol. 67:2266-2275.
Strauss, J. H., and E. G. Strauss. 1994. The alphaviruses: gene expression, replication, and evolution. Microbiol. Rev. 58:491-562.
Tellinghuisen, T. L., A. E. Hamburger, B. R. Fisher, R. Ostendorp, and R. J. Kuhn. 1999. In vitro assembly of alphavirus cores by using nucleocapsid protein expressed in Escherichia coli. J. Virol. 73:5309-5319.
Wengler, G., U. Boege, H. Bischoff, and K. Wahn. 1982. The core protein of the alphavirus Sindbis virus assembles into core-like nucleoproteins with the viral genome RNA and with other single-stranded nucleic acids in vitro. Virology 118:401-410.
You, L. R., C. M. Chen, T. S. Yeh, T. Y. Tsai, R. T. Mai, C. H. Lin, and Y. H. Lee. 1999. Hepatitis C virus core protein interacts with cellular putative RNA helicase. J. Virol. 73:2841-2853.
Zhu, N., A. Khoshnan, R. Schneider, M. Matsumoto, G. Dennert, C. Ware, and M. M. Lai. 1998. Hepatitis C virus core protein binds to the cytoplasmic domain of tumor necrosis factor (TNF) receptor 1 and enhances TNF-induced apoptosis. J. Virol. 72:3691-3697.(Kevin C. Klein, Sheri R. )
Department of Medicine, University of Washington, Seattle, Washington 98195
ABSTRACT
Significant advances have been made in understanding hepatitis C virus (HCV) replication through development of replicon systems. However, neither replicon systems nor standard cell culture systems support significant assembly of HCV capsids, leaving a large gap in our knowledge of HCV virion formation. Recently, we established a cell-free system in which over 60% of full-length HCV core protein synthesized de novo in cell extracts assembles into HCV capsids by biochemical and morphological criteria. Here we used mutational analysis to identify residues in HCV core that are important for capsid assembly in this highly reproducible cell-free system. We found that basic residues present in two clusters within the N-terminal 68 amino acids of HCV core played a critical role, while the uncharged linker domain between them was not. Furthermore, the aspartate at position 111, the region spanning amino acids 82 to 102, and three serines that are thought to be sites of phosphorylation do not appear to be critical for HCV capsid formation in this system. Mutation of prolines important for targeting of core to lipid droplets also failed to alter HCV capsid assembly in the cell-free system. In addition, wild-type HCV core did not rescue assembly-defective mutants. These data constitute the first systematic and quantitative analysis of the roles of specific residues and domains of HCV core in capsid formation.
INTRODUCTION
Hepatitis C virus (HCV) is a major public health concern worldwide. Two percent of the world's population is infected with the virus (54), which is the major cause of non-A, non-B hepatitis and which often leads to cirrhosis of the liver or hepatocellular carcinoma (42). While current treatments have improved, they are poorly tolerated, not completely efficacious, and not available in all settings (31). Development of better treatments and vaccines has been hampered by the lack of virus replication systems. Standard cell culture systems do not support HCV replication, and the chimpanzee is the only animal model capable of being infected and yielding high HCV titers (23). These limitations have been partly overcome by the development of HCV replicon systems, which support autonomous replication of HCV RNA (2, 10, 36, 37). Replicon systems have allowed specific requirements of HCV RNA replication to be studied; however, HCV particles, or even nucleocapsids, are not produced in these systems (9, 27, 53). Consistent with this observation, infectious particles are not released from these cells (27). As a result, the requirements for critical steps in virion formation, including capsid assembly, genome encapsidation, budding, and release, remain largely unknown.
HCV is an enveloped, single-stranded, positive-sense RNA virus in the Flaviviridae family (3). The HCV genome has a single open reading frame that codes for a 3,000-amino-acid polyprotein. The structural proteins, core, E1, and E2, are the N-terminal products in the polyprotein and are released from the polyprotein by host proteinases (34, 43, 57). Once cleaved from the polyprotein, the 173-amino-acid mature core protein assembles into HCV capsids at the cytoplasmic face of the endoplasmic reticulum (ER) (7, 8, 45). Core is known to interact with the HCV envelope glycoprotein E1 at the ER (35), and assembled capsids are thought to acquire their envelopes by budding into the ER (4, 7, 8, 35). However, the specific details of these late events in the viral life cycle have not been elucidated.
Besides forming the capsid that houses the HCV genomic RNA, core is also known to modulate diverse cellular functions. Core is carcinogenic when expressed in transgenic mice (47), has pro- and antiapoptotic functions (29), alters the transcription of other viral promoters (58, 59), and appears to induce steatosis (48) and the formation of lipid droplets (1). Core is known to interact directly with many different cellular proteins that most likely play a role in modulating these diverse functions (42) or may modulate the ability of core to assemble. Most interacting proteins have been identified through yeast two-hybrid approaches. All reported interactions of core with intracellular proteins, including the cytoplasmic domain of tumor necrosis factor (TNF) receptor 1 (65), lymphotoxin ? receptor (15, 41), hnRNP K (26), DDX3 (51), Cap-Rf (64), and PA28 (46), are mediated by the N-terminal two-thirds of core. The observation that no cellular proteins associate with the C terminus of core may reflect an artifact of the yeast two-hybrid assay (many screens have been performed with truncations of core, e.g., 41) or may be because the hydrophobic C terminus is less accessible to interactions with cellular proteins. However, the interaction of the HCV envelope glycoprotein E1 with core has been mapped to the C terminus (35, 39). Core is also known to target to lipid droplets via its C terminus, where it colocalizes with apolipoprotein AII (1). Deleting amino acids 153 to 169 or mutating the prolines at positions 138 and 143 abolishes lipid droplet association (25, 43). While the significance of core trafficking to lipid droplets remains unclear, this targeting event may have implications for replication or pathogenesis (42). Besides interacting with cellular proteins, core is known to self-associate. Amino acids from 82 to 102, referred to as a homotypic interacting domain, mediate core-core interactions in a yeast two-hybrid assay (50). Core is also highly basic and binds RNA (19, 57), and this association is dependent on the basic N terminus (57). Because interactions of core with cellular proteins can be studied in many cultured cell lines, much is known about these interactions, including which domains of core are required. However, due to the lack of systems for studying assembly in a quantitative manner, no systematic mutational analysis of core has been performed to define domains that are important for its ability to assemble into an HCV capsid.
We have recently established a cell-free system (CFS) that supports robust HCV core assembly and forms bona fide HCV capsids, as demonstrated by velocity sedimentation, buoyant density, and transmission electron microscopic analyses (28). Thus, HCV capsids produced in this system are nearly indistinguishable from authentic capsids isolated from the serum of infected patients (28). This system links de novo translation of core to HCV capsid assembly in eukaryotic extracts, recapitulating events beginning with synthesis of HCV core through the completion of capsid assembly. Additional advantages and features of this system are that it is amenable to quantitation, highly reproducible, can be manipulated, and supports synthesis of mutant core constructs. Previously we used this system to define characteristics and requirements of HCV assembly and showed the importance of the N-terminal RNA binding region for capsid assembly (28). Here we use this permissive HCV capsid assembly system to further characterize important residues in the N terminus and to examine domains elsewhere in the core protein that may be involved in capsid assembly. We find that clusters of basic charges are critical for HCV assembly and that certain uncharged regions in both the N terminus and the remainder of the protein are largely dispensable for HCV capsid assembly.
MATERIALS AND METHODS
DNA plasmids. Cell-free plasmid vectors were derived from SP64 vector (Promega) into which the 5' untranslated region (UTR) of Xenopus laevis globin had been inserted at the HindIII site (44). The parental plasmid used for this study contained the HCV core coding region from an HCV 1b isolate as described previously (28). Using the parental plasmid as the template, mutant core constructs were prepared by either site directed mutagenesis (Stratagene), or by two-step-PCR and cloned into the BglII and EcoRI sites of the parental plasmid (data not shown). All coding regions were verified by sequencing.
In vitro transcription and cell-free translation/assembly. In vitro transcription was performed using the SP6 polymerase (New England Biolabs) and the SP64 expression plasmids described above. Resulting transcripts (or transcription buffer for mock transcript, as indicated) were used to program cell-free translation and assembly reactions for 120 min, either using wheat germ extracts at 26°C or using rabbit reticulocyte lysate at 37°C. Reactions were radiolabeled using [35S]methionine (ICN Biochemicals), as described previously (18, 33, 52).
Gradient analysis of cell-free reactions and cellular lysate. Calibration of gradients to determine S value positions has been described previously (32, 33). Ten to 20 microliters of cell-free assembly reactions, diluted into 200 μl final volume containing 0.625% NP-40 detergent, 10 mM Tris-acetate, pH 7.4, 50 mM KAc, 100 mM NaCl, and 4 mM Mg acetate, were layered onto gradients containing sucrose prepared in the same buffer. Velocity sedimentation was performed on step gradients containing 400 μl each of 10%, 20%, 30%, and 40% sucrose with a 200-μl 50% sucrose cushion. Centrifugation of velocity sedimentation gradients was performed at 201,000 x g for 55 min at 4°C in a TLS55 rotor (Beckman Coulter Optima Max centrifuge), and 200-μl fractions were collected serially from the top. Equivalent aliquots of each gradient fraction were trichloroacetic acid (TCA) precipitated and subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by autoradiography (AR).
Chymotrypsin digestions. Chymotrypsin digestions were done as previously reported (30). Either fractions 1 and 2 or fractions 6 and 7 from velocity sedimentation gradients were pooled and then added to chymotrypsin reactions. Chymotrypsin reactions were set up on ice and contained 300 μl of sample, 50 mM Tris, pH 8.0, 20 mM CaCl2, and 1 ng/μl chymotrypsin (Bovine Pancreas; Calbiochem) in a volume of 1.2 ml. At the indicated time points, 200 μl of the reaction were removed, quenched with TCA, TCA precipitated, and analyzed by SDS-PAGE and AR. The concentration of chymotrypsin was decreased 10-fold from the previous report (30) to compensate for greatly reduced amount of substrate in our samples.
Electron microscopy. Cell-free reactions were subjected to velocity sedimentation centrifugation, and fractions 6, 7, and 8 were pooled and dialyzed (molecular weight cutoff, 55,000) for 1 h against phosphate-buffered saline (PBS) at room temperature. Dialyzed samples were then settled onto carbon-coated grids (Ted Pella) for 2 to 3 min, stained with 1% uranyl acetate for 30 s, and visualized using a JEOL 1010 transmission electron microscope. An experienced electron microscopist identified and examined reactions and controls in single-blinded fashion in three separate experiments. Histograms were prepared by measuring the diameters of all capsids in three to six fields (containing 120 capsids). Criteria for excluding small particles that were not capsids have been described previously (28).
Quantitation. Autoradiographs were digitized using an AGFA Duoscan T1200 scanner and Adobe Photoshop 5.5 software (Adobe Systems Incorporated). Mean band densities were determined and adjusted for band size and background. Amount of assembly was quantitated as the amount of core present in fractions 6, 7, and 8.
Statistics. P values were obtained using a paired student t test (one-tailed).
RESULTS
Internal deletions within the N-terminal 42 amino acids diminish HCV capsid assembly. We previously established a CFS that supports HCV capsid assembly and demonstrated that approximately 60 to 80% of newly synthesized HCV core polypeptides assemble into 100S complexes that migrate in fractions 6 to 8 on our velocity sedimentation gradients. The 100S complexes correspond to HCV capsids by multiple criteria (28). Additionally, we showed that HCV capsids are heterogeneous in size as is seen in vivo, and the 100S cell-free capsids range from 75S to 120S upon analysis in higher resolution gradients (28). In our previous study, we demonstrated that both forms of full-length HCV core that are found in vivo (C191, which includes the E1 signal sequence, and the mature, signal-cleaved version of core, termed C173) assemble to an equivalent extent. Furthermore, there was no difference in assembly when wheat germ extract (WG) versus rabbit reticulocyte lysate (RRL) was utilized as a source of cellular factors for the CFS (28). Thus, unless otherwise noted, experiments in the current study were performed in the WG CFS using C173 as wild-type (WT) core, with % assembly defined as % of radiolabeled core polypeptides present in gradient fractions 6 to 8.
Additionally, in the previous study, we showed that the highly basic N-terminal 68 amino acids of HCV core contain residues important for HCV capsid assembly, since N-terminal truncations of HCV core altered assembly while truncations in the C-terminal half of core had no effect. We found that deleting the N-terminal 10 amino acids (N10) had no effect on cell-free assembly (79% assembly) but deleting the N-terminal 42 (N42) or 68 (N68) amino acids decreased assembly to 12% and 8%, respectively (28). Qualitatively this was seen in velocity sedimentation profiles in which WT C173 migrated largely in fractions 6 to 8, and N42 and N68 migrated at the top of the gradient in fractions 1 and 2, representing complexes of <10S (28).
Based on these observations, we wanted to further identify specific residues or motifs that are important for capsid assembly. Therefore, we generated a series of internal deletions within the N-terminal 42 amino acids (Fig. 1A). When total translations were analyzed by SDS-PAGE, we found that all the deletion mutants expressed to similar levels as WT C173 and migrated at their expected sizes (Fig. 1B). We then tested their ability to assemble into HCV capsids in the CFS by velocity sedimentation (Fig. 1C), as described previously (28). Because core is known to be extremely lipophilic, the velocity sedimentation gradients contained 0.625% NP-40 to solubilize nonspecific proteins aggregating with core as well as associating membranes, allowing the true sedimentation value of HCV core and assembled HCV capsids to be assessed. Deletion of 32 amino acids (designated 11-42) resulted in a statistically significant decrease to 21% assembly, compared to WT C173, which yielded 60% assembly. This was consistent with results obtained with the previously described N-terminal truncation mutants (28) and confirmed that residues required for assembly are present between amino acids 11 to 42. However, we found that smaller deletions that removed 11 to 22 residues within this region (11-21, 21-42, and 31-42) inhibited assembly to only a modest extent (Fig. 1C), suggesting that this region does not contain a single discrete motif that is critical for assembly.
Basic residues in the N terminus of core are critical for HCV capsid assembly. Given these findings, we hypothesized that residues important for assembly are likely to be dispersed between amino acids 11 to 42 and might function cooperatively. The N terminus contains two regions with a high density of basic amino acids linked by a neutral region of 15 amino acids (Fig. 2A). Therefore, we hypothesized that the two basic regions and possibly the neutral linking region are important for HCV assembly. To test this possibility, we constructed targeted deletions of the two basic clusters (1 and 2) and the neutral linker (L). The 1 construct (amino acids 8 to 25 deleted) removed 7 basic residues, while 2 (amino acids 39 to 64 deleted) removed 10 basic residues, and L (amino acids 27 to 38 deleted) removed no basic residues (Fig. 2A). The deletion mutants expressed to similar levels and migrated on SDS-PAGE at their expected sizes (Fig. 2B). When we analyzed their ability to assemble, we found that deleting either cluster of basic amino acids (1 or 2) diminished the ability of core to assemble to approximately 25%. Surprisingly, deleting the entire neutral region (L) between the basic clusters had no effect on HCV capsid assembly in the CFS (Fig. 2C and D).
Next, we directly assessed the contribution of the basic residues in the N terminus by substituting alanines for basic residues in the N terminus. Fifteen mutants encoding as few as one and as many as 16 basic residue substitutions in one or both basic clusters were constructed (Fig. 3A). We found that the ability of the mutant constructs to assemble in the CFS decreased as more basic residues were mutated (Fig. 3B and C). When only one basic residue was mutated (RK1A and RK1B, Fig. 3) there was little or no effect on assembly (60% assembly for each construct compared to 61% assembly for WT core). However, as the number of basic residues mutated increased, a progressive decrease in the ability of core to assemble was observed (i.e., 48% for RK2A and RK2B; 30% for RK3A and RK3B; 25% for RK4A and RK4B), suggesting that basic charge in this region is critical for capsid assembly. It should be noted that not all mutants expressed as well as WT (Fig. 3B, line graph), but this likely does not account for the lack of assembly because we have previously demonstrated that HCV assembly is minimally dependent on core concentration in the CFS (28). Consistent with this observation, when WT core expression was lowered to less than that of RK4B, were observed (data not shown). Additionally, some basic charge substitution mutants displayed equivalent amounts of translation relative to WT C173 but had significantly reduced levels of assembly (i.e., RK6, RK9, RK13, and RK16 in Fig. 3B). These findings are consistent with basic charge in this region being required for efficient capsid assembly.
Table 1 summarizes all the constructs encoding N-terminal mutations (truncations, deletions, and substitutions) that we have studied in the CFS and is organized by the number of basic residues that are altered. The amount of assembly for each construct is indicated as % of WT core assembly (with WT assembly set at 100%). This comparison reveals that removal or mutation of one or two basic residues was generally well tolerated, while further increasing the number of basic residues progressively diminished HCV capsid assembly. The apparent correlation between the number of basic residues present in the N terminus and the ability of HCV core to assemble suggests that these basic residues are an important determinant of HCV capsid assembly.
HCV core polypeptides encoding mutations form capsids that are stable. We repeatedly observed that core mutants displaying intermediate levels of assembly migrated differently in velocity sedimentation gradients compared to fully assembly-competent or assembly-incompetent constructs. In addition to being present in the assembled fractions, core polypeptides that assembled to intermediate levels were dispersed throughout the gradient (i.e., 1 and 2 in Fig. 2D) rather than solely forming peaks in specific fractions as was the case for assembly-competent (i.e., fractions 6 and 7 for WT C173 in Fig. 2D) or assembly-incompetent mutants (i.e., fractions 1 and 2 for N68 in Fig. 6B). This led us to ask whether constructs that assemble to intermediate levels form unstable complexes that give rise to the more dispersed migration pattern during the velocity sedimentation analysis. We examined stability by isolating capsids (in fractions 6 and 7) formed by C173 or core mutants as well as other complexes (in fractions 3 and 4) formed by core mutants that display a limited level of assembly. We then subjected these complexes to a second round of velocity sedimentation on the same type of gradients. If the capsids and/or complexes are inherently stable, they should migrate in the same fractions upon repeat velocity sedimentation. Capsids of WT C173 and two mutants (2 and RK7) were stable, since they migrated to fractions 6 and 7 on the second velocity sedimentation gradient (Fig. 4). RK7 did appear to spread out somewhat during the second velocity sedimentation analysis but mostly migrated in fractions 6 and 7. In contrast, when smaller complexes (i.e., fractions 3 and 4) of the two mutants 2 and RK7 were reanalyzed by velocity sedimentation, they migrated mostly to the top of the gradient (fraction 1, which contains monomers and complexes of <10S), suggesting that they lack stability. Thus, it appears that some mutant constructs that assemble to intermediate or low levels form at least two types of complexes: stable capsid-like complexes that migrate in the 100S assembled fractions and complexes that migrate in other regions of the gradient that are unstable.
Unassembled core polypeptides are sensitive to chymotrypsin, unlike assembled capsids. To further address whether mutant HCV core constructs assemble into capsids at a low level in the cell-free system, we modified a previously described assay that was used to demonstrate that assembled recombinant HCV core is resistant to chymotrypsin digestion, whereas unassembled recombinant HCV core is relatively susceptible to chymotrypsin digestion (30). If the mutant complexes in fractions 6 and 7 are intact capsids, they should be resistant to chymotrypsin digestion. To test this, we isolated 100S capsid like complexes, as well as unassembled HCV core, from cell-free reactions expressing WT or mutant constructs and subjected them to chymotrypsin digestion. Figure 5 shows that WT core from the assembled fractions 6 and 7 (right panels) was relatively resistant to chymotrypsin, while the very small amount of WT core present in the top two fractions (left panels) was protease sensitive. Both the protease sensitivity of unassembled HCV core and the protease resistance of assembled capsids in the CFS resembled the findings previously reported for unassembled and assembled recombinant HCV core (30). Similarly, despite different input amounts for different constructs, mutant core present in unassembled top fractions was relatively protease sensitive while core from assembled fractions 6 and 7 was relatively protease resistant for all constructs examined, including an assembly-competent mutant construct (RK2B), mutant constructs with basic amino acids mutated in either the first or second cluster (RK4B and RK6, respectively), and a construct containing a large deletion (2) (Fig. 5). Thus, the protease resistance of the small amounts of 100S complexes formed by mutant constructs supports the conclusion that these constructs undergo a limited amount of capsid assembly. Additionally, the examined protease sensitivity of the unassembled fractions suggests that the mutations are unlikely to grossly affect protein folding, although it is impossible to rule out very subtle effects on folding.
Wild-type core does not rescue assembly-defective mutants. In certain systems it is possible to rescue defective mutants by expressing a competent construct in trans. For example, budding-defective human immunodeficiency virus type 1 (HIV-1) constructs that are missing the p6 domain of the HIV-1 capsid protein can be complemented in trans in cells with constructs that encode regions of the p6 domain (40). In the case of HIV-1 capsid assembly, assembly-defective nonmyristoylated capsid precursor GagPol proteins can be rescued by coexpression of myristoylated assembly-competent capsid proteins (60). Conversely, mutant capsid proteins can interfere with the function of WT proteins by acting as dominant-negative mutants, as shown by studies of specific mutations in HIV-1 Gag (14). Together, these data indicate that in the case of HIV-1 capsid formation, capsid proteins produced from different transcripts interact during the assembly process. To test whether WT C173 rescues the assembly of assembly-defective constructs or whether core mutants act in a dominant-negative manner to inhibit assembly of WT C173, we programmed cell-free reactions with equivalent amounts of transcripts encoding WT and mutant constructs (N68, 11-42, or N2). These constructs migrate at different positions on SDS-PAGE from WT C173 (Fig. 6A) and therefore can be distinguished when expressed in the same reaction. When we analyzed their ability to assemble by velocity sedimentation, we found that WT C173 assembled to an equivalent extent in the absence or presence of mutant constructs, suggesting that none of the mutants examined behaved in a dominant-negative manner (compare panels in Fig. 6A to C173 panel in B; graphed in C). Moreover, assembly of the mutant constructs did not increase when coexpressed with WT, demonstrating that WT C173 does not complement the assembly-defective mutants (compare panels in Fig. 6A to three lower panels in B; graphed in D). Additionally, coexpressing WT and mutant constructs did not alter the distribution on velocity sedimentation gradients of the mutant core constructs examined, showing that formation or accumulation of other complexes was unaffected (compare Fig. 6A to B). Equivalent results were found when an assembly-competent mutant (82-102; see Fig. 7) was coexpressed with the mutants in Fig. 6 (data not shown).
Mutations in other regions of core do not alter HCV capsid assembly. Other groups have suggested that regions outside of the N terminus of core are important for HCV capsid assembly. Amino acids 82 to 102 were found to contain a homotypic interaction domain by a yeast two-hybrid assay that measures core-core interactions (50). A morphological study of capsid assembly done in BHK21 cells using a Semliki forest virus expression system implicated an aspartic acid at amino 111 as important for HCV assembly (8), although it was unclear to what extent WT core or the mutant assembled in this system. Additionally, other regions in the central region of core called domain 2 (42) are known to be important for targeting of the HCV core protein (24, 25, 43) but have not been examined with respect to capsid assembly. For example, two proline residues (at amino acids 138 and 143) form a proline knot motif that is important for targeting core to lipid droplets (25). Because core-core interactions and targeting are likely to be important for HCV capsid formation, we wanted to quantitatively examine what effect these mutations, shown in Fig. 7A, had on the ability of core to assemble in the cell-free system. In contrast to the assembly-defective N-terminal mutants shown previously, core constructs encoding 82-102, D111A, and mutations in prolines 138 and 143 assembled as well as WT C173 by velocity sedimentation (Fig. 7B and C). Similar results were obtained when D111A was examined using a cell-free system containing RRL rather than WG (Fig. 7B, compare right and left sides). To confirm these findings we also subjected fractions containing 100S capsids encoding the D111A and 82-102 mutations to transmission electron microsocopy (TEM; Fig. 8A to C). We had previously shown that capsids formed by WT C173 in the CFS closely resemble capsids obtained from patient serum (28). We found that the assembly-competent 82-102 and D111A mutants formed abundant numbers of capsids that resembled capsids produced by WT C173 in the CFS. Measurements revealed a bimodal distribution of diameters for capsids formed by the 82-102 mutant. This distribution closely resembled the distribution of diameters displayed by WT C173 capsids prepared in parallel (Fig. 8D). In contrast, capsids encoding the D111A mutation had larger diameters while maintaining a bimodal distribution of diameters (Fig. 8D). These data suggest that, at least in the CFS, these residues are not essential for HCV capsid assembly, although the aspartic acid at position 111 does appear to influence capsid size.
Phosphorylated serines are not essential for capsid assembly. The fact that cultured cells do not support HCV assembly suggests that this process can be regulated; however, this regulation has not been studied in detail. One common mechanism frequently used to regulate cellular processes and cellular proteins is phosphorylation. HCV core has been shown to be phosphorylated in vitro at serines at amino acids 53, 99, and 116 by protein kinase A and C (58) and are phosphorylated in human liver cell lines (38). These serines are highly conserved across different HCV genotypes (11), suggesting they may play an important role for HCV. Studies show that phosphorylation of core may regulate targeting of core to the nucleus (38), and the ability of HCV core to suppress HBV transcription (58). Whether these phosphorylated serine residues play a role in capsid assembly has not been addressed previously. To examine this, serines at 53, 99, and 116 were sequentially converted to alanines (Fig. 9A). Constructs encoding these mutations translated to an equivalent extent and assembled as well as WT C173 in both WG and RRL cell extracts (Fig. 9B and C). Thus, it appears that serines in core that are thought to undergo phosphorylation are not required for HCV capsid assembly in the cell-free system, even when a mammalian extract is used. Conversion of serine 53 to aspartic acid (S53D) to mimic the phosphorylated state also had no effect (Fig. 9B and C). The effect of conversion of other serines to aspartic acids has not yet been examined.
DISCUSSION
Here we have systematically analyzed regions of the HCV core protein that are important for capsid assembly using a recently described HCV cell-free system that allows for a quantitative analysis of assembly in a cellular context. Because it supports efficient capsid assembly (28), this system differs from almost all reported cellular systems in which HCV core fails to assemble into capsids. Because of the lack of systems for studying HCV capsid assembly, surrogate assays for assembly, such as core-core interactions in yeast two-hybrid assays and nonquantitative assays, have been used to define residues that appeared to be important for HCV assembly. It has been difficult to fully interpret results from these studies, leaving a gap in our understanding of which residues and domains are important for HCV assembly. Here we showed that the N-terminal basic region plays a key role in HCV assembly in the CFS. Interestingly, this region is also known to bind RNA, which is thought to nucleate capsid assembly for a number of viruses (5, 12, 13, 55, 62). Conversely, we found that regions outside the N terminus (i.e., distal to amino acid 82) did not influence HCV capsid assembly per se.
The N terminus of HCV core contains two highly conserved clusters of basic residues separated by a neutral linker region (11). Within the N terminus, the most important residues for assembly appeared to be the basic residues, because deleting the entire neutral linker did not affect HCV assembly. Truncations, deletions, and point mutants encompassing basic residues diminished the ability of core to assemble. Analyzing all of these mutants together revealed that removing more than four basic residues severely impaired the ability to assemble in a cell-free system containing WG extract (Table 1). Our data also suggest that no specific motif for assembly exists because mutating several basic amino acids, regardless of their exact location within the two N-terminal basic clusters, effectively inhibited HCV assembly (see Table 1). These data suggest that the overall charge of the N terminus is important for HCV assembly; however, we cannot rule out that specific residues in this region are more important than others for HCV capsid assembly. Additionally, the contribution of many of the uncharged amino acids has not been addressed. Note that even though its ability to assemble is greatly diminished, the RK16 mutant (encoding 16 basic to neutral amino acid substitutions) maintained some capacity to assemble (see Fig. 3B). This small amount of residual assembly may reflect contributions of basic residues that remain in the N terminus, contributions of basic residues distal to the N-terminal 62 amino acids that we examined, or the contribution of nonbasic residues for assembly.
Our finding that basic charge in the N terminus of HCV core is important for HCV capsid assembly is consistent with studies of assembly of other RNA viruses. While the effect of mutations in capsid proteins on assembly have not been studied systematically for flaviviruses, mutational analyses have been performed for other RNA viruses. Most extensive have been mutational analyses of the retroviral capsid protein, Gag (6). Typically, these studies have demonstrated that basic charge in the Gag nucleocapsid domain (NC) is important for nonspecific association with RNA and for assembly. For example, in the case of HIV-1 Gag, charged residues in NC are required for capsid assembly (16, 17, 56). Consistent with these findings, studies of assembly of purified retroviral capsid proteins in vitro have demonstrated the importance of RNA in nucleating assembly of capsid proteins (12, 13). Studies of alphavirus capsid proteins have also demonstrated the importance of basic residues for RNA encapsidation (61). However, the role of basic charge in promoting capsid formation remains less clear for alphaviruses than for retroviruses. Studies of purified alphavirus capsid proteins in vitro showed that nonspecific association with RNA is important for nucleating assembly of alphavirus core proteins (49, 62, 63). A study of Semliki Forest virus capsid protein mutants expressed in mammalian cells found that complete deletion of the basic charged region in the N terminus eliminated assembly, but large partial deletions were tolerated (20). In contrast, deletions in the Ross River virus N terminus that eliminated most of the basic residues did not impair capsid protein interactions or virus particle production, although viral RNA content was drastically reduced (22). Nevertheless, most studies of RNA virus assembly support a model in which regions of basic charge in capsid proteins interact nonspecifically with RNA, which in turn promotes assembly. Data presented here for HCV capsid assembly fit this model. The exact mechanism by which basic charge acts to promote assembly remains unclear, but the identification of key residues presented here will allow future studies to address this question.
We also investigated whether expressing WT or mutant core constructs together would alter assembly of either construct. Surprisingly, WT C173 failed to rescue any mutants tested, nor did any of the mutant constructs examined act in a dominant-negative manner, suggesting that the core encoded by separate constructs are unable to interact. This is consistent with the possibility that there exist microenvironments where assembly occurs and that these microenvironments contain only one type of core polypeptide. If such microenvironments contain a single mRNA, core polypeptides translated from that particular mRNA might be preferentially incorporated together into capsids and fail to interact with core polypeptides translated from a different mRNA. Such close coordination of translation with assembly could explain the failure of WT C173 to complement core mutants in our experiments and would also be consistent with our previous findings that HCV assembly occurs very rapidly in pulse-chase experiments and is relatively independent of core polypeptide concentration (28). However, this model remains to be tested directly.
Morphological studies of core assembly and two-hybrid assays for core-core interactions have been used in the past to evaluate mutations in HCV core. When a construct encoding a deletion in a domain thought to be involved in homotypic interactions (amino acids 82 to 102) (50) was programmed into our cell-free assembly system, we observed no difference in its ability to assemble relative to WT C173. This region, which is thought to facilitate core dimerization (50), may not be essential for assembly but instead may be important for core to interact with cellular factors that regulate other functions. We also found that converting an aspartic acid to alanine at amino acid 111 (D111A) had no effect on HCV capsid assembly, in contrast to another study that implicated a role for the aspartic acid at 111 in HCV assembly (8). The amount of assembly in the earlier study was not quantified, so the severity of the reported defect is unclear. Interestingly, the D111A mutation in core creates a PSAP motif, which is known to facilitate budding for other viruses through interaction with TSG101 and ESCRT proteins (21). Thus, it is possible that the D111A mutation causes core to interact with cellular machinery such as TSG101 or ESCRT proteins, which may mistarget core and thereby inhibit the ability of core to assemble properly in intact cells, but this hypothesis remains to be tested.
It is likely that interaction of core with other cellular factors plays a role in regulating its ability to assemble. For example, core is known to target to lipid droplets in mammalian cells (1, 24, 25, 43, 53), and this is dependent on proline residues located in the C terminus of core. When prolines at amino acids 138 and 143 are mutated, core no longer targets to lipid droplets, and these core mutants are degraded (25). When core constructs containing these same mutations were expressed in the CFS, they were stable and assembled into capsids to the same extent as WT core. This underscores the need to take into account that although cell-free systems faithfully recapitulate many cellular processes, the CFS described here may not reproduce some of the regulatory events that occur in human liver cells, the natural target of HCV. This dissociation of HCV capsid assembly in the CFS from negative (or positive) regulatory processes is useful because it allows assembly to be studied independently. However, this caveat must be kept in mind when interpreting the assembly data. Note that more complex cell-free systems can be devised to reconstitute regulatory events associated with assembly. Indeed, we have found that by adding liver cell lysates to cell-free reactions, WT HCV capsid assembly is partially inhibited, suggesting that in the presence of specific mammalian cellular factors, HCV assembly can be down-regulated (28). While the critical residues for HCV assembly appear to reside at the N terminus, it is possible that other domains in the core protein (i.e., the C terminus and the E1 signal sequence) play important roles in regulating HCV assembly in vivo. It will be interesting to determine whether mutations in these regions alter the ability of mammalian cell lysates to inhibit HCV capsid assembly in the CFS.
ACKNOWLEDGMENTS
This work was supported by a pilot project grant to J.R.L. from Puget Sound Partners (#26145) and an NIH training grant award to K.C.K. (NIH T32 CA09229).
RRL was a gift from V. Lingappa at the University of California at San Francisco. We thank L. Caldwell at the Fred Hutchinson Cancer Research Center for assistance with electron microscopy; L. Walker for technical assistance; and J. Dooher, V. Lingappa, M. Linial, M. Newman, M. Orr, J. Overbaugh, and L. Walker for helpful discussions.
REFERENCES
Barba, G., F. Harper, T. Harada, M. Kohara, S. Goulinet, Y. Matsuura, G. Eder, Z. Schaff, M. J. Chapman, T. Miyamura, and C. Brechot. 1997. Hepatitis C virus core protein shows a cytoplasmic localization and associates to cellular lipid storage droplets. Proc. Natl. Acad. Sci. USA 94:1200-1205.
Bartenschlager, R., and V. Lohmann. 2001. Novel cell culture systems for the hepatitis C virus. Antiviral Res. 52:1-17.
Bartenschlager, R., and V. Lohmann. 2000. Replication of hepatitis C virus. J. Gen. Virol. 81(Pt. 7):1631-1648.
Baumert, T. F., S. Ito, D. T. Wong, and T. J. Liang. 1998. Hepatitis C virus structural proteins assemble into viruslike particles in insect cells. J. Virol. 72:3827-3836.
Beckett, D., H. N. Wu, and O. C. Uhlenbeck. 1988. Roles of operator and non-operator RNA sequences in bacteriophage R17 capsid assembly. J. Mol. Biol. 204:939-947.
Berkowitz, R., J. Fisher, and S. P. Goff. 1996. RNA packaging. Curr. Top. Microbiol. Immunol. 214:177-218.
Blanchard, E., D. Brand, S. Trassard, A. Goudeau, and P. Roingeard. 2002. Hepatitis C virus-like particle morphogenesis. J. Virol. 76:4073-4079.
Blanchard, E., C. Hourioux, D. Brand, M. Ait-Goughoulte, A. Moreau, S. Trassard, P. Y. Sizaret, F. Dubois, and P. Roingeard. 2003. Hepatitis C virus-like particle budding: role of the core protein and importance of its Asp111. J. Virol. 77:10131-10138.
Blight, K. J., A. A. Kolykhalov, and C. M. Rice. 2000. Efficient initiation of HCV RNA replication in cell culture. Science 290:1972-1974.
Blight, K. J., J. A. McKeating, J. Marcotrigiano, and C. M. Rice. 2003. Efficient replication of hepatitis C virus genotype 1a RNAs in cell culture. J. Virol. 77:3181-3190.
Bukh, J., R. H. Purcell, and R. H. Miller. 1994. Sequence analysis of the core gene of 14 hepatitis C virus genotypes. Proc. Natl. Acad. Sci. USA 91:8239-8243.
Campbell, S., and A. Rein. 1999. In vitro assembly properties of human immunodeficiency virus type 1 Gag protein lacking the p6 domain. J. Virol. 73:2270-2279.
Campbell, S., and V. M. Vogt. 1995. Self-assembly in vitro of purified CA-NC proteins from Rous sarcoma virus and human immunodeficiency virus type 1. J. Virol. 69:6487-6497.
Chazal, N., B. Gay, C. Carriere, J. Tournier, and P. Boulanger. 1995. Human immunodeficiency virus type 1 MA deletion mutants expressed in baculovirus-infected cells: cis and trans effects on the Gag precursor assembly pathway. J. Virol. 69:365-375.
Chen, C. M., L. R. You, L. H. Hwang, and Y. H. Lee. 1997. Direct interaction of hepatitis C virus core protein with the cellular lymphotoxin-beta receptor modulates the signal pathway of the lymphotoxin-beta receptor. J. Virol. 71:9417-9426.
Cimarelli, A., S. Sandin, S. Hoglund, and J. Luban. 2000. Basic residues in human immunodeficiency virus type 1 nucleocapsid promote virion assembly via interaction with RNA. J. Virol. 74:3046-3057.
Dawson, L., and X. F. Yu. 1998. The role of nucleocapsid of HIV-1 in virus assembly. Virology 251:141-157.
Erickson, A. H., and G. Blobel. 1983. Cell-free translation of messenger RNA in a wheat germ system. Methods Enzymol. 96:38-50.
Fan, Z., Q. R. Yang, J. S. Twu, and A. H. Sherker. 1999. Specific in vitro association between the hepatitis C viral genome and core protein. J. Med. Virol. 59:131-134.
Forsell, K., M. Suomalainen, and H. Garoff. 1995. Structure-function relation of the NH2-terminal domain of the Semliki Forest virus capsid protein. J. Virol. 69:1556-1563.
Freed, E. O. 2002. Viral late domains. J. Virol. 76:4679-4687.
Frolov, I., E. Frolova, and S. Schlesinger. 1997. Sindbis virus replicons and Sindbis virus: assembly of chimeras and of particles deficient in virus RNA. J. Virol. 71:2819-2829.
Gale, M., Jr., and M. R. Beard. 2001. Molecular clones of hepatitis C virus: applications to animal models. Ilar J. 42:139-151.
Hope, R. G., and J. McLauchlan. 2000. Sequence motifs required for lipid droplet association and protein stability are unique to the hepatitis C virus core protein. J. Gen. Virol. 81:1913-1925.
Hope, R. G., D. J. Murphy, and J. McLauchlan. 2002. The domains required to direct core proteins of hepatitis C virus and GB virus-B to lipid droplets share common features with plant oleosin proteins. J. Biol. Chem. 277:4261-4270.
Hsieh, T. Y., M. Matsumoto, H. C. Chou, R. Schneider, S. B. Hwang, A. S. Lee, and M. M. Lai. 1998. Hepatitis C virus core protein interacts with heterogeneous nuclear ribonucleoprotein K. J. Biol. Chem. 273:17651-17659.
Ikeda, M., M. Yi, K. Li, and S. M. Lemon. 2002. Selectable subgenomic and genome-length dicistronic RNAs derived from an infectious molecular clone of the HCV-N strain of hepatitis C virus replicate efficiently in cultured Huh7 cells. J. Virol. 76:2997-3006.
Klein, K. C., S. J. Polyak, and J. R. Lingappa. 2004. Unique features of hepatitis C virus capsid formation revealed by de novo cell-free assembly. J. Virol. 78:9257-9269.
Kountouras, J., C. Zavos, and D. Chatzopoulos. 2003. Apoptosis in hepatitis C. J. Viral Hepat. 10:335-342.
Kunkel, M., and S. J. Watowich. 2002. Conformational changes accompanying self-assembly of the hepatitis C virus core protein. Virology 294:239-245.
Lindsay, K. L. 2002. Introduction to therapy of hepatitis C. Hepatology 36:S114—S120.
Lingappa, J. R., R. L. Hill, M. L. Wong, and R. S. Hegde. 1997. A multistep, ATP-dependent pathway for assembly of human immunodeficiency virus capsids in a cell-free system. J. Cell Biol. 136:567-581.
Lingappa, J. R., R. L. Martin, M. L. Wong, D. Ganem, W. J. Welch, and V. R. Lingappa. 1994. A eukaryotic cytosolic chaperonin is associated with a high molecular weight intermediate in the assembly of hepatitis B virus capsid, a multimeric particle. J. Cell Biol. 125:99-111.
Liu, Q., C. Tackney, R. A. Bhat, A. M. Prince, and P. Zhang. 1997. Regulated processing of hepatitis C virus core protein is linked to subcellular localization. J. Virol. 71:657-662.
Lo, S. Y., M. J. Selby, and J. H. Ou. 1996. Interaction between hepatitis C virus core protein and E1 envelope protein. J. Virol. 70:5177-5182.
Lohmann, V., F. Korner, A. Dobierzewska, and R. Bartenschlager. 2001. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J. Virol. 75:1437-1449.
Lohmann, V., F. Korner, J. Koch, U. Herian, L. Theilmann, and R. Bartenschlager. 1999. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 285:110-113.
Lu, W., and J. H. Ou. 2002. Phosphorylation of hepatitis C virus core protein by protein kinase A and protein kinase C. Virology 300:20-30.
Ma, H. C., C. H. Ke, T. Y. Hsieh, and S. Y. Lo. 2002. The first hydrophobic domain of the hepatitis C virus E1 protein is important for interaction with the capsid protein. J. Gen. Virol. 83:3085-3092.
Martin-Serrano, J., and P. D. Bieniasz. 2003. A bipartite late-budding domain in human immunodeficiency virus type 1. J. Virol. 77:12373-12377.
Matsumoto, M., T. Y. Hsieh, N. Zhu, T. VanArsdale, S. B. Hwang, K. S. Jeng, A. E. Gorbalenya, S. Y. Lo, J. H. Ou, C. F. Ware, and M. M. Lai. 1997. Hepatitis C virus core protein interacts with the cytoplasmic tail of lymphotoxin-beta receptor. J. Virol. 71:1301-1309.
McLauchlan, J. 2000. Properties of the hepatitis C virus core protein: a structural protein that modulates cellular processes. J. Viral Hepat. 7:2-14.
McLauchlan, J., M. K. Lemberg, G. Hope, and B. Martoglio. 2002. Intramembrane proteolysis promotes trafficking of hepatitis C virus core protein to lipid droplets. EMBO J. 21:3980-3988.
Melton, D. A., P. A. Krieg, M. R. Rebagliati, T. Maniatis, K. Zinn, and M. R. Green. 1984. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res. 12:7035-7056.
Mizuno, M., G. Yamada, T. Tanaka, K. Shimotohno, M. Takatani, and T. Tsuji. 1995. Virion-like structures in HeLa G cells transfected with the full-length sequence of the hepatitis C virus genome. Gastroenterology 109:1933-1940.
Moriishi, K., T. Okabayashi, K. Nakai, K. Moriya, K. Koike, S. Murata, T. Chiba, K. Tanaka, R. Suzuki, T. Suzuki, T. Miyamura, and Y. Matsuura. 2003. Proteasome activator PA28gamma-dependent nuclear retention and degradation of hepatitis C virus core protein. J. Virol. 77:10237-10249.
Moriya, K., H. Fujie, Y. Shintani, H. Yotsuyanagi, T. Tsutsumi, K. Ishibashi, Y. Matsuura, S. Kimura, T. Miyamura, and K. Koike. 1998. The core protein of hepatitis C virus induces hepatocellular carcinoma in transgenic mice. Nat. Med. 4:1065-1067.
Moriya, K., H. Yotsuyanagi, Y. Shintani, H. Fujie, K. Ishibashi, Y. Matsuura, T. Miyamura, and K. Koike. 1997. Hepatitis C virus core protein induces hepatic steatosis in transgenic mice. J. Gen. Virol. 78(Pt 7):1527-1531.
Mukhopadhyay, S., P. R. Chipman, E. M. Hong, R. J. Kuhn, and M. G. Rossmann. 2002. In vitro-assembled alphavirus core-like particles maintain a structure similar to that of nucleocapsid cores in mature virus. J. Virol. 76:11128-11132.
Nolandt, O., V. Kern, H. Muller, E. Pfaff, L. Theilmann, R. Welker, and H. G. Krausslich. 1997. Analysis of hepatitis C virus core protein interaction domains. J. Gen. Virol. 78:1331-1340.
Owsianka, A. M., and A. H. Patel. 1999. Hepatitis C virus core protein interacts with a human DEAD box protein DDX3. Virology 257:330-340.
Perara, E., and V. R. Lingappa. 1985. A former amino terminal signal sequence engineered to an internal location directs translocation of both flanking protein domains. J. Cell Biol. 101:2292-2301.
Pietschmann, T., V. Lohmann, A. Kaul, N. Krieger, G. Rinck, G. Rutter, D. Strand, and R. Bartenschlager. 2002. Persistent and transient replication of full-length hepatitis C virus genomes in cell culture. J. Virol. 76:4008-4021.
Poynard, T., V. Ratziu, Y. Benhamou, P. Opolon, P. Cacoub, and P. Bedossa. 2000. Natural history of HCV infection. Baillieres Best Pract. Res. Clin. Gastroenterol. 14:211-228.
Sacher, R., R. French, and P. Ahlquist. 1988. Hybrid brome mosaic virus RNAs express and are packaged in tobacco mosaic virus coat protein in vivo. Virology 167:15-24.
Sandefur, S., R. M. Smith, V. Varthakavi, and P. Spearman. 2000. Mapping and characterization of the N-terminal I domain of human immunodeficiency virus type 1 Pr55(Gag). J. Virol. 74:7238-7249.
Santolini, E., G. Migliaccio, and N. La Monica. 1994. Biosynthesis and biochemical properties of the hepatitis C virus core protein. J. Virol. 68:3631-3641.
Shih, C. M., C. M. Chen, S. Y. Chen, and Y. H. Lee. 1995. Modulation of the trans-suppression activity of hepatitis C virus core protein by phosphorylation. J. Virol. 69:1160-1171.
Shih, C. M., S. J. Lo, T. Miyamura, S. Y. Chen, and Y. H. Lee. 1993. Suppression of hepatitis B virus expression and replication by hepatitis C virus core protein in HuH-7 cells. J. Virol. 67:5823-5832.
Smith, A. J., N. Srinivasakumar, M. L. Hammarskjold, and D. Rekosh. 1993. Requirements for incorporation of Pr160gag-pol from human immunodeficiency virus type 1 into virus-like particles. J. Virol. 67:2266-2275.
Strauss, J. H., and E. G. Strauss. 1994. The alphaviruses: gene expression, replication, and evolution. Microbiol. Rev. 58:491-562.
Tellinghuisen, T. L., A. E. Hamburger, B. R. Fisher, R. Ostendorp, and R. J. Kuhn. 1999. In vitro assembly of alphavirus cores by using nucleocapsid protein expressed in Escherichia coli. J. Virol. 73:5309-5319.
Wengler, G., U. Boege, H. Bischoff, and K. Wahn. 1982. The core protein of the alphavirus Sindbis virus assembles into core-like nucleoproteins with the viral genome RNA and with other single-stranded nucleic acids in vitro. Virology 118:401-410.
You, L. R., C. M. Chen, T. S. Yeh, T. Y. Tsai, R. T. Mai, C. H. Lin, and Y. H. Lee. 1999. Hepatitis C virus core protein interacts with cellular putative RNA helicase. J. Virol. 73:2841-2853.
Zhu, N., A. Khoshnan, R. Schneider, M. Matsumoto, G. Dennert, C. Ware, and M. M. Lai. 1998. Hepatitis C virus core protein binds to the cytoplasmic domain of tumor necrosis factor (TNF) receptor 1 and enhances TNF-induced apoptosis. J. Virol. 72:3691-3697.(Kevin C. Klein, Sheri R. )