Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M501826200 on March 9, 2005

J. Biol. Chem., Vol. 280, Issue 18, 17737-17748, May 6, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/18/17737    most recent
M501826200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boni, S.
Right arrow Articles by Cahour, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boni, S.
Right arrow Articles by Cahour, A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Hepatitis C Virus Core Protein Acts as a trans-Modulating Factor on Internal Translation Initiation of the Viral RNA*

Sébastien Boni{ddagger}§, Jean-Pierre Lavergne¶, Steeve Boulant§, and Annie Cahour{ddagger}||

From the {ddagger}Laboratoire de Virologie, Centre Européen de Recherche en Virologie et Immunologie, Unité Propre de Recherche et d'Enseignement Supérier EA 2387, IFR 113 Immunité et Infection, Groupe Hospitalier Pitié-Salpêtrière, 83 Boulevard de l'Hôpital, 75651 Paris Cedex 13 and Laboratoire de Bioinformatique et RMN Structurale, Institut de Biologie et Chimie des Protéines, UMR5086 CNRS, Université Claude Bernard Lyon I, IFR 128 Biosciences Lyon-Gerland, 7 Passage du Vercors, 69367 Lyon Cedex 07, France

Received for publication, February 17, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Translation initiation of hepatitis C virus (HCV) RNA occurs through an internal ribosome entry site (IRES) located at its 5' end. As a positive-stranded virus, HCV uses the genomic RNA template for translation and replication, but the transition between these two processes remains poorly understood. HCV core protein (HCV-C) has been proposed as a good candidate to modulate such a regulation. However, current data are still the subject of controversy in attributing any potential role in HCV translation to the HCV core protein. Here we demonstrate that HCV-C displays binding activities toward both HCV IRES and the 40 S ribosomal subunit by using centrifugation on sucrose gradients. To gain further insight into these interactions, we investigated the effect of exogenous addition of purified HCV-C on HCV IRES activity by using an in vitro reporter assay. We found that HCV IRES-mediated translation was specifically modulated by HCV-C provided in trans, in a dose-dependent manner, with up to a 5-fold stimulation of the IRES efficiency upon addition of low amounts of HCV-C, followed by a decrease at high doses. Interestingly, mutations within some domains of the IRES as well as the presence of an upstream reporter gene both lead to changes in the expected effects, consistent with the high dependence of HCV IRES function on its overall structure. Collectively, these results indicate that the HCV core protein is involved in a tight modulation of HCV translation initiation, depending on its concentration, and they suggest an important biological role of this protein in viral gene expression.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Hepatitis C virus (HCV)1 is the primary causative agent of non-A and non-B hepatitis, frequently associated with high rates of progressive liver disease, leading to cirrhosis and hepatocellular carcinoma (1). Since its identification in 1989 (2), appreciation of the significant worldwide impact of chronic HCV infection has grown as ~170 million people, i.e. 3% of the world's population, are infected with HCV. HCV is an enveloped positive-stranded RNA virus whose genome of about 9600 nucleotides (nt) encodes a polyprotein that is processed by both host and viral proteases to yield individual structural and nonstructural HCV proteins (3).

HCV, along with pestiviruses and GB virus B (GBV-B), are members of the Flaviviridae family that utilize a cap-independent mechanism to initiate translation of their viral RNA that starts at the AUG codon (position 341 for HCV), following ribosome recognition of the upstream internal ribosome entry site (IRES) (4). HCV IRES is a cis-acting element, highly folded into a complex secondary and tertiary structure consisting of several stem-loop domains, which has been shown to be crucial for its functionality (reviewed in Ref. 5). Of the four domains delineated within the HCV 5'-untranslated region (5'-UTR), only the first seems not to be part of the IRES. The minimal sequence necessary for IRES activity extends approximately from nt 42 to a short stretch (12–30 nt) downstream of the initiation codon (6). However, the 3' border of the IRES element remains controversial because of conflicting data on the requirement of some viral coding sequences between the AUG codon and the reporter gene for its efficiency. The following explanation has been proposed to reconcile the different reports: the HCV IRES-dependent translation negatively correlates with the stability of the putative stem-loop IV involving base pairing between the extreme 3' end of the HCV 5'-UTR and the 5' proximal open reading frame following the initiation codon, located in the loop of that hairpin structure (7). Several studies have demonstrated a significantly enhanced expression of different reporters when part of the viral coding sequence is retained (6, 8–10), although no specificity in that coding sequence seems to be required, as demonstrated with chimeric constructs harboring either CSFV or HCV coding sequences (11).

The mechanism of the internal initiation of HCV IRES, because of its extreme sensitivity to secondary structure in the vicinity of the initiation codon, resembles the situation described in prokaryotic systems rather than that observed for picornavirus RNAs (12). First, there is formation of a binary complex between the basal part of domain III of the IRES (loop IIId) and the 40 S ribosomal subunit, followed by binding of the eukaryotic initiation factor 3 to the apical domain IIIb (13, 14). This complex is further stabilized by the apical loop of domain II, independently of any other canonical translation initiation factors, leading to the terminal 80 S ribosomal assembly by binding of the 60 S subunit and allowing translation initiation (12) of the viral RNA. However, it has been suggested that trans-acting noncanonical factors either of cellular or of viral origin interact with HCV RNA in the stem-loop region IV, tightly modulating translation (5). It appears, by analogy with picornaviruses, that the interaction of the HCV IRES with cellular factors has been studied more than the effect of HCV-encoded proteins on the HCV translation. Because of its particularity of promoting persistent infection, HCV has been suggested to possess a self-modulating mechanism to maintain a low level of replication. Therefore, it can be assumed that viral protein(s) might modulate RNA balance between the following three key steps of the viral life cycle: translation, replication, and packaging. Some of these proteins have been proposed to influence HCV IRES activity, such as core protein (15,16), NS4A and NS4B (17), and NS4B and NS5A (18). However, there are conflicting results, mostly those concerning the HCV core protein that still appears to be a good candidate that might interact with HCV IRES for the following reasons. First, it is supposed to be present in the infected cell with E1 and E2 glycoproteins after decapsidation as a component of the nucleocapsid and also to be the first protein synthesized in that cell due to its most N-terminal position in the HCV polyprotein. Second, in addition to its role as a structural protein in the assembly and morphogenesis to form viral nucleocapsids, HCV core protein is involved in a wide range of biological activities such as cell transformation (19), apoptosis (20), dysregulation of lipid metabolism (21), modulation of immunological functions (22), regulation of cell signaling (23), and of cellular transcriptional promoters (24) (for review see Refs. 25–27). Therefore, it is still the subject of debate as to whether HCV core protein acts as a translational regulator. Some controversial reports assert that the core protein-coding sequence modulates HCV IRES function, acting through a long range RNA-RNA interaction that is either nonspecific (28) or specific, as described between nt 24 and 38 within the HCV 5'-UTR and nt 428 and 442 of the core coding sequence (29). In contrast, other studies indicate that the HCV core protein reduces the efficiency of HCV IRES translation by binding to the IRES either (i) through more than two of the four identified clusters of basic amino acid residues (15) or (ii) through two different amino acids sequences of core protein described as follows: amino acids 34–44 (16) and amino acids 1–20 as well as the corresponding synthetic peptide (30). In either case, a down-regulation of the IRES activity was reported.

In this study, to gain insight into the potential role of the HCV core protein in influencing the efficiency of the HCV IRES-directed translation, we further examined the effect of the wild-type purified HCV core protein on the expression of reporter genes in an in vitro system. We found that the core protein provided in trans specifically modulates the HCV IRES activity in a dose-dependent manner. These findings suggest that the HCV core protein may play a role in the regulation of viral gene expression, which is discussed as one possible molecular mechanism facilitating viral persistence.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmid Constructs—Enzymes used for cloning and modifying DNA were from New England Biolabs and Roche Applied Science. Bicistronic vectors (vb) were derived from the pIRF vector, as described previously (10). Briefly, this plasmid contains the CMV and T7 promoters for transcription, the firefly luciferase (FLuc) gene, followed by the HCV 5'-UTR sequence and the Renilla luciferase gene (RLuc) in the pcDNA3.1Zeo (+) vector (Invitrogen). In this system, the upstream reporter FLuc (control) is translated in a cap-dependent fashion, whereas the downstream reporter RLuc (assay) is under the control of the HCV IRES. The constructs are described in Fig. 1. Vectors containing different HCV IRES sequences were obtained by inserting in the intercistronic region of pIRF, in place of the original HCV-1a IRES, HCV IRES sequences amplified from human patient sera by reverse transcription-PCR performed on the corresponding HCV RNAs and digested with BamHI and PstI enzymes. HCV IRES of genotype 1b was described in Ref. 31, and mutated IRESs of genotype 1a, as depicted in Fig. 1, have been reported in previous studies as follows: Q12 and Q22 in Ref. 10 and Q511 and Q519 in Ref. 32. Heterologous IRES sequences were amplified with appropriate primers containing BamHI and PstI sites (i) from pCITE vector (Novagen) for encephalomyocarditis virus (EMCV) IRES (586 nt) (33), (ii) from pA/C-4 (a gift from Dr. C. Wychowski) for classical swine fever virus (CSFV) IRES (Alfortville strain) (373 nt) (34), and (iii) from pSV CAT BiP Luc (kindly provided by Dr. P. Sarnow) for human immunoglobulin heavy chain-binding protein (BiP) IRES (220 nt) (35). Monocistronic vectors (mv) consist of (i) vm FLuc, containing the FLuc gene alone and cloned into pcDNA3.1, and (ii) vm IRLuc, the second plasmid concomitantly used with vm FLuc, which includes the RLuc gene expressed under the different IRES sequences, as described above, and inserted upstream of it.



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 1.
Schematic diagrams of different constructs used. The upper panel shows the bicistronic (left) and monocistronic (right) reporter vectors containing CMV and T7 promoters. In the bicistronic system, the upstream FLuc gene and the downstream RLuc gene are separated by different IRES sequences. The different IRES sequences used (*) are described under "Materials and Methods" and are indicated in the lower frame. Left, heterologous IRESs are as follows: two viral, EMCV and CSFV IRESs, and the cellular BiP IRES. Right, homologous HCV IRESs are based on the representation of Honda et al. (52) from genotype 1b. HCV IRES sequence consists of the complete 5'-UTR and 30-nucleotide core-coding sequence (nt 1–371). IRES from genotype 1a differs from that of genotype 1b by only a few nucleotides as specified in the text. IRES-1a mutants Q511, Q519, Q12, and Q22 as described elsewhere are depicted. I, II, III, and IV, four structural domains shaded; the initiator AUG codon in stem-loop IV is circled.

 
The constructs for expression of HCV and GBV-B core proteins were obtained by inserting into the pT7-7 (His6) vector (36) the following NdeI-PstI-digested fragments as amplified previously by PCR: (i) a 507-bp fragment corresponding to amino acids 1–169 for the HCV-1b core protein (GenBankTM accession number D89872 [GenBank] ) encoded by the plasmid pCMV-C980 (a gift from Dr. K. Shimotohno (37)); and (ii) a 354-bp fragment corresponding to amino acids 1–118 for GBV-B (GenBank TM accession number AF179612 [GenBank] ) encoded by the plasmid pGBB (a gift from Dr. J. Bukh (38)). The correct orientations and sequences of all these constructs were confirmed by nucleotide sequencing.

Purification of Core Proteins and 40 S Ribosomal Subunit—Core-(His6) fusion proteins were produced in the inclusion bodies and purified as described previously (39). Briefly, Escherichia coli strain BL21 SI (Invitrogen) was transformed with the plasmids encoding the respective core proteins of HCV and GBV-B. Transformants were grown at 37 °C in Luria broth medium without sodium chloride (NaCl), and expression of the recombinant fusion protein was induced by adding NaCl to a final concentration of 200 mM. Thereafter, two extractions were performed with 6 M urea in the presence of 10 mM {beta}-mercaptoethanol, and the proteins were purified on a nickel-nitrilotriacetic acid-agarose column (Qiagen) by elution with 250 mM imidazole and were submitted to a step of reverse phase high pressure liquid chromatography. After lyophilization, each of the proteins was dissolved in 20 mM Tris-HCl, pH 7.4, 1 mM dithiothreitol buffer, containing or not 0.1% n-dodecyl maltoside (DM). The overall conformation and the oligomeric properties of the core proteins in solution were investigated by using circular dichroism, analytical centrifugation, intrinsic fluorescence measurements, and proteolysis with endoproteinase Glu-C. The results indicated that core proteins produced aggregate in the absence of detergent but behave as soluble {alpha}-helical folded proteins in the presence of mild detergent. The folded form displayed an exchange between monomer and dimer state.2

HIV-1 IIIB p24 core was from Advanced Biotechnologies, Inc. (Tebu, France), and consisted of the native viral protein obtained from HIV-1 IIIB-infected HUT76 cells, purified by immunoaffinity, and solubilized in phosphate-buffered saline, pH 7.2. The 40 S ribosomal subunit was prepared by zonal centrifugation, as described previously, using free polysomes (40).

In Vitro Transcription—Plasmids were linearized at the NotI site located immediately downstream of the gene of interest. The linearized sequences were purified by QIAquick PCR purification kit (Qiagen) and served as templates for in vitro transcription of uncapped RNAs from the T7 promoter using large scale RNA production T7 RiboMAX system (Promega). After a 90-min incubation at 37 °C followed by treatment with RQ1 DNase (Promega) for 15 min at 37 °C, the RNAs synthesized were extracted and purified using the RNeasy kit (Qiagen). Then their content was determined by UV spectrophotometry, and their integrity was confirmed by SDS-agarose gel electrophoresis. The 32P-labeled HCV IRES RNA was obtained from vm IRLuc plasmid linearized with PstI, by in vitro transcription, and treatment with T4 polynucleotide kinase (Promega) in the presence of [{gamma}-32P]adenine triphosphate (3000 Ci/mmol, Amersham Biosciences), under standard conditions.

Western Blot Analysis—Protein analysis by Western was performed with standard protocols after migration on a 12.5% SDS-polyacrylamide gel. After transfer to a nitrocellulose membrane (Amersham Biosciences), the HCV core protein was probed with primary monoclonal anti-core antibody (19D96D6; gift of Dr. C. Jolivet-Reynaud) (1:10,000) and horseradish peroxidase-conjugate goat anti-mouse monoclonal antibody (1:3,000) as secondary reagent and then revealed by the ECL+ detection procedure (Amersham Biosciences).

Sucrose Gradient Centrifugation Analysis of 40 S Ribosomal Complexes—Purified 40 S ribosomal subunit (5 pmol), purified HCV core protein 1b (2.5 pmol), and 32P-radiolabeled HCV IRES RNA (2.5 pmol) were incubated according to different combinations in a final volume of 100 µl of gradient buffer (20 mM Tris-HCl, pH 7.4, 25 mM KCl, 5 mM MgCl2, 20 mM {beta}-mercaptoethanol, and 0.1% dodecyl maltoside) for 15 min at 30 °C. The reactions were then layered over 5–20% (w/v) linear sucrose gradient and centrifuged in a Beckman SW 60 swing bucket rotor at 50,000 rpm for 1 h at 4 °C. Fractions (400 µl) were collected from the bottom of the tube, and the absorbance profile at A280 was monitored. Alternatively, the total radioactivity in each gradient fraction was estimated in a Packard Tri-Carb counter, and the presence of core protein was determined after 10% (v/v) trichloroacetic acid precipitation, followed by SDS-PAGE and Western blotting.

In Vitro Translation and Luciferase Assay—In vitro translations in micrococcal nuclease-treated rabbit reticulocyte lysates (Promega) were performed according to the supplier's instructions, after a 1-h incubation at 30 °C, in a final volume of 10 µl containing 200 ng of RNA, with the addition, when appropriate, of 1 µl of core protein in increasing amounts varying from 10 to 1000 ng. When starting from plasmid DNA constructs, the TNT-coupled reticulocyte lysate system (Promega) was primed with 200 ng of linearized DNA and increasing concentrations of core protein for transcription-translation with T7 RNA polymerase as indicated by the manufacturer in a 10-µl reaction mixture for 60 min at 30 °C. DNA and RNA amounts were in the translation linear response range for the corresponding batch of lysate.

Luciferase activities were assayed on 6 µl of the reaction mixture by using dual-luciferase reporter assay system (Promega). The FLuc and RLuc values were expressed relative to the control transcript without the addition of HCV-C protein, which was assigned a value of 1 (100% efficiency). Errors were calculated as the standard deviation of three calculated IRES activities and expressed as percentages of the average activity.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
HCV Core Protein Interferes with the Formation of the HCV IRES-40 S Ribosomal Subunit Primary Complex—As an initial examination of the impact of HCV core (HCV-C) protein on IRES-directed translation, we investigated the potential interaction of that viral protein with the HCV IRES-40 S complex. Because the HCV IRES specifically binds to the 40 S ribosomal subunit, even in the absence of any initiation factors, it is likely that additional factor(s) may modulate internal initiation of translation, interfering with IRES-40 S assembly. We addressed this possibility for HCV core protein, taking into account that it has already been reported to interact with HCV IRES (as mentioned above) and with the ribosome (41). To this end, we analyzed the potential interaction between the three constituents, HCV IRES, 40 S ribosomal subunit, and HCV-C protein, by ultracentrifugation on sucrose gradients of the complexes resulting from different combinations between them (Fig. 2). After characterization of the constituents taken individually, their putative association with each other was assessed either by counting radioactivity for IRES or by immunoblotting for core protein. IRES RNA-core association was confirmed (Fig. 2, B and C-b) in fractions 2–6, suggesting the formation of complexes slightly entering the gradient. Moreover, a complete cosedimentation of the 40 S subunit with HCV-C protein was observed (Fig. 2C-c). The sedimentation pattern of IRES plus the 40 S subunit indicated that the HCV IRES was recovered in two regions, in addition to the top of the gradient (Fig. 2B): fractions 7–9 corresponding to the association with 40 S, and fractions 2–5 that might result from interaction of IRES with ribosome element(s), due to the background noted in the distribution of the absorbance profile at 280 nm (Fig. 2A). Accordingly, it has been reported recently that HCV IRES exhibited a strong affinity for the ribosomal S5 protein that prevented the recruitment of the 40 S subunit (42). Therefore, it appears in this experiment that any event that might impair the observed unexplained binding of IRES in fractions 3–6 could improve IRES-40 S interaction. This happened in the last step, with the addition of purified core protein as shown in Fig. 2, B and C-d, with a shift in IRES distribution, corresponding to an enhanced association of IRES-40 S in fractions 7–9. Taken together, these data indicate that HCV core protein displays binding activity toward both HCV IRES and the 40 S ribosomal subunit. Moreover, they suggest that HCV-C may help IRES RNA-40 S assembly.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 2.
Interference of HCV core protein with HCV IRES-40 S ribosomal subunit primary complex. Reaction mixtures containing 40 S ribosomal subunit, purified HCV core protein 1b, and 32P-labeled HCV IRES RNA according to different combinations were overlaid onto a 5–20% sucrose gradient and centrifuged for 1 h at 4 °C. Gradient fractions were collected from the bottom as indicated (fractions 1–9), and the components in each fraction were analyzed as described under "Materials and Methods." Distributions of 40 S ribosomal subunit (absorbance at 280 nm) (A), of labeled HCV IRES (radioactivity, cpm) (B), and of HCV core protein (Western blots a–d) (c) are shown.

 
HCV Core Protein Stimulates HCV IRES Activity in a Dose-dependent Manner—The results reported above prompted us to elucidate which regulatory mechanism, if any, would modulate HCV IRES efficiency through HCV core protein. We first used an in vitro coupled transcription-translation TNT system. All HCV IRESs used contained the first 30 nucleotides of the viral coding sequence, as described previously (10), between the HCV initiation codon and the Renilla luciferase (RLuc) reporter. The core protein of genotype 1b was expressed in a prokaryotic system and purified as described under "Materials and Methods." In this study, we used amino acids 1–169 of the protein, which is similar in size to the mature protein as suggested earlier by Santolini et al. (41). The TNT mix was primed with 200 ng of bicistronic DNA reporter vector vb IRF-1b containing the genotype 1b HCV IRES (31) and supplemented, when appropriate, with increasing amounts of purified recombinant HCV core protein diluted in buffer with or without DM. This mild detergent was expected to avoid core aggregation as the protein has the propensity to multimerize via an interacting domain mapped within its N-terminal hydrophobic region (amino acids 36–102) (43, 44). Because an equilibrium between both dimeric and multimeric forms was suspected in the infected cell (43), we were interested in testing their respective effects on HCV IRES activity.

Luciferase activities were measured by using the Stop and Glo dual luciferase assay system (Promega) and were plotted against HCV-C concentrations (Fig. 3A, left). The results showed modest enhancement of RLuc activity (Fig. 3A, left, shaded bars) at increasing concentrations (10, 100, and 1000 ng) of core protein. However, the activity observed for FLuc, expressed in a cap-dependent manner, was almost the same (Fig. 3A, left, striped bars). Therefore, the effect of core addition on HCV IRES efficiency could be evaluated by comparing the ratio of RLuc activity relative to that of FLuc, with the ratio obtained with the same vector in the absence of core protein, arbitrarily considered to be 1 (Fig. 3B, left). A gradual increase was observed upon the addition of HCV-C protein to the translation reactions, suggesting an impact of the viral protein on the HCV IRES-mediated translation. This effect is dose-dependent, because RLuc expression was the highest in the presence of 1000 ng of core protein, with a relative IRES efficiency of 1.84 when diluted in buffer with DM (Fig. 3B, left, solid bars). When diluted in the absence of DM (Fig. 3B, left, open bars), i.e. under conditions favoring its aggregation, the core protein was less potent, with a 1.45-fold increase at 1000 ng, suggesting that the multimeric state may have an impact on its effect.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 3.
The stimulatory effect of exogenous addition of purified HCV core protein 1b on HCV IRES 1b translation efficiency. In vitro coupled transcription-translation (TNT) reactions were programmed with 200 ng of bicistronic reporter vector vb IRF of genotype 1b (left panels) and with 200 ng of each of monocistronic vectors vm FLuc plus vm IRLuc (right panels). Varying amounts of the core protein (10–1000 ng) were added the respective reporter translation reactions as depicted on the x axis. A, luciferases activities are indicated in the presence of increasing amounts of HCV-C protein diluted in buffer containing DM. FLuc activity (cap-dependent) (striped bars); RLuc activity (HCV-IRES-dependent) (shaded bars). B, changes in HCV-IRES-directed RLuc activity relative to FLuc activities upon increasing concentrations of HCV-C protein diluted in buffer with DM (solid bars) and without DM (open bars). Relative HCV IRES efficiency was estimated from proportional luciferase activity (RLuc/FLuc) of in vitro translation products with that of the corresponding vector in the absence of core protein normalized to 1 (100% activity). Left panel, bicistronic context; right panel, monocistronic context. Each column represents the average IRES activity from triplicate translation samples, and error bars indicate the standard deviations.

 
In a second step, (i) considering the well documented structure of HCV IRES as well as its diverse interactions with cellular factors (5), and (ii) because it has been reported that the context of an IRES element may be important for its activity (45), it was tempting to determine whether HCV-C protein influences the IRES function according to its context within the plasmid reporter vector used. We therefore reproduced the same experiment as described above with monocistronic vectors vm FLuc and vm IRLuc. As shown in Fig. 3A, right, shaded bars), RLuc activities regularly increased and were higher than in a bicistronic context, whereas those obtained for FLuc were not affected in the presence of any amount of core protein (Fig. 3A, right, striped bars). Therefore, a gradual enhancement of HCV IRES relative efficiency was observed in proportion to core addition, with an RLuc/FLuc ratio above that obtained in a bicistronic context, up to 3-fold in the presence of 1000 ng of HCV-C protein (Fig. 3B, right, solid bars). These results suggest easier accessibility of the HCV IRES in the monocistronic context. Moreover, values obtained for core dilution in the absence of DM (Fig. 3B, right, open bars) were still below those obtained with dilution in DM (solid bars), with an RLuc/Fluc ratio of 1.72 versus 3 in the presence of 1000 ng of core protein. We did not succeed in gathering relevant data (either increase or decrease in HCV IRES activity) with a larger amount of core protein (such as 5000 ng), even in a greater reaction volume.

HCV Core Protein Acts on HCV IRES Activity at the Translational Level—The TNT system was convenient for roughly assessing the potential impact of HCV-C protein on the corresponding IRES efficiency. However, our next goal was to specify this effect, taking into account the two following points. First, the HCV core protein has been implicated in interactions with DNA molecules due to its overall high basic charge through nonspecific binding to negative DNAs and through regulation of transcriptional host and viral promoters (24, 46). Moreover, a cellular promoter activity has been depicted recently within the HCV IRES (47). Consequently, it was considered important to exclude these putative interfering mechanisms in order to clarify whether the modulation of the IRES activity by the core protein, observed above, occurs at the translational level. Second, when the transcription-translation-coupled TNT system is used, the estimation of the IRES RNA template interacting with core protein, and their stoichiometric ratio, is not available. In this regard, further experiments were performed starting with RNA instead of DNA, restricting the observed events to the translation and allowing quantitation of RNA. Rabbit reticulocyte lysate reaction mixture was programmed with the corresponding RNAs transcribed in vitro from either bi- or monocistronic reporter plasmids containing HCV IRES and supplemented with increasing amounts of core protein. Similarly to what was observed with the TNT system, changes in luciferases activities were observed only for RLuc, although the FLuc activity was almost the same upon any addition of HCV core protein (data not shown). Therefore, the effect of core on HCV IRES efficiency could be estimated through the activity of RLuc relative to FLuc (Fig. 4). As RLuc expression is directly under the control of HCV IRES, and the only one to vary upon supplementation with core protein, this reflects the effective influence of HCV-C on translation efficiency. A dose-dependent effect was observed on HCV IRES activity but was slightly different in the two systems. In the bicistronic system (Fig. 4A, solid bars) a 2-fold maximal enhancement of translation (1.73) was achieved with 100 ng of HCV-C, although in the monocistronic system (Fig. 4B, solid bars) a stimulation up to 5-fold (4.9) was induced by the addition of 1000 ng of core protein. One possible explanation lies in the fact that in a bicistronic context added HCV-C protein may rapidly become saturating due to the blockage resulting from the presence of the upstream cistron. As first underlined with the TNT system, these observations are consistent with the idea that in the absence of any upstream reporter gene, HCV IRES is more accessible to factors that putatively affect its function. Unlike assays performed with the TNT system, amounts greater than 1000 ng of HCV-C protein could be tested, which means at least up to 5000 ng to remain available in monitoring luminometric activities. Most interestingly, a decrease in HCV IRES efficiency was observed at such high concentrations, from 4.9 to 2.8, in the monocistronic system (Fig. 4B, solid bars), and a slight inhibition, from 1.73 to 0.96, in the bicistronic context (Fig. 4A, solid bars), strongly arguing in favor of a tight modulation of HCV IRES activity by the HCV core protein, according to its concentration. Moreover, as noted previously, the activation observed in this last series of assays was more effective when aggregation of core protein was prevented; compare solid bars (+DM) and open bars (–DM) in Fig. 4. However, in the presence of 5000 ng of core protein, this difference is less important, probably due to the fact that at high concentration HCV-C protein tends to oligomerize, even if diluted with detergent.



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 4.
HCV protein acts on HCV IRES efficiency at the translational level. In vitro translation reactions of rabbit reticulocyte lysates (RRLs) were programmed with 200 ng of transcripts obtained in vitro from bicistronic (A) or monocistronic (B) plasmid reporters and supplemented with increasing amounts of HCV-C protein diluted in buffer with (solid bars) or without (open bars) DM as indicated. Relative IRES efficiencies were determined as described in Fig. 3. Each column represents the average IRES activity from triplicate translation samples, and error bars indicate the standard deviations.

 
HCV IRES-directed Translation Is Specifically Modulated by HCV-C Protein Compared with Heterologous Viral Core Proteins—To determine whether the purified recombinant HCV-C protein was capable of a specific modulation of HCV IRES translation, we studied in vitro translation of reporter transcripts vb IRF-1b and vm IR-1b plus vm FLuc as described above in rabbit reticulocyte lysates, but replacing the HCV-C protein by the GBV-B core protein or the HIV-1 p24 core protein, as described under "Materials and Methods," whose size and affinity for RNA are close to those of the HCV core protein. These proteins were added to the translation reaction mixture in increasing amounts ranging from 10 to 1000 ng. In contrast to the modulatory effect observed for the HCV-C protein, none of the heterologous core proteins tested had any effect on HCV IRES activity in either a bi- or monocistronic context and even at the highest concentration (Fig. 5). These data suggest that the observed modulation of HCV IRES-directed translation is specific to the HCV-C protein.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 5.
Effect of heterologous core proteins on HCV IRES-dependent translation efficiency. RRL system reactions were programmed with bicistronic (A) and monocistronic (B) reporters, respectively, supplemented as depicted on the x axis with various core proteins from HCV (solid bars), HIV (shaded bars), or GBV-B (open bars), and diluted in DM at the indicated concentrations. Relative IRES efficiencies were determined as described in Fig. 3. The columns represent the means, and the bars the standard deviations of the experiment performed in triplicate.

 
HCV Core Protein Specifically Modulates HCV IRES Activity Compared with That of Heterologous IRESs—Alternatively, due to results obtained with heterologous core proteins, it was tempting to determine whether or not the modulating effect of HCV-C protein on IRES-directed translation was restricted to HCV IRES. For this purpose, the impact of HCV-C was assessed on different heterologous IRESs as follows: two viral IRESs (i) EMCV IRES, belonging to the Picornaviridae family, and (ii) CSFV IRES, belonging to the same Flaviviridae family as HCV, and BiP, one cellular IRES. HCV IRES in each bi- and monocistronic plasmid reporter was replaced by one of the other IRESs, and the resulting RNA transcripts were used to program in vitro reactions, as described above, in the absence or presence of increasing amounts of HCV-C protein (Fig. 6). Unlike what was observed for HCV IRES, luciferases activities obtained with bicistronic RNA containing EMCV IRES behaved differently upon addition of the HCV-C protein. FLuc activity dramatically decreased ~6-fold in the presence of increasing amounts of HCV-C (see Fig. 6A, left, striped bars). In contrast, RLuc activity did not change significantly (Fig. 6A, left, shaded bars). Consequently, it can be assumed that the apparent activation of the EMCV IRES relative efficiency (RLuc/FLuc), up to 7.85-fold, as depicted in Fig. 6B, left, was irrelevant. By taking into account that an inhibitory effect of EMCV IRES on capped RNAs translation has been described previously, presumably resulting from the ability of the J-K region within the IRES RNA to sequester cellular factors needed for cellular mRNA translation (48), this surprising result could be explained as follows. Rather than faithfully reflecting an actual role of HCV-C in EMCV translation activity, the observed effect might result from facilitation of the sequestration of cellular factors in the vicinity of EMCV IRES inserted between luciferase genes, thus being detrimental to cap-dependent FLuc translation.



View larger version (40K):
[in this window]
[in a new window]
  		FIG. 6.
Effect of HCV-C protein on heterologous IRESs. In vitro translation reactions of RRL were programmed with 200 ng of in vitro transcripts obtained from bicistronic (left panels) and monocistronic (right panels) reporter vectors containing the following heterologous EMCV (squares), CSFV (circles), and BiP (triangles) IRESs and supplemented with increasing amounts of HCV-C protein as indicated. A, luciferases activities obtained for transcripts containing EMCV IRES in the presence of increasing amounts of HCV-C protein diluted in buffer with DM; left, bicistronic RNA; right, monocistronic RNA; FLuc activity (cap-dependent) (striped bars); RLuc activity (EMCV-IRES-dependent) (shaded bars). B, relative IRES efficiencies (RLuc/Fluc ratios) determined as described in Fig. 3, after programmed of translation reactions with respective transcripts and addition in trans of increasing amounts of HCV-C protein diluted in buffer with DM. C, identical experiments were performed with HCV-C protein diluted without DM. Each value represents the average relative IRES activity from triplicate translation samples, and the error bars represent the standard deviation.

 
In contrast to EMCV, translation reactions programmed with transcripts containing CSFV or BiP IRES in the bicistronic context displayed luciferases activities with tendencies, upon addition of HCV-C protein, similar to those observed with the HCV IRES, i.e. no variation in FLuc activity in comparison to RLuc activity, which varied in amplitude according to the IRES considered (data not shown). Thus, due to stabilization of FLuc activity, the variations in CSFV and BiP IRES efficiencies could be assessed through their relative RLuc activity (Fig. 6B, left). Accordingly, for CSFV IRES, the enhancement was ~2-fold (1.93) on addition of 100 ng of HCV-C protein, and then tended to plateau (1.80-fold) between 100 and 1000 ng of recombinant protein. The increase observed at 100 ng of core protein, close to that of 1.84 depicted in Fig. 3B, left (solid bars) for HCV IRES activity in the same conditions, could be explained by a possible interaction between HCV-C protein and the loop IIId domain of CSFV IRES in the presence of a small amount of HCV protein. The loop IIId domain is well conserved among both HCV and pestiviruses (49), and its secondary structure is the most efficient to interact with HCV core protein (50). Such an interaction may become irrelevant at higher concentration, leading to a plateau. Unlike CSFV IRES, a slight 0.7-fold decrease was obtained for BiP IRES activity (Fig. 6B, left). Noteworthy, luminometric measures were unsuccessful in the presence of 5000 ng of HCV core protein. Therefore, subsequent analyses performed with heterologous IRESs were limited to the addition of 1000 ng of HCV-C protein. Moreover, for each IRES considered, the relative similarity of the curves obtained with and without DM (Fig. 6C left, compare B and C), although efficiency was slightly lower in the absence of DM, suggests the lack of actual influence of HCV-C in either state, aggregated or not, on heterologous IRESs.

In the case of monocistronic systems, patterns for luciferase activities for each of the heterologous IRESs tested were comparable with those obtained in a bicistronic context, except for EMCV IRES (Fig. 6A, right). In that case, RLuc activity still stabilized around 40 x 106 light units (Fig. 6A, right, shaded bars), whereas that for FLuc dramatically decreased upon addition of core protein (striped bars), rendering the estimation of the EMCV IRES relative activity in those conditions inappropriate to reflect any actual modulation of its efficiency by HCV core protein. The apparent stimulation observed in the presence of HCV-C protein in this monocistronic context (Fig. 6B, right), which was less than that noted in a bicistronic context, could be explained by facilitation of the primary tendency of EMCV IRES free of any upstream reporter gene to sequester cellular factors, thus dramatically reducing the cap-dependent translation. However, increasing amounts of HCV-C protein might be responsible when accumulated around EMCV IRES for the partial impairment of its attraction of cellular factors to the benefit of capped FLuc RNA, which therefore becomes translatable, leading to a plateau. Unlike EMCV, because of stabilization of FLuc activity, whatever the amount of added HCV-C protein, the respective variations in efficiency of CSFV and BiP IRESs could be investigated through relative RLuc activity (Fig. 6B, right). For CSFV IRES, a 4-fold stimulation (4.14) occurred upon the addition of 100 ng of HCV-C, and a ratio of 4.09 was obtained with 1000 ng. Such an increase could be explained by the same mechanism as in the bicistronic system, i.e. by binding of HCV-C protein to the IRES IIId domain, though somehow facilitated in the monocistronic context. Furthermore, the tendency to reach a plateau at 1000 ng of HCV-C might result from a saturating nonspecific impact of core protein. Such a common behavior of EMCV and CSFV IRESs at high concentrations of HCV-C protein underlines its nonspecificity. Moreover, in a monocistronic context, no effect on BiP IRES activity was observed upon addition of HCV-C protein (Fig. 6B, right). This was expected because BiP IRES is unrelated to HCV IRES. In addition, less structuring of its sequence, because of its shortness, might account for the absolute lack of susceptibility to the HCV core protein. IRES activities were substantially lower when HCV-C protein was aggregated (buffer – DM) than when diluted in DM (Fig. 6C, right, compare B and C). Profiles were almost identical for EMCV and CSFV IRESs, indicating an apparent increase in their efficiency as HCV-C protein concentration increased. These results are not completely ambiguous and suggest that in aggregated form HCV-C protein might exert a nonspecific stimulatory effect even at low concentrations (see "Discussion"). Altogether, these data are in favor of a specific interaction of HCV-C protein with homologous IRES and reinforce the importance of the context of the IRES tested for its efficiency.

Modulation of HCV Internal Initiation of Translation by the Viral Core Protein Is Tightly Dependent on the Overall Integrity of the HCV IRES Structure—To assess further the extent of such specific interference of HCV-C protein in HCV IRES-directed translation, we examined its impact on HCV IRES variants. To this end, we used the HCV-1a IRES as contained in pIRF vector (vb IRF-1a) and various natural mutants previously isolated and described under "Materials and Methods" and in Fig. 1. HCV-1a IRES differs from HCV-1b by only a few changes, which are clustered at four loci (UGA, nt 11–13 instead of GAU, GA at nt 34–35 instead of AG, C204A, and G243A) (51). Mutations in the variants studied were located as follows: Q519 (C18T) in domain I, Q511 (G76A) in domain II, Q22 (G266A) in domain IIId, and Q12 (G301C) between domains IIIe and IIIf.

Like HCV IRES 1b, in both bi- and monocistronic contexts, luciferase activities obtained for each reporter gene exhibited the same tendency for HCV IRES 1a and corresponding variants, i.e. gradual variations for IRES-dependent RLuc activities and almost stable values for FLuc translation (data not shown). Therefore, the respective efficiencies of these different IRESs could be drawn as shown in Fig. 7. In a bicistronic context, maximum relative activity of HCV IRES 1a upon addition of increasing amounts of HCV-C-1b protein was about that observed with IRES 1b, i.e. 1.93 versus 1.84 with 100 ng of core protein in the monomeric state (Fig. 7A, left) and 1.69 versus 1.76 in the multimeric state (Fig. 7B, left). However, this activity drastically decreased to 0.83 with 1000 ng of HCV-C protein in buffer with DM and to 0.55 in the absence of DM. This suggests a less stringent effect of HCV-C 1b on HCV IRES 1a than in the closest homologous system (HCV-C 1b/HCV IRES 1b) for which a decrease in activity was observed only with amounts of HCV-C higher than 1000 ng.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 7.
Effect of HCV-C protein on HCV IRES 1b and different variants. In vitro translation reactions of RRL were programmed with 200 ng of in vitro transcripts obtained from bicistronic (left) and monocistronic (right) reporter plasmids containing the following HCV-1a IRES (squares) and different mutants Q12 (open triangles), Q22 (open squares), Q511 (triangles), and Q519 (circles), as described under "Materials and Methods," and supplemented with increasing amounts of HCV-C protein as indicated, diluted in buffer with DM (A) and without DM (B). Relative IRES efficiencies were determined as described in Fig. 3. Each value represents the average relative IRES activity from triplicate translation samples, and the error bars represent the standard deviation.

 
We further assessed the activity of mutated IRES 1a in order to determine whether some variations in overall structure of the HCV IRES 1a might either restore the relative efficiency obtained with IRES 1b, or emphasize the difference observed between the two genotypes. That purpose was based on the following observations. (i) The HCV IRES whole conformation is fully required for its efficiency (52). (ii) The HCV IRES-40 S association, the primary step for HCV translation initiation, requires an intact IIId loop to allow 40 S subunit binding to IRES RNA (14). (iii) Mutations in the bulge loop IIId, mostly those affecting the G content, are deleterious in the HCV-C/IRES interaction.3 Patterns of curves obtained with vb Q511 and vb Q519 mutants were almost identical to that of vb 1a with less activity (Fig. 7A, left), and still slightly lower in buffer without DM (Fig. 7B, left). However, both vb Q22 and vb Q12 mutants gave different results, with a gradual inhibition of the IRES function on addition of HCV-C core protein, to 0.73 for vb Q22 and to 0.27 for Q12. When HCV-C protein was diluted in buffer without DM, data obtained with both of these mutants were not significantly different from those obtained with DM, suggesting that HCV-C protein exerts a nonspecific effect on the activity of these IRESs.

In a monocistronic context (Fig. 7, A and B, right), HCV 1a IRES was as much stimulated as IRES 1b (3.69 versus 3.61) in the presence of 100 ng of monomeric HCV-C. However, this effect rapidly decreased to 2.45 after addition of 1000 ng of HCV-C protein, instead of continuing to increase as observed for IRES-1b (Fig. 4B). The vm Q519 mutant behaved exactly the same, which was not surprising due to the location of the mutation (C18T) in the first structural domain I, which is not considered to belong to the HCV IRES sequence (6). Similarly, the vm Q511 mutant exhibited approximately the same efficiency in the presence of HCV-C protein as that of vm 1a and vm Q519 IRESs, except for a slightly smaller decrease with 1000 ng of HCV-C protein (3.43 versus 2.45 and 2.51), tending to a plateau. The G76A mutation responsible is outside the region involved in RNA-RNA and RNA-protein(s) interactions. In the presence of the multimeric HCV-C protein (Fig. 7B, right), these mutants reacted differently with higher stimulation of both mutants vm Q511 and vm Q519 than vm 1a with 100 and 1000 ng of core protein, respectively, suggesting the importance of the form of HCV-C interacting with HCV IRES, leading to the possibility of an unexpected interaction with core when aggregated.

The variants vm Q22 and vm Q12 behaved differently in that context. A surprising initial phase of activation was observed with 10 and 100 ng of HCV-C, up to a factor approaching 2.5, followed by near inhibition with a factor <1 (0.75 and 0.85, respectively) in the presence of 1000 ng of core protein (Fig. 7A, right). These results might be explained during the first phase by interaction between HCV-C protein and IRES, helping IRES to display a conformation recognizable by the translation machinery (40 S subunit). In the second phase, at a higher concentration of core protein, the IRES conformation might be unsuited to subsequent positive interaction with the 40 S subunit due to a saturating mechanism, leading to partial inhibition of IRES activity. Such an interpretation is emphasized by examination of the data obtained with the same mutants in a bicistronic context (Fig. 7, A and B, left). Whatever the concentration of HCV-C, the near inhibition of the corresponding IRESs was observed, probably accounting for their inaccessibility by the core protein. Data obtained in the presence of the HCV-C protein diluted in buffer without DM (Fig. 7B, right) appeared to be identical to those with DM, suggesting that in the case of IRES harboring Q22 and Q12 mutations, and thus important structural changes, HCV core protein was expected to act as a helper rather than a specific modulator of translation initiation.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Because of conflicting data on the modulatory mechanism of HCV IRES translation initiation by HCV core protein, in this study we attempted to clarify the impact of increasing amounts of a purified HCV-C recombinant protein, provided in trans, on the expression of a reporter gene in an in vitro system. From the data obtained, four major conclusions can be drawn. First, HCV-C protein substantially modulates HCV IRES function in a dose-dependent manner. Second, the effects observed occur at the translational level. Third, they are specific to both interacting partners HCV IRES and HCV-C protein. Fourth, the modulatory effects concern the fidelity of translation initiation, being tightly dependent on the overall structural integrity around the AUG initiating codon. Moreover, even though nonspecific, a non-negligible impact of HCV-C protein could be observed on EMCV or CSFV IRES activity. This effect might be explained, being amplified with the aggregated form of HCV-C protein, by molecular attraction between basic domain(s) of HCV-C protein and positively charged RNA sequences that would confer on HCV core protein a kind of chaperone role toward some cellular factors gathered around the IRES. HCV core protein has been described recently to exhibit a chaperone activity toward nucleic acids (53), but such activity has not yet been tested on HCV IRES.

Our study strongly supports a relevant biological role for HCV core protein in IRES efficiency, showing it to have a tight modulatory effect, depending on its concentration, and not restricted to inhibition as reported previously by others. Although the interpretation of such a discrepancy remains unclear, it might result from several experimental features as follows: (i) the choice of reporter system and, more importantly, as demonstrated here, the importance of its use in a mono- or bicistronic context; (ii) the method of purification and storage (i.e. the use of detergent in buffer) for recombinant HCV-C protein tested; and (iii) the ratio between core protein and IRES RNA interacting with each other. Concerning the first point about the relative proximity of IRES and core protein at the translation level in assessing HCV IRES efficiency, our results point to its importance through the comparative use of bi- and monocistronic systems. Increased relevance in a monocistronic context can be easily justified due to the 5' position of the IRES regulatory sequence within the HCV genome, without any other upstream coding sequence, which would be expected to disrupt the overall structure, thus hampering interaction with cellular and/or viral factors. This clearly appears in Table I in which the relative efficiencies of IRESs studied without any addition of HCV core protein, taken as 1 (100% efficiency when calculation of relative RLuc activity), are depicted in both contexts. This is emphasized as follows: (i) for EMCV IRES (1.4 in bicistronic context versus 19.7 in monocistronic context) due to its ability to sequester cellular factors, as mentioned above; and (ii) for BiP IRES (0.4 versus 6), because of its small size. Moreover, this is confirmed when considering the impact of HCV-C on the IRES efficiency placed in a monocistronic context, which is approximately twice (2.8 with homologous system HCV-C 1b/IRES 1b and 1.9 with HCV-C 1b/IRES 1a system) that in the situation in which IRES is inserted between the two reporter genes.


View this table:
[in this window]
[in a new window]
  		TABLE I
Comparison of relative IRES efficiencies from IRESs used in this study and in absence of HCV core protein

 
In this report we limited our study to an in vitro assay for which we obviously cannot ascertain the native conformation of the HCV-C protein and RNAs used. Nevertheless, concerning the recombinant core protein produced, a lot of precautions were taken to ensure a high degree of purity (39) and satisfactory solubilization by adding 1% dimethyl maltoside to buffer. The modulation of HCV IRES function we observed, which depended on the amount of HCV-C protein, argues in favor of an RNA-protein interaction, rather than an RNA-RNA interaction. Binding of HCV core protein to the HCV 5' end RNA has already been described, specifically (15, 50, 54) or not (41), in order to elucidate the encapsidation mechanism without really attempting to assess its involvement in translation, except for the work of Shimoike et al. (15). No mechanism would seem to be favored over any other. Nonetheless, it can be assumed that both events might be distinguished because of their order of appearance within the life cycle. Accordingly, the following have been reported: (i) the multimerization state of HCV core protein, which is common to both its membrane-bound and free forms, might occur at a later stage of the viral cycle (43), and (ii) HCV-C protein appeared more active on IRES efficiency in the monomeric or dimeric state than when multimerized (the present study), thus reinforcing the possibility of sequential and multiple involvement of HCV core protein in the infected cell.

Concerning the conformation of the different HCV IRESs used, we are confident of their integrity because of the following well known structural/functional features of HCV IRES: (i) the adoption of an autonomous folding of the 5'-UTR structure depending on the ionic environment (55), and (ii) the prokaryotic-like mechanism of HCV translation initiation that does not require most of the eukaryotic initiation factors (12). Accordingly, HCV IRES can be considered not to be hampered in its function in an acellular system, compared with in vivo, despite the low cellular factor content of rabbit reticulocyte lysates. Moreover, the use of HCV IRES 1a mutants revealed a satisfactory sensitivity of the assay used to investigate their respective efficiencies as described above. The conclusions obtained are in agreement with the following: (i) recent work reporting the strict requirement of intact loop IIId to allow primary interaction between HCV IRES and 40 S subunit (14), and (ii) the interference of HCV core protein in the binary IRES-40 S complex assembly as suggested in Fig. 2. In addition, as already well documented, our results confirm a better efficiency of the HCV IRES 1b than that of genotype 1a, implying a further selection between variants in the infected cell when coinfection with several genotypes has occurred.

Furthermore, an advantage that deserves to be underlined in our in vitro assay is the possibility of determining the exact ratio between the RNA of interest (here different IRESs) and the core protein potentially interacting with each other. This ensures high data reproducibility, which could not be achieved in vivo where interaction of the partners, even after their calibrated input into the cell, depends on parameters such as transfection efficiency and functional features specific to the recipient cell. However, because of the lack of an efficient cellular system for producing HCV, we do not have any idea of the relative proportion of HCV RNA IRES and HCV core protein interacting with each other. Recently, an average value of 25 pg (ranging from 7 to 56 pg per cell) has been estimated in single hepatocytes by means of laser capture microdissection (56). Moreover, an average value of 33 viral genomes (ranging from 7 to 64 RNA molecules) is currently accepted per productively infected hepatocyte (57), thus enabling determination of the ratio of HCV-C protein to RNA potentially able to interact in one infected cell. Consequently, the relatively high proportion of 24 x 106 molecules of core protein (C) for one of viral RNA genome (IRES extremity) was found (24 x 106 C/1 IRES) (Table II), arguing for a biological impact of HCV core protein in the infected cell. When calculated in our assays, proportions varied from one molecule of HCV core protein (1C) to 500 (500C) for one of RNA (IRES) when supplementation was carried out with 10 ng (1 C/1 IRES) to 5000 ng (500 C/1 IRES) of HCV-C protein, respectively (Table II). The ratio was thus significantly 4 x 104-fold lower, even with the highest concentration of core protein tested (5000 ng), than that expected in the cell. In other words, a rather small amount of HCV-C protein (100 molecules) is sufficient in our system to exert a 5-fold stimulation of one molecule of HCV IRES, although a little more (five times) HCV-C protein, which is still little compared with its cellular availability for one molecule of HCV RNA, is already enough to diminish this effect nearly 2-fold. This is indicative of a certain quality of the HCV core protein used in this work, and of the relevance of its sharp modulatory effect on HCV IRES efficiency, excluding any artifactual IRES-core protein interaction due to saturation of RNA with core protein. Obviously, HCV core protein detected in the cell interacts not only with HCV IRES but is also involved in other cooperative functions because of its broad cell distribution (25–27).


View this table:
[in this window]
[in a new window]
  		TABLE II
Comparison between HCV-C protein to HCV IRES proportions (core protein/IRES) observed within the infected cell and those calculated in our in vitro assay

 
Based on the poliovirus model for which the nonstructural (NS) proteins 2A and 3CD, with the aid of cellular factors, regulate coordination between translation and replication (58), the two mechanisms being unable to take place simultaneously (59), HCV core protein has been proposed to play a role in viral persistence by down-regulating the translation process to the benefit of replication (15, 16, 30). Accordingly, a model has been proposed by Zhang et al. (16), suggesting a specific arrest of the HCV genomic RNA translation by core binding to the IRES, when all viral proteins are sufficiently synthesized for replication, and then allowing initiation of the negative-strand synthesis. However, such a scheme proposes that core protein at cellular high levels impairs HCV IRES, without addressing the possibility that lower levels may have a different effect. In view of our results, an alternative, albeit complementary, pathway of viral life cycle events is possible, considering the importance of the core concentration for its modulatory effect. Why might the HCV-C protein not play a stimulatory role at an early stage of the infection? Keep in mind that in the infected cell (i) it remains close to viral RNA after decapsidation, and (ii) it is the first protein expressed from the polyprotein. Thus, it would seem extraordinary that this protein might exert an inhibitory effect on the subsequent expression of other viral proteins rather than promote their accumulation to initiate further viral life cycles, ensuring viral propagation inside the infected cell. According to this new scheme, the events leading to transition from translation to replication might be as follows. A first step would involve HCV-C protein for active translation, in order to produce each of the individual viral proteins to a level consistent with the formation of replication and assembly complexes at endoplasmic reticulum membranes. Later on, when the process is sufficiently under way and the concentration of core protein tends toward saturation, its own modulatory effect on HCV translation initiation might be less important. Therefore, it is not excluded that other viral proteins such as nonstructural protein(s) might cooperate in regulation of the IRES function, thereby facilitating initiation of replication and taking over from the core protein, which could then become available to exert some of its multiple functions such as multimerization at the endoplasmic reticulum membrane for virus assembly, attachment to lipid droplets (60), and putative involvement in diverse cellular processes (25–27). A potential cooperative effect of another viral protein has already been investigated in two studies, although their data are at odds, one claiming an inhibitory effect of both NS4A and NS4B on HCV IRES (17), and the other a stimulatory effect of NS4B and NS5A (18). Such a discrepancy concerning NS proteins, in addition to the controversy mentioned for HCV-C protein, in their modulation of HCV IRES function, highlights the difficulty of ascertaining any biological relevance of a result aimed at elucidating translation initiation mechanisms, due to the absence of any in vitro infection system. To date, replicons, even though the most efficient systems, are not available to study the regulatory mechanisms of HCV IRES translation nor to distinguish them from replication. Rather, reporter systems have been designed as convenient tools for in vitro and in vivo evaluation of the HCV IRES activity on the expression of the gene located downstream of it.

In vivo systems are useful both to determine the relevance of a possible biological effect detected in vitro, and to approach mechanisms occurring in the cell. Nonetheless, in the absence of any available virus infection system, in vivo affordable systems can mimic, although not replicate, a putative biological mechanism via heterologous expression vectors. Of note, these surrogate tools do not allow the control of the actual concentration of the expressed protein due to the following: (i) the well known overexpression resulting from these systems; (ii) the possible variation in expression over time during the cellular cycle; and (iii) the potential multiplicity of subcellular localizations of the expressed protein. This is precisely the case for HCV core protein as mentioned before (25), and it may also account for discrepancies between published data because of the different experimental procedures used. In cell-based systems, unlike acellular assays, there is an important membranous network that affects core protein distribution, depending on cell type and maybe on its actual effect on HCV IRES. Moreover, data reflect the overall steady-state inside the cell, without distinguishing different phases of the cell cycle during which HCV IRES efficiency has been reported to vary (61). In an in vitro assay, the system is more constrained, and the observed phenomenon is thus expected to faithfully represent the spontaneous interaction between HCV IRES and core protein. Nonetheless, we are aware of the necessity of a subsequent in vivo study to confirm the data presented here. This work is in progress, using a mammalian expression system allowing modulatory production of core protein.

In summary, our results collectively support a specific modulatory effect of HCV core protein on the translational activity of viral IRES, although the biological relevance of these data requires further careful examination. Due to conflicting data obtained in this field, it is likely that the HCV IRES-core protein interaction has evolved to regulate the HCV life cycle, gifting the virus with a strategy ensuring its amplification via modulation of the host cell translation apparatus. Further studies are needed to dissect these molecular aspects, and such an RNA-protein interaction might be considered as an excellent target for future therapy (62).


   FOOTNOTES
 
* This work was supported in part by the Association pour la Recherche Contre le Cancer Grant 4301 (to J.-P. L.), the Agence Nationale pour la Recherche sur le Sida, the CNRS, and INSERM (ATC Hépatite C). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

§ Recipient of doctoral grant from the Ministère de l'Education Nationale, de la Recherche et de la Technologie. Back

|| To whom correspondence should be addressed: Laboratoire de Virologie, Centre Européen de Recherche en Virologie et Immunologie, Unité Propre de Recherche et d'Enseignement Supérieur EA 2387, Groupe Hospitalier Pitié-Salpêtrière, 83 Bd. de l'Hôpital, 75651 Paris Cedex 13, France. Tel.: 33-145-82-6298; Fax: 33-145-82-6314; E-mail: cahour{at}idf.ext.jussieu.fr.

1 The abbreviations used are: HCV, hepatitis C virus; IRES, internal ribosome entry site; UTR, untranslated region; nt, nucleotide; EMCV, encephalomyocarditis virus; CSFV, classical swine fever virus; DM, n-dodecyl maltoside; HIV, human immunodeficiency virus, type 1; GBV-B, GB virus B; RRLs, rabbit reticulocyte lysates; NS, nonstructural; C, core protein. Back

2 S. Boulant, C. Vanbelle, C. Ebel, F. Penin, and J. P. Lavergne, submitted for publication. Back

3 T. Shimoike, personal communication. Back


   ACKNOWLEDGMENTS
 
We are grateful to Dr. Peter Sarnow for supplying the plasmid SV CAT BiP Fluc containing the BiP IRES sequence and to Dr. Cezslaw Wychowski for supplying the plasmid pA/C-4 containing the CSFV IRES. We are indebted to Dr. Kunitada Shimotohno for providing us with the plasmid pCMV-980, Dr. John Bukh for the gift of plasmid pGBB, and Dr. Colette Jolivet-Reynaud for the monoclonal antibody 19D96D6. We also thank Sofia Lourenço and Frédéric Bertrand for their help.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hoofnagle, J. H. (1997) Hepatology 26, S15–S20[CrossRef][Medline] [Order article via Infotrieve]
  2. Choo, Q. L., Kuo, G., Weiner, A. J., Overby, L. R., Bradley, D. W., and Houghton, M. (1989) Science 244, 359–362[Abstract/Free Full Text]
  3. Reed, K. E., and Rice, C. M. (2000) Curr. Top. Microbiol. Immunol. 242, 55–84[Medline] [Order article via Infotrieve]
  4. Tsukiyamakohara, K., Iizuka, N., Kohara, M., and Nomoto, A. (1992) J. Virol. 66, 1476–1483[Abstract/Free Full Text]
  5. Rijnbrand, R. C., and Lemon, S. M. (2000) Curr. Top. Microbiol. Immunol. 242, 85–116[Medline] [Order article via Infotrieve]
  6. Reynolds, J. E., Kaminski, A., Kettinen, H. J., Grace, K., Clarke, B. E., Carroll, A. R., Rowlands, D. J., and Jackson, R. J. (1995) EMBO J. 14, 6010–6020[Medline] [Order article via Infotrieve]
  7. Honda, M., Brown, E. A., and Lemon, S. M. (1996) RNA (N. Y.) 2, 955–968
  8. Zhao, W. D., Lahser, F. C., and Wimmer, E. (2000) J. Virol. 74, 6223–6226[Abstract/Free Full Text]
  9. Malet, I., Wychowski, C., Huraux, J. M., Agut, H., and Cahour, A. (1998) Biochem. Biophys. Res. Commun. 253, 257–264[CrossRef][Medline] [Order article via Infotrieve]
  10. Laporte, J., Malet, I., Andrieu, T., Thibault, V., Toulme, J. J., Wychowski, C., Pawlotsky, J. M., Huraux, J. M., Agut, H., and Cahour, A. (2000) J. Virol. 74, 10827–10833[Abstract/Free Full Text]
  11. Fletcher, S. R., Ali, I. K., Kaminski, A., Digard, P., and Jackson, R. J. (2002) RNA (N. Y.) 8, 1558–1571
  12. Pestova, T. V., Shatsky, I. N., Fletcher, S. P., Jackson, R. J., and Hellen, C. U. (1998) Genes Dev. 12, 67–83[Abstract/Free Full Text]
  13. Kieft, J. S., Zhou, K., Jubin, R., and Doudna, J. A. (2001) RNA (N. Y.) 7, 194–206
  14. Ji, H., Fraser, C. S., Yu, Y., Leary, J., and Doudna, J. A. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 16990–16995[Abstract/Free Full Text]
  15. Shimoike, T., Mimori, S., Tani, H., Matsuura, Y., and Miyamura, T. (1999) J. Virol. 73, 9718–9725[Abstract/Free Full Text]
  16. Zhang, J., Yamada, O., Yoshida, H., Iwai, T., and Araki, H. (2002) Virology 293, 141–150[CrossRef][Medline] [Order article via Infotrieve]
  17. Kato, J., Kato, N., Yoshida, H., Ono-Nita, S. K., Shiratori, Y., and Omata, M. (2002) J. Med. Virol. 66, 187–199[CrossRef][Medline] [Order article via Infotrieve]
  18. He, Y., Yan, W., Coito, C., Li, Y., Gale, M., Jr., and Katze, M. G. (2003) J. Gen. Virol. 84, 535–543[Abstract/Free Full Text]
  19. Ray, R. B., Lagging, L. M., Meyer, K., and Ray, R. (1996) J. Virol. 70, 4438–4443[Abstract]
  20. Marusawa, H., Hijikata, M., Chiba, T., and Shimotohno, K. (1999) J. Virol. 73, 4713–4720[Abstract/Free Full Text]
  21. Barba, G., Harper, F., Harada, T., Kohara, M., Goulinet, S., Matsuura, Y., Eder, G., Schaff, Z., Chapman, M. J., Miyamura, T., and Brechot, C. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 1200–1205[Abstract/Free Full Text]
  22. Large, M. K., Kittlesen, D. J., and Hahn, Y. S. (1999) J. Immunol. 162, 931–938[Abstract/Free Full Text]
  23. Tellinghuisen, T. L., and Rice, C. M. (2002) Curr. Opin. Microbiol. 5, 419–427[CrossRef][Medline] [Order article via Infotrieve]
  24. Ray, R. B., Lagging, L. M., Meyer, K., Steele, R., and Ray, R. (1995) Virus Res. 37, 209–220[CrossRef][Medline] [Order article via Infotrieve]
  25. Lai, M. M., and Ware, C. F. (2000) Curr. Top. Microbiol. Immunol. 242, 117–134[Medline] [Order article via Infotrieve]
  26. McLauchlan, J. (2000) J. Viral Hepat. 7, 2–14[CrossRef][Medline] [Order article via Infotrieve]
  27. Ray, R. B., and Ray, R. (2001) FEMS Microbiol. Lett. 202, 149–156[CrossRef][Medline] [Order article via Infotrieve]
  28. Wang, T. H., Rijnbrand, R. C., and Lemon, S. M. (2000) J. Virol. 74, 11347–11358[Abstract/Free Full Text]
  29. Kim, Y. K., Lee, S. H., Kim, C. S., Seol, S. K., and Jang, S. K. (2003) RNA (N. Y.) 9, 599–606
  30. Li, D., Takyar, S. T., Lott, W. B., and Gowans, E. J. (2003) J. Gen. Virol. 84, 815–825[Abstract/Free Full Text]
  31. Laporte, J., Bain, C., Maurel, P., Inchauspe, G., Agut, H., and Cahour, A. (2003) Blood 101, 52–57[Abstract/Free Full Text]
  32. Malet, I., Belnard, M., Agut, H., and Cahour, A. (2003) J. Virol. Methods 109, 161–170[CrossRef][Medline] [Order article via Infotrieve]
  33. Jang, S. K., Krausslich, H. G., Nicklin, M. J., Duke, G. M., Palmenberg, A. C., and Wimmer, E. (1988) J. Virol. 62, 2636–2643[Abstract/Free Full Text]
  34. Ruggli, N., Tratschin, J. D., Mittelholzer, C., and Hofmann, M. A. (1996) J. Virol. 70, 3478–3487[Abstract]
  35. Macejak, D. G., and Sarnow, P. (1991) Nature 353, 90–94[Medline] [Order article via Infotrieve]
  36. Cortay, J. C., Negre, D., Scarabel, M., Ramseier, T. M., Vartak, N. B., Reizer, J., Saier, M. H., Jr., and Cozzone, A. J. (1994) J. Biol. Chem. 269, 14885–14891[Abstract/Free Full Text]
  37. Tanaka, T., Kato, N., Nakagawa, M., Ootsuyama, Y., Cho, M. J., Nakazawa, T., Hijikata, M., Ishimura, Y., and Shimotohno, K. (1992) Virus Res. 23, 39–53[CrossRef][Medline] [Order article via Infotrieve]
  38. Bukh, J., Apgar, C. L., and Yanagi, M. (1999) Virology 262, 470–478[CrossRef][Medline] [Order article via Infotrieve]
  39. Boulant, S., Becchi, M., Penin, F., and Lavergne, J. P. (2003) J. Biol. Chem. 278, 45785–45792[Abstract/Free Full Text]
  40. Reboud, A. M., Dubost, S., Buisson, M., and Reboud, J. P. (1980) J. Biol. Chem. 255, 6954–6961[Abstract/Free Full Text]
  41. Santolini, E., Migliaccio, G., and La Monica, N. (1994) J. Virol. 68, 3631–3641[Abstract/Free Full Text]
  42. Ray, P. S., and Das, S. (2004) Nucleic Acids Res. 32, 1678–1687[Abstract/Free Full Text]
  43. Matsumoto, M., Hwang, S. B., Jeng, K. S., Zhu, N., and Lai, M. M. (1996) Virology 218, 43–51[CrossRef][Medline] [Order article via Infotrieve]
  44. Nolandt, O., Kern, V., Muller, H., Pfaff, E., Theilmann, L., Welker, R., and Krausslich, H. G. (1997) J. Gen. Virol. 78, 1331–1340[Abstract]
  45. Hunt, S. L., Kaminski, A., and Jackson, R. J. (1993) Virology 197, 801–807[CrossRef][Medline] [Order article via Infotrieve]
  46. Bergqvist, A., and Rice, C. M. (2001) J. Virol. 75, 772–781[Abstract/Free Full Text]
  47. Dumas, E., Staedel, C., Colombat, M., Reigadas, S., Chabas, S., Astier-Gin, T., Cahour, A., Litvak, S., and Ventura, M. (2003) Nucleic Acids Res. 31, 1275–1281[Abstract/Free Full Text]
  48. Scheper, G. C., Thomas, A. A., and Voorma, H. O. (1991) Biochim. Biophys. Acta 1089, 220–226[Medline] [Order article via Infotrieve]
  49. Brown, E. A., Zhang, H. C., Ping, L. H., and Lemon, S. M. (1992) Nucleic Acids Res. 20, 5041–5045[Abstract/Free Full Text]
  50. Tanaka, Y., Shimoike, T., Ishii, K., Suzuki, R., Suzuki, T., Ushijima, H., Matsuura, Y., and Miyamura, T. (2000) Virology 270, 229–236[CrossRef][Medline] [Order article via Infotrieve]
  51. Honda, M., Rijnbrand, R., Abell, G., Kim, D., and Lemon, S. M. (1999) J. Virol. 73, 4941–4951[Abstract/Free Full Text]
  52. Honda, M., Beard, M. R., Ping, L. H., and Lemon, S. M. (1999) J. Virol. 73, 1165–1174[Abstract/Free Full Text]
  53. Cristofari, G., Ivanyi-Nagy, R., Gabus, C., Boulant, S., Lavergne, J. P., Penin, F., and Darlix, J. L. (2004) Nucleic Acids Res. 32, 2623–2631[Abstract/Free Full Text]
  54. Fan, Z., Yang, Q. R., Twu, J. S., and Sherker, A. H. (1999) J. Med. Virol. 59, 131–134[CrossRef][Medline] [Order article via Infotrieve]
  55. Kieft, J. S., Zhou, K., Jubin, R., Murray, M. G., Lau, J. Y., and Doudna, J. A. (1999) J. Mol. Biol. 292, 513–529[CrossRef][Medline] [Order article via Infotrieve]
  56. Sansonno, D., Lauletta, G., and Dammacco, F. (2004) J. Viral Hepat. 11, 27–32[Medline] [Order article via Infotrieve]
  57. Chang, M., Williams, O., Mittler, J., Quintanilla, A., Carithers, R. L., Jr., Perkins, J., Corey, L., and Gretch, D. R. (2003) Am. J. Pathol. 163, 433–444[Abstract/Free Full Text]
  58. Hambidge, S. J., and Sarnow, P. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 10272–10276[Abstract/Free Full Text]
  59. Gamarnik, A. V., and Andino, R. (1998) Genes Dev. 12, 2293–2304[Abstract/Free Full Text]
  60. McLauchlan, J., Lemberg, M. K., Hope, G., and Martoglio, B. (2002) EMBO J. 21, 3980–3988[CrossRef][Medline] [Order article via Infotrieve]
  61. Honda, M., Kaneko, S., Matsushita, E., Kobayashi, K., Abell, G. A., and Lemon, S. M. (2000) Gastroenterology 118, 152–162[CrossRef][Medline] [Order article via Infotrieve]
  62. Jubin, R. (2003) Curr. Opin. Investig. Drugs 4, 162–167[Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
L. Jaffrelo, S. Chabas, S. Reigadas, A. Pflieger, C. Wychowski, J. Rumi, M. Ventura, J.-J. Toulme, and C. Staedel
A functional selection of viral genetic elements in cultured cells to identify hepatitis C virus RNA translation inhibitors
Nucleic Acids Res., September 1, 2008; 36(15): e95 - e95.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
E. Steinmann, C. Brohm, S. Kallis, R. Bartenschlager, and T. Pietschmann
Efficient trans-Encapsidation of Hepatitis C Virus RNAs into Infectious Virus-Like Particles
J. Virol., July 15, 2008; 82(14): 7034 - 7046.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
R. Ivanyi-Nagy, J.-P. Lavergne, C. Gabus, D. Ficheux, and J.-L. Darlix
RNA chaperoning and intrinsic disorder in the core proteins of Flaviviridae
Nucleic Acids Res., February 11, 2008; 36(3): 712 - 725.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
L. K. McMullan, A. Grakoui, M. J. Evans, K. Mihalik, M. Puig, A. D. Branch, S. M. Feinstone, and C. M. Rice
Evidence for a functional RNA element in the hepatitis C virus core gene
PNAS, February 20, 2007; 104(8): 2879 - 2884.
[Abstract] [Full Text] [PDF]

Home page
 J. Gen. Virol.Home page
P. R. MacCallum, S. C. Jack, P. A. Egan, B. T. McDermott, R. M. Elliott, and S.-W. Chan
Cap-dependent and hepatitis C virus internal ribosome entry site-mediated translation are modulated by phosphorylation of eIF2{alpha} under oxidative stress.
J. Gen. Virol., November 1, 2006; 87(Pt 11): 3251 - 3262.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/18/17737    most recent
M501826200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boni, S.
Right arrow Articles by Cahour, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boni, S.
Right arrow Articles by Cahour, A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement