Processing of the encephalomyocarditis virus capsid precursor protein studied in rabbit reticulocyte lysates incubated with N-formyl-[35S]methionine-tRNAfMet

Complete translation of the hepatitis C virus genome in vitro: membranes play a critical role in the maturation of all virus proteins except for NS3

We developed an in vitro translation extract from Krebs-2 cells that translates the entire open reading frame of the hepatitis C virus (HCV) strain H77 and properly processes the viral protein precursors when supplemented with canine microsomal membranes (CMMs). Translation of the C-terminal portion of the viral polyprotein in this system is documented by the synthesis of NS5B. Evidence for posttranslational modification of the viral proteins, the N-terminal glycosylation of E1 and the E2 precursor (E2-p7), and phosphorylation of NS5A is presented. With the exception of NS3, efficient generation of all virus-specific proteins is CMM dependent. A time course of the appearance of HCV products indicates that the viral polyprotein is cleaved cotranslationally. A competitive inhibitor of the NS3 protease inhibited accumulation of NS3, NS4B, NS5A, and NS5B, but not that of NS2 or structural proteins. CMMs also stabilized HCV mRNA during translation. Finally, the formyl-[35S]methionyl moiety of the initiator tRNA(Met) was incorporated exclusively into the core protein portion of the polyprotein, demonstrating that translation initiation in this system occurs with high fidelity.

Mengo virus 3C proteinase: recombinant expression, intergenus substrate cleavage and localization in vivo

Mengo virus 3C proteinase was cloned and expressed to high levels in a bacterial vector system. The protein was solubilized from inclusion bodies then purified to homogeneity (> 95%) by ion exchange chromatography. The recombinant enzyme was proteolytically active in cell-free processing assays with a Mengo capsid precursor substrate, L-P1-2A, correctly and proficiently cleaving it into L, 1AB, 1C, 1D and 2A protein products. Further analyses with synthetic peptide substrates encompassing the Mengo or rhinovirus-14 2C/3A cleavage sequences, showed the Mengo 3C could recognize and process specific glutamine-glycine sites within these peptides. The reactivity with the rhinovirus peptide was unexpected, because cross-reactivity between a picornavirus 3C enzyme and a protein substrate from different genus of this family has otherwise never been observed. In reciprocal reactions, a rhinovirus-14 3C preparation was unable to cleave the Mengo-derived synthetic peptide substrate. The recombinant Mengo 3C reactions were also characterized with regard to substrate Km, optimum pH and temperature. The protein was additionally used to raise monoclonal antibodies (mAbs) in mice, which in turn localized natural 3C, 3ABC, 3CD and P3 in immunoblots, immunoprecipitations and indirect immunofluorescence assays of Mengo-infected HeLa cells. The monoclonals showed cross-reactivity with 3C and 3C-containing precursors from encephalomyocarditis virus (EMCV), but did not react with 3C proteins from rhinovirus-14 or poliovirus-1M.

Polyprotein processing of Theiler's murine encephalomyelitis virus

To investigate polyprotein processing of Theiler's murine encephalomyelitis viruses, we analyzed in vitro translation reactions programmed by in vitro-derived transcripts from an infectious full-length cDNA clone of the DA strain of Theiler's virus. To help identify the proteinases that carried out the processing, we modified the DA cDNA clone transcription template by linearization with different restriction endonucleases that generate templates of different lengths or by constructing linker insertion or deletion mutations or both in putative proteinase-coding regions. Protein 3C carried out most of the cleavages of the polyprotein, as is true for the other picornaviruses that have been studied. A second proteinase also appeared active at the LP12A-2B junction. A protein of slightly faster mobility than the leader protein was seen with translation of transcripts derived from DA cDNA but not GDVII cDNA. This protein may be synthesized from an alternative initiation site in the DA leader-coding region out of phase with the polyprotein reading frame. Our findings are relevant to ongoing investigations of the abnormal virus expression seen in DA virus late demyelinating disease, since polyprotein processing is critical in regulating picornaviral gene expression.

Encephalomyocarditis virus 3C protease: efficient cell-free expression from clones which link viral 5' noncoding sequences to the P3 region

All picornaviral peptides are derived by progressive posttranslational cleavage of a giant precursor polyprotein. Translation of encephalomyocarditis virus (EMC) RNA in rabbit reticulocyte extracts produces active viral peptides, including protease 3C, which is responsible for many cleavage reactions within the processing cascade. DNA plasmids containing 5' noncoding sequences of EMC linked to other portions of the viral genome were constructed and transcribed into RNA. Like virion RNA, the clone-derived transcripts directed efficient protein translation in vitro. The 5'-linked constructions may represent examples of a general method for cell-free expression of any cloned gene segment. One construction produced a self-cleaving P3 region precursor, which contained active 3C protease. A genetically engineered insertion within the 3C sequences eliminated endogenous self-cleavage activity without altering the ability of the P3 peptide to serve as substrate in bimolecular reactions with added 3C. Another plasmid encoding the L-VP0 portion of the capsid region was used to demonstrate that scission between the leader peptide (L) and capsid protein VP0 can be catalyzed by 3C. The enzyme responsible for this step was previously unidentified. A rapid purification scheme for isolation of 3C from EMC-infected HeLa cells is also presented.

A detailed kinetic analysis of the in vitro synthesis and processing of encephalomyocarditis virus products

Translation of encephalomyocarditis virus RNA in rabbit reticulocyte lysates has been used to analyse the pathway of proteolytic processing of the primary translation products. A minimum of two distinct proteases is required to account for the results: one for the excision of the capsid precursor protein, A1, from the nascent polyprotein, and the other for all other cleavages including cleavage at the F/C junction. The excision of A1 is an extremely rapid reaction which occurs as soon as the cleavage site has been synthesised and is resistant to all the proteolytic inhibitors tested and to high temperature, characteristics which are more consistent with an intramolecular cleavage catalysed by a virus-coded protease than cleavage by an endogenous reticulocyte protease. Once excised, A1 remains stable until translation has reached the middle of the region of the genome coding for C, at which time a number of events occur in rapid succession: F is excised in its mature form probably via an intramolecular cleavage; a proteolytic activity capable of secondary processing of A1 to A, B, D1, alpha, gamma and epsilon appears; and a polypeptide of molecular weight about 32,000 appears. This protein (p32) originates from the N-terminal portion of C, and maps in the same position on the genome as p22, the protein previously identified as the virus-coded protease. Polypeptide p32 is derived from C by a single step cleavage generating E as the other product, a processing pathway at least as important, if not more important than the step-wise route via D as an intermediate. Since p32 first appeared at the same time as the start of secondary processing, whilst p22 was first detected much later, it is argued that at least the early stages of processing of the capsid precursor must have been carried out by p32 rather than p22.

Modulation of the expression of poliovirus proteins in reticulocyte lysates

The translation of poliovirus RNA into specific viral proteins in mRNA-dependent reticulocyte lysates (MDLs) was found to be highly dependent on individual lysate preparations. Under optimal conditions, the first polypeptide detected was always P3-1b (formerly NCVP 1b), the product of the 3' portion of the poliovirus genome; the formation of P1-1a (formerly NCVP 1a) followed as shown by time-course and pulse-chase experiments. However, some lysates synthesized little or no P1-1a despite their ability to synthesize P3-1b and to translate normally other cellular and viral mRNAs. When an MDL competent in synthesizing P1-1a was diluted ca. twofold, while maintaining optimal concentrations of salts, tRNA, DTT, creatine phosphate, and amino acids, P1-1a formation was virtually eliminated, while the synthesis of P3-1b, presumably as a consequence of a more downstream initiation, was maintained. The synthesis of P1-1a in a diluted MDL was restored, and P3-1b synthesis suppressed, by the addition of a S10 fraction prepared from uninfected or virus-infected HeLa cells. Nuclease treatment and dialysis of the S10 fraction did not inhibit its activity. These findings indicate that individual MDLs either possess limiting quantities of, or occasionally are deficient in, a factor(s) that promotes the utilization of the presumed 5' proximal initiation site (the AUG at nucleotide position 781-783) and that a homologous factor(s) exists in HeLa cells. The implication of these findings for the strategy of poliovirus replication is discussed.

Multiple proteases in foot-and-mouth disease virus replication

Translation of foot-and-mouth disease virus RNA in a rabbit reticulocyte lysate for short time intervals resulted in the production of the peptides P20a , P16, and P88 (Lab, Lb, and P1) (R. R. Rueckert , Recommendations of the 3rd European Study Group on Molecular Biology of Picornavirus, Urbino , Italy, 1983). If further translation was prevented, the structural protein precursor P88 was not cleaved, even after prolonged incubation. This result indicates that the mechanism of the cleavage between P20a -P16 and P88 and of that between P88 and P52 (P2) differs from the mechanism of the secondary cleavages which produce the structural proteins. Furthermore, treatment of foot-and-mouth disease virus-infected cells with the protease inhibitor D-valyl phenylalanyl lysyl chloromethyl ketone prevented the in vivo cleavage between P20a -P16 and P88 but had no effect on any of the other cleavage events. These results suggest that the cleavage of the foot-and-mouth disease virus polyprotein utilizes two different host proteases.

Systematic nomenclature of picornavirus proteins

An easily learned convention for systematizing the nomenclature of picornavirus proteins is described. The convention is based upon an idealized map, called the L434 diagram, of the picornavirus polyprotein.

Partial N-terminal amino acid sequences of polypeptides p14 and p12 of encephalomyocarditis virus are identical and correspond to the N-terminus of the viral polyprotein

Our previous data suggested that translation in an EMC virus RNA-programmed cell-free system from Krebs-2 cells is initiated predominantly at a single site and that the earliest amino acid sequences synthesized correspond to non-structural 'leader' polypeptides p14 and p12 [(1982) FEBS Lett. 141, 153-156]. Here, polypeptides p14 and p12 were labelled in vitro by tritiated amino acids, isolated and subjected to automated Edman degradation. Both polypeptides (after the loss of the N-terminal methionine) were shown to contain alanine in position 1 and glutamic acid in positions 5 and 7. These and other data demonstrate that p14 and p12 share a common N-terminal sequence. This sequence coincides precisely with the N-terminus of EMC virus polyprotein sequence deduced from the primary structure of the viral genome [(1984) Nucleic Acids Res., in press]. Thus, the single initiation site operating in our translation system corresponds to the start of the polyprotein molecule.

Biochemical map of polypeptides specified by foot-and-mouth disease virus

Pulse-chase labeling of foot-and-mouth disease virus-infected bovine kidney cells revealed stable and unstable viral-specific polypeptides. To identify precursor-product relationships among these polypeptides, antisera against a number of structural and nonstructural viral-specific polypeptides were used. Cell-free translations programmed with foot-and-mouth disease virion RNA or foot-and-mouth disease virus-infected bovine kidney cell lysates, which were shown to contain almost identical polypeptides, were immunoprecipitated with the various antisera. To further establish identity, some proteins were compared by partial protease digestion. Evidence for a membrane association of the polypeptides coded for by the middle genome region is also presented. A biochemical map of the foot-and-mouth disease virus genome was established from the above information.

Orientation of the Cleavage Map of the 200-Kilodalton Polypeptide Encoded by the Bottom-Component RNA of Cowpea Mosaic Virus

The genomic organization of the bottom-component RNA of cowpea mosaic virus was studied. In vivo, this RNA encodes at least eight different polypeptides of 170, 110, 87, 84, 60, 58, 32, and 4 kilodaltons (K), the last polypeptide representing the genome-bound protein VPg. In rabbit reticulocyte lysates, bottom-component RNA is translated into a 200K polypeptide which is then processed to give the 32 and 170K polypeptides also found in vivo. By pulse-labeling the 200K primary translation product, we now show that the 32 and 170K polypeptides are derived from the NH 2 -terminal and COOH-terminal parts of this polypeptide, respectively. Comparison of the proteolytic peptide patterns of 170K polypeptides synthesized in vitro and pulse-labeled at either the NH 2 -terminal or the COOH-terminal end with the patterns of the 170 and 110K polypeptides found in vivo demonstrates that the order within the 200K primary translation product of cowpea mosaic virus bottom-component RNA is as follows: NH 2 -32K polypeptide-58K polypeptide-VPg-24K polypeptide-87K polypeptide-COOH.