No abstract

Infection of HeLa cells with poliovirus leads to rapid shut-off of host cell transcription by RNA polymerase II. Previous results have suggested that both the basal transcription factor TBP (TATA-binding protein) and transcription activator proteins such as CREB (cyclic AMP-responsive element-binding protein) and Oct-1 (the octamer-binding factor) are cleaved by the viral-encoded protease, 3C(Pro). Here we demonstrate that the transcriptional activator (and tumor suppressor) p53 is degraded by the viral protease 3C both in vivo and in vitro. Unlike other transcription factors that are directly cleaved by 3C(pro), degradation of p53 requires a HeLa cell activity in addition to 3C(Pro). The degradation of p53 by 3C(Pro) does not appear to involve the ubiquitin pathway of protein degradation. Vaccinia virus infection of HeLa cells leads to inactivation of the cellular activity required for 3C(Pro)-mediated degradation of p53. The vaccinia-encoded protein (CrmA) is known to inhibit caspase I (ICE protease) that converts inactive IL-1beta to an active secreted form. Incubation of HeLa cells with caspase I inhibitor Z-VAD-fmk does not interfere with 3C(Pro)-mediated degradation of p53. The cellular activity present in extracts of HeLa cells can be fractionated through phosphocellulose. A partially purified fraction that elutes at 0.6 M KCl from phosphocellulose contains the activity that degrades p53 in a 3C(Pro)-dependent manner. These results suggest that both poliovirus-encoded protease 3C(Pro) and a cellular activity are required for the degradation of p53 observed in cells infected with poliovirus.

Poliovirus-encoded nonstructural polypeptide 2C is a multifunctional protein that plays an important role in viral RNA replication. 2C interacts with both intracellular membranes and virus-specific RNAs and has ATPase and GTPase activities. Extensive computer analysis of the 2C sequence revealed that in addition to the known ATPase-, GTPase-, membrane-, and RNA-binding domains it also contains several "serpin" (serine protease inhibitor) motifs. We provide experimental evidence suggesting that 2C is indeed capable of regulating virus-encoded proteases. The purified 2C protein inhibits Poliovirus is the prototype member of the picornavirus family with a plus-sense RNA genome of 7,440 nucleotides, which is covalently linked to a small viral protein (VP g ) at its 5Ј end and polyadenylated at its 3Ј end (34, 56). The positive-strand mRNA, which lacks VP g , codes for a single large polyprotein, which is processed into mature proteins by viral proteases 2A pro , 3C pro and 3CD pro (reviewed in references 39 and 71). Biochemical and genetic evidence suggests that most of the poliovirus nonstructural proteins are involved in viral RNA replication in the form of precursors, mature polypeptides, or both (71). The viral RNA polymerase precursors 3CD pro , the VP g precursor 3AB and a cellular polypeptide, poly(rC)-binding protein, have been shown to interact with the 5Ј-cloverleaf structure of viral RNA, leading to the formation of a functional complex important for viral RNA replication (3,4,31,47,72). The precursor proteins 3AB and 3CD pro also interact with the 3Ј-untranslated region of viral RNA in the absence of other proteins (31). The initiation of poliovirus RNA synthesis appears to be primed by a protein-nucleotidyl covalent complex (VP g -pU or VP g -pUpU) (10, 50, 52).Although it was initially thought to be a member of the cysteine protease family, mutational analyses, amino acid sequence comparison, and three-dimensional modeling have suggested that 3C pro adopts a fold similar to that found in serine proteases such as chymotrypsin (12,28,30,32,33,37,45). The other viral protease, 2Apro , also possesses a chymotrypsin-like fold that is related to smaller serine proteases such as ␣-lytic proteinase (12). Although the major function of these proteases is to process viral precursor polypeptides, both proteases are also involved in the shutoff of host cell metabolism. Although 2A pro is involved in the shutoff of host cell translation (13, 29, 41), 3C pro has been shown to cleave and inactivate a number of host cell transcription factors leading to the inhibition of cellular transcription (19,61,70,75).The 2C protein of poliovirus is 329 amino acids long (37.5 kDa) and is highly conserved among picornaviruses (5). Genetic analyses have implicated the 2C polypeptide in a number of functions during viral replication such as uncoating (40), host cell membrane rearrangement (18), RNA replication (reviewed in reference 71), and encapsidation (69). The exact role of 2C in these processes, however, is not known. Many ...

Soon after infection, poliovirus (PV) shuts off host-cell transcription, which is catalysed by all three cellular RNA polymerases. rRNA constitutes more than 50 % of all cellular RNA and is transcribed from rDNA by RNA polymerase I (pol I). Here, evidence has been provided suggesting that both pol I transcription factors, SL-1 (selectivity factor) and UBF (upstream binding factor), are modified and inactivated in PV-infected cells. The viral protease 3C pro appeared to cleave the TATA-binding protein-associated factor 110 (TAF 110 ), a subunit of the SL-1 complex, into four fragments in vitro. In vitro protease-cleavage assays using various mutants of TAF 110 and purified 3C pro indicated that the Q 265 G 266 and Q 805 G 806 sites were cleaved by 3C pro . Both SL-1 and UBF were depleted in PV-infected cells and their disappearance correlated with pol I transcription inhibition. rRNA synthesis from a template containing a human pol I promoter demonstrated that both SL-1 and UBF were necessary to restore pol I transcription fully in PV-infected cell extracts. These results suggested that both SL-1 and UBF are transcriptionally inactivated in PV-infected HeLa cells.

Previous studies have implicated disulfide bonds between Vp1 molecules in the stabilization of the simian virus 40 (SV40) capsid. To identify the cysteine residues involved in intermolecular disulfide interactions, systematic oligo-directed mutagenesis of cysteine codons to serine codons was initiated. Wild-type and mutant Vp1 proteins were produced in rabbit reticulocyte lysates and were allowed to interact post-translationally. Disulfide-linked Vp1 complexes were assessed via nonreducing SDS-PAGE and via sucrose-gradient sedimentation. Wild-type Vp1 forms 7S pentamers followed by 12S disulfide-linked multi-pentameric complexes in cell-free lysates. Mutagenesis of all seven cysteine codons abolished Vp1 12S complexes, but did not affect pentamer formation. A quadruple Vp1 mutant at Cys 49 , Cys 87 , Cys 254 and Cys 267 continued to form 12S complexes, whereas the major products of the Cys 9 , Cys 104 and Cys 207 triple mutant Vp1 were 7S pentamers. Single and double mutant Vp1 proteins at the three cysteines affected continued to form 12S complexes, but to a lesser extent. Thus, inter-pentamer disulfide bonds at Cys 9 , Cys 104 and Cys 207 are essential and sufficient for stabilization of Vp1 complexes in cell-free lysates. These results are in agreement with previous structural studies of SV40 that implicated the same three residues in disulfide linkage in the capsid. Possible parameters for the involvement of the three cysteines are discussed.