Expression and site-specific mutagenesis of the poliovirus 3C protease in Escherichia coli.

Abstract

We have engineered a segment of the poliovirus genome (nucleotides 5438-6061) that encodes the 183 amino acid residues of the 3C region and 25 residues of the 3D region of the viral polyprotein into an Escherichia coli expression vector. The 3C region is a virus-specific protease, which, when expressed in E. coli, is shown to be active and autocatalytic. In our system, three poliovirus-specific proteins are produced: a precursor polyprotein (3C-3D), an internal initiation product, and the mature protease (3C). Mutants in the 3C region have been constructed by oligonucleotidedirected mutagenesis and their effect on the proteolytic activity has been assayed by the in vivo production of the mature protease. The mutation of highly conserved residues (cysteine-47 or histide-161) produced an inactive enzyme, while the mutation of a nonconserved residue (cysteine-153) had a negligible effect on the proteolytic activity.Poliovirus, a small RNA virus, is a member of the picornavirus family, which includes rhinovirus, encephalomyocarditis virus, and foot and mouth disease virus. During the viral life cycle, the genomic RNA is initially translated into a >200-kDa polyprotein, which, while nascent, is cleaved to a series of polypeptide intermediates, termed P1, P2, and P3. Further processing yields the majority of mature viral products (structural and nonstructural) and is dependent on a viral protease(s). The poliovirus P3 intermediate consists of three nonstructural proteins-the genome-linked virion protein (VPg), the protease (3C), and the replicase (3D)-and two proteins of unknown function (3C' and 3D') (1, 2). Hanecak et al. (3) have demonstrated that the 3C region of the poliovirus genome represents the core sequence responsible for the proteolytic processing (at Gln-Gly) of the precursor polypeptides. In addition, the poliovirus 3C protease has been reported to be autocatalytic (4), in agreement with results for encephalomyocarditis virus (5) and foot and mouth disease virus (6).The enzymatic mechanism of these viral proteases has not been well characterized. Inhibitors such as iodoacetamide, N-ethyl maleimide, and para-chloromercuribenzoate have implicated a cysteine residue in the active site (7,8). However, diisopropylphosphofluoridate, an inhibitor of serine proteases, has also been reported as an effective inhibitor of poliovirus protein processing (9). A comparative study (10) of the derived protease proteins from several picornaviruses as well as the related plant virus cowpea mosaic virus (11) revealed one region of significant homology located in the carboxyl third ofthe protein. Within this region a cysteine and a histidine residue have been strictly conserved. These observations suggest that the picornavirus proteases most probably belong to the class of cysteine proteases.To obtain sufficient quantities of the protease proteins for study, we have engineered a cDNA clone of the poliovirus 3C-3D region into an Escherichia coli expression vector, which produced the protease polypeptides in high yield...

Processing and Assembly of Foot-and-Mouth Disease Virus Proteins Using Subgenomic RNA

“…This protease has been characterized as a cysteine protease (Argos et al, 1984) and by sequence comparison of several different picornaviruses two conserved amino acids have been implicated as reactive residues in the active site (Werner et al, 1986). Furthermore these amino acids have been shown to be directly involved in the active site of 3C from PV (Ivanoff et al, 1986). Sequence analysis of the 3C-3D junction (Fig.…”

Cysteine proteases of positive strand RNA viruses and chymotrypsin‐like serine proteases

“…[5]). The functional importance of these two residues has been subsequently confirmed by sitedirected mutagenesis [9], though direct test of the hypothesis has not been reported. Due to the different location of the (putative) catalytic Cys residues and to the lack of overall sequence similarity, it was suggested that 3C-like proteases were not evolutionarily related to other cysteine proteases [5], their formal analogy being explained by convergence.…”

“…Despite the similarity in the positioning of the (putative) catalytic residues in 3C p~° and chymotrypsin-like enzymes, Cys and Ser are not as easily interchangeable in the triad as could be imagined. Thus Cys 147 to Ser substitution in poliovirus 3C Pr° completely abolished its protease activity [9]. In nature, however, such substitutions appear to work as exemplified by the putative protease of SBMV; presumably this is gratified by some compensatory substitution(s).…”

“…Specifically, substitution of Val or Ala for GIy 51 (hereafter in this section the poliovirus numbering is used), presumably disrupting a/~-turn, was lethal, whereas substitution of Asp (a residue frequently occurring in flturns [23]) in the same position resulted in a viable virus [24]. Substitutions in strands E and F which could cause local deformations of the/3-sheet exerted relatively mild effects on viral reproduction [24][25][26], and a substitution of Ser for Cys 153 in strand J appeared to be without effect on the activity of 3C P'° expressed in E. coli [9]. Moreover, the processing defects inflicted by substitutions in strands E and F were similar (namely, impairment of the cleavage at the C-terminus of 3C Pr° itself [24,25]), in accord with our proposal that these strands might interact with each other in native 3C Pt°.…”

Proteolytic processing of tomato ringspot nepovirus 3C-like protease precursors: definition of the domains for the VPg, protease and putative RNA-dependent RNA polymerase.

“…The 3C and 3C-related proteases also contain a substratebinding pocket which determines the cleavage site specificity (Bazan & Fletterick, 1988 ;Dougherty et al, 1989 ;Gorbalenya et al, 1989 ;Nienaber et al, 1993 ;Allaire et al, 1994 ;Matthews et al, 1994). In particular, a histidine residue located in the Cterminal region of the protease is directly involved in the recognition of a glutamine (or glutamate) residue at the k1 position of the cleavage sites of picorna-, poty-and comovirus polyproteins (Bazan & Fletterick, 1988 ;Gorbalenya et al, 1989 ;Allaire et al, 1994 ;Matthews et al, 1994 ;Ivanoff et al, 1986 ;Dougherty et al, 1989 ;. The proteases of nepoviruses of subgroups a and b do not contain a histidine residue in their substrate-binding pockets and recognize cleavage sites that differ from the (Q, E)\(G, S, M) consensus of the aforesaid polyproteins (see Mayo & Robinson, 1996 ;Sanfaçon, 1995 and references therein).…”