key: cord-0004896-r7ofnxf1
authors: Rozanov, M. N.; Drugeon, G.; Haenni, A. -L.
title: Papain-like proteinase of turnip yellow mosaic virus: a prototype of a new viral proteinase group
date: 1995
journal: Arch Virol
DOI: 10.1007/bf01309862
sha: e234ddbac9446c5e8473f1d89cfadcee42a9de9e
doc_id: 4896
cord_uid: r7ofnxf1

Sequence comparisons predicted a potential papain-like proteinase domain in the N-terminal cleavage product (NRP) of the large nonstructural replicase polyprotein (RP) of turnip yellow mosaic virus (TYMV). Replacement of the predicted catalytic amino acids, Cys-783 by Ser, or of His-869 by Glu, abolished cleavage of the 206K RP into a ∼150K NRP and a ∼78K C-terminal product in reticulocyte lysates, while other substitutions exerted no apparent influence on proteolysis. The proteinase-deficient mutant RPs could not be cleaved in trans by as much as an eight-fold molar excess of wild-type proteinase. Deletion experiments have excluded the possible influence on autoproteolysis of amino acid sequences 1–708 and 982–1204 flanking the proteinase domain. Thus, the proteinase of TYMV with a papain-like dyad of essential amino acids has been mapped just upstream from the putative NTPase domain. Statistically significant sequence similarities with the TYMV proteinase were found for the similarly located domains of the replicase polyproteins of carlaviruses, capilloviruses, apple stem pitting virus and apple chlorotic leaf spot virus as well as for those of other tymoviruses and for the domain located downstream from the putative NTPase domain of the large polyprotein of beet necrotic yellow vein furovirus. All these domains are not significantly similar to other known proteinases, although they conserve papain-like Cys- and His-containing motifs. Thus these domains constitute a compact group of related enzymes, the tymo-like proteinases, within the proposedpapainlike proteinase supergroup. The resulting alignment of 10 tymo-like proteinase sequences has revealed a third highly conserved residue — Gly (Gly821 in TYMV RP) followed by a hydrophobic residue. We speculate that all the tymo-like proteinase domains of the viral replicative proteins may share common biochemical and biological features.

Genomes of all animal positive strand RNA viruses encode proteinases, most of which have been classified as either chymotrypsin-like El, 2] or papain-like [3, 4] enzymes. The plant "picorna-like viruses" [5] , e.g. comoviruses, nepoviruses and potyviruses share this feature with their animal counterparts [6] . Chymotrypsin-like proteinase domains were also predicted for sobemo-like viruses El, 6]. However, it was not known whether plant viruses belonging to another large virus supergroup, the Sindbis-like viruses [6] , contain proteinase genes. The only data suggesting the existence of such genes for viruses of this latter group were obtained for turnip yellow mosaic virus (TYMV).

TYMV is the type member of the tymovirus group. The spherical virion of TYMV contains a single molecule of positive RNA, 6318 nucleotides long, with an mTG cap at its 5' end and a tRNA-like structure at its 3' end [7] . TYMV RNA encodes three polypeptides. The larger nonstructural protein of 206 K, involved in RNA replication [8] , designated here RP (replicase protein), and the smaller cell-to-cell movement protein of 69 K [9] , designated OP (overlapping protein) are translated from overlapping reading frames on the genomic RNA [7, 10] . The 3' adjacent coat protein (CP) gene is expressed via a subgenomic RNA.

Based on sequence comparisons, 3 putative functional domains, a methyltransferase Ell], an NTPase/helicase and a polymerase [7] , have been highlighted in RP (Fig. 1) . These data unequivocally linked tymoviruses with the Sindbis-like supergroup of viruses [5] . In addition, it has come as a surprise that the large replicative proteins of spherical tymoviruses closely resemble those of filamentous potexviruses [12] , another distinct group of plant Sindbislike viruses.

The large nonstructural RP of TYMV undergoes autocleavage [ 13] yielding at least two products of ~150K (NRP, or the N-terminal part of RP) and 78 K (CRP, or the C-terminal part of RP). Similar autocleavages were recently observed for 5 different tymoviruses [14-1. The probable proteinase recognition/ cleavage site has been mapped between amino acid residues 1253-1261 of the TYMV polyprotein (Fig. 1) . The responsible proteinase may be located somewhere in the 688-amino acid stretch upstream from this site [15] . However, attempts to locate the proteinase domain by similarity with known cellular and viral proteinases were unsuccessful ( [6, 7, 13, 15] ; E.V. Koonin, pers. comm.) . This led to the speculation that there are significant differences in the expression strategies of plant and animal viruses E16].

In this paper we describe the prediction of a papain-like proteinase domain in the TYMV RP polyprotein and of its potential catalytically active amino acids, and confirm this prediction by mutation and deletion analyses. From these data and by sequence comparison we have revealed potential tymo-like Numbers indicate amino acid positions in the sequence of the TYMV RP. Thick solid lines show the deletions (dNN* or dS*A) made to map the proteinase domain; xxx, the sequences previously shown to be dispensable for autoproteolysis [15] . The effect of each mutation or deletion on autocleavage of the RP is indicated by (-), no cleavage, or (+), apparently unaltered cleavage proteinase domains in the large nonstructural proteins of carlaviruses, capilloviruses, furoviruses, apple stem pitting virus and apple chlorotic leaf spot virus. These domains constitute a compact group of related enzymes, the tymo-like proteinases within the proposed papain-like proteinase supergroup.

The GENEBEE package [17] was used for sequence database searches and to look for local similarities between two sequences. Alignments were made by the MULTALIN [18] and OPTAL [19] programs, and then analyzed by visual inspection. The significance of alignments was assessed by the OPTAL program as the number of standard deviation (SD) above the mean score for 25 random permutations of the same sequences. Alignments scoring 6 SD or more were taken into consideration. The sequences of the following viruses, found to possess (putative) proteinase domains, were included in the final alignment: the tymoviruses TYMV [7] , kennedya yellow mosaic virus, KYMV [20] , ononis yellow mosaic virus, OYMV [21] , eggplant mosaic virus, EMV [22] and erysimum latent virus, ELV [23] ; potato carlavirus M, PVM, [24] ; apple stem pitting virus, ASPV [25] ; apple chlorotic leaf spot virus, ACLSV [26] ; apple stem grooving capillovirus, ASGV [27] ; and beet necrotic yellow vein furovirus, BNYVV [28] .

In all constructs, designations of plasmids and transcripts are preceded b37 the letter 'p' and 't' respectively. Most DNA manipulations were performed according to standard protocols [29] . Minipreps for the final transcription/translation experiments were prepared using a plasmid mini kit (Qiagen) as indicated by the supplier. Plasmids were screened by restriction analysis. A plasmid construct derived from pEMBL19 and containing a fulllength copy of the TYMV cDNA, pTYFL7, was obtained as described [30] . A single-stranded DNA matrix for mutagenesis, MX-wt, was obtained by cloning an MroI1846-XhoI3282 blunt-ended fragment of pTYFL7 (1436 base pairs) into the EcoRI-SmaI digested and blunt-ended M13mpl8 vector. Mutants were generated according to the USB protocol using the following synthetic oligodeoxynucleotides containing mismatches (bold type letters), and artificially introduced restriction sites (underlined) and codon changes: 5'GGAGAGAAGAGAATTCAAGGY (EcoRI) for the Cys783 to Ser mutation, MX-C783S; 5'GCCGGGTGAGAATTCGGATGGCGGTCCY (EcoRI) for the His869 to Glu mutation, MX-H869E; 5'CTTCATGTTTAAAATCAAGTTC3' (DraI) for the Ser925 to Leu mutation, MX-S925L; 5'CGGATATGTCGACCCACAGCCAGC3' (SatI) for the Lys982 to Ser mutation, MX-K982S; 5'GGAGCCCATGGGGGGGTCCGC3' to introduce an additional NcoI site, designated NcoI*. Mutations from recombinant phages were transferred to pTYFL7 by replacement of the wild-type MroI-XhoI fragment by the mutant counterparts yielding the full-length TYMV cDNA mutants, designated pC783S, pH869E, pS925L, pK982S, and pNcoI*, respectively. Selected plasmid constructs were partially sequenced using the T7 DNA polymerase sequencing kit (Pharmacia).

To obtain the polyprotein containing a deletion upstream of the proteinase domain, the full-length mutant pNcoI* was digested with NcoI. After heating to destroy the enzymatic activity, the sample was divided into two parts. The first was treated with the Klenow fragment ofDNA polymerase I to fill in sticky ends, and the other was left untreated. After electrophoretic separation in an agarose gel and purification, both the sticky (s) and blunt larger DNA fragments (containing the vector and most of the TYMV cDNA sequence) were religated and transformed into competent Escherichia coli cells. The resulting plasmids (pdNN*s and pdNN*) contained open reading frames for the predicted proteinase domain downstream from the 5'-proximal 41 codons of the OP gene, or from the 5'-proximal 39 codons of the RP gene, respectively.

Polyprotein with a deletion between the proposed proteinase domain and the putative recognition/cleavage site was obtained from pK982S bearing the artificially-introduced SatI* site. Its TYMV RNA-specific SaII*3047-EcoRI region of the plasmid was replaced by the corresponding ApaI3708-EcoRI fragment using a synthetic SaII-ApaI linker (5'TCGAGTTCTAGACTGGCCAGGGGCC3', (+), upper strand, and 5'CCTGGCCAG-TCTAGAACY, (-), lower strand).

The deleted and mutated plasmids were linearized by EcoRI (located in the multiple cloning site) or by SmaI (position 6061 in the TYMV cDNA) respectively. Transcription with T7 RNA polymerase (BRL) was performed essentially as recommended by the supplier, using 1 mM each of ATP, CTP and UTP, 0.5 mM GTP and 2 mM mTGpppG. After purification on a G-50 Sephadex column, phenol extraction and ethanol precipitation, integrity and size of the resulting transcripts were estimated by native gel electrophoresis, and concentration by A26 o measurements.

The standard translation conditions (total volume 10 gl) using a rabbit reticulocyte lysate, and 370kBq of L-[35S]cysteine (~48TBq/mmol; Amersham) or of L-[3SS]methionine (~40 TBq/mmol; ICN) were as described [13] . Incubation conditions and final RNA concentrations were as indicated in Fig. 3 . The translation products (generally 8 gl) were analyzed on 12.5% acrylamide-0.1% bis-acrylamide-SDS gels [31] . The gels were then fixed and fiuorographed.

To develop a hypothesis about the possible location of the TYMV proteinase domain, we analyzed sequences of large proteins of tymo-and potexviruses ( Fig. 1 ). These proteins proved to be closely related in the regions of the putative methyttransferase, NTPase and polymerase domains [11, 12] . However, it seemed conceivable that the proteinase domain would be conserved among the tymovirus sequences but not among the potexvirus sequences, because potexviruses did not seem to utilize proteolysis as a strategy for the expression of their proteins (A. Karasev, pets. comm.). After the entire sequences of the large proteins of tymo-and potexviruses were aligned (not shown), stretches of conserved tymovirus sequences that had no counterparts in potexvirus sequences were detected. Such tymovirus-specific portions of the alignment were primary targets of our search.

The tymoviral sequences contained neither chymotrypsin-like nor aspartic proteinase motifs known to occur in polyproteins of many viruses. However, it was also known that the genomes of animal Sindbis-like viruses as well as of several plant and animal viruses belonging to other supergroups encode papainlike proteinases [4] . The updated list of such viruses includes alphaviruses [3, 32] , rubiviruses [33] , corona-and artheriviruses [34, 35] , potyviruses [36] , hypovirulence-associated virus (HyAV) of the chestnut blight fungus [37, 28] ( Fig. 2A) , probably aphthoviruses [4, 39] , bymoviruses [40, 41] , hepatitis E virus [16] , and possibly, pestiviruses [42] . The proteinases of these viruses have little or no sequence similarity with each other or with cellular papain-like proteinases except for two regions, containing (putative) catalytically active Cys and His residues ( Fig. 2A ). Our analysis of the tymovirus sequences revealed such a conserved amino acid dyad upstream of the putative NTPase domain. In the TYMV sequence, these amino acids, Cys783 and His869, were located within the 688-amino acid region which is reported to be essential for autocleavage [15] . The conserved Cys is followed by Leu (or Met in the case of ELV); this is in contrast to other papain-like proteinases, which have an aromatic amino acid at this position ( Figs. 2A, B) , with one exception, the Staphylococcal proteinase. Similar to most other papain-like proteinases, both the Cys and His residues are preceded by regions rich in structure breakers, such as Pro, Gly, Asn, Set, although the conserved amino acid pattern is unique for the tymovirus sequences (Figs 2A, B) Using the level of similarity between tymovirus sequences as the main criterion, and by analogy with known papain-like proteinases, the N-terminal border of the TYMV proteinase domain was predicted to be not further than 49 amino acid residues upstream of the essential Cys783. The C-terminal border of the proteinase could be located in a fairly short variable stretch of the sequence, between His869 and the putative NTPase/helicase domain. As the borders of the latter were not determined, "motif" A [19] , GFAGCGKT, where K is the invariant Lys982, had to be considered its N-terminal border.

Thus, an unusual papain-like proteinase with catalytic Cys783 and His869 was predicted in tymoviral replicase polyproteins close to, and upstream of, the putative NTPase domain.

This hypothesis was tested by mutational and deletion analyses of the predicted proteinase domain ( Fig. 1; Fig. 3A , B). Site-directed point mutations were introduced into a TYMV cDNA clone, pTYFL7. Their influence on autoproteolysis of RP was then investigated by in vitro translation of corresponding mutant transcripts using both standard (Fig. 3A) and temperature shift conditions [13] (data not shown) for each of them. In full agreement with our predictions, replacement of Cys783 by Ser (C783S) or of His869 by Glu (H869E) abolished RP to NRP and CRP hydrolysis (Fig. 3A, lanes 3, 4) . Other substitutions such as replacement of the invariant Ser925 for Leu ($925L; lane To localize the proteinase domain in a short stretch of sequence containing the essential Cys and His, two deletion variants of pTYFL7 were constructed, pdNN* and pdS*A. To delete the 5-proximal half of the RP gene, a single point mutation T2217 to C was introduced into the TYMV cDNA upstream of, and close to, the predicted 5'-border of the proteinase gene so as to create a second NcoI site referred to as NcoI*. The transcripts obtained from the NcoI-NcoI* deletion variant of the TYMV cDNA (Fig. 1) gave rise upon translation in reticulocyte lysates to a truncated polypeptide (dNN*) of 131 K, which underwent autoproteolysis yielding the NRP product and the unaltered CRP product (Fig. 3B, lanes 7-9) under conditions identical to those observed with a full-length transcript (wt; Fig. 3A , lane 2) or with TYMV RNA (Fig. 3B,  lanes 1-3) . Thus, the deleted amino acid sequences are not required for autoproteotysis of the polyprotein in vitro. In addition, the fact that the two transcripts tdNN* and tdNN*s, which differ in the first 39 codons upstream of the NcoI site, gave identical results (not shown), eliminated the possibility that the N-terminus of RP participates in autoproteolysis. Hence, in agreement with our prediction, the N-terminal border of the proteinase is downstream from amino acid 709 in the polyprotein.

Deletion of amino acids 982-t204 removed most of the NTPase domain (dS*A; Fig. 1 ). Again, this deletion did not abolish autoproteolysis of the expected 182 K RP precursor to yield the NRP product and the unaltered CRP product (Fig. 3B, lanes 4-6) . The possibility that the proteinase domain is located downstream from its cleavage/recognition site was excluded [15] (our unpubl, data). We conclude that the amino acid stretch located between residue 709 and the NTPase domain, displaying some similarity to the papain-like proteinases and bearing the Cys and His residues essential for proteinase activity, indeed harbors the TYMV proteinase.

To investigate whether the proteinase can act in trans, tC783S was translated in vitro in the presence of transcripts encoding the wild-type proteinase domain, either tDNN* (not shown) or tTYFL1 (Fig. 3C) . No cleavage of the undeleted tC783S mutant polyprotein could be detected even with an 8-or 5-fold molar excess of tTYFLI or tDNN*, respectively, over tC783S (not shown).

The TYMV proteinase is the prototype of a new distinct group of (putative) viral proteinases Figure 2B shows that, as mentioned before, the TYMV-like proteinase domain is strictly conserved among all tymoviruses. To determine whether other Sindbis-like plant viruses have (putative) TYMV-like proteinase domains, we examined their polypeptides for the presence of the tymo-like proteinase sequence motifs (in the case of BNYVV it was kindly done by E. V. Koonin and V. V. Dolja, pers. comm. and [43] ). Stretches of sequences which contained such motifs were analyzed for overall similarity to the tymovirus proteinase. Figures 2B, C show the results of our analysis. Of the five sequences that were significantly similar to the TYMV proteinase, four (PVM, ASPV, ASGV, and ACLSV) were aligned with each other (Fig. 2C ) and then compared with the alignment of the 5 tymoviral proteinase sequences. The resulting alignment score of 6.5 SD supports our hypothesis on the relationship of these sequences of the carlavirus PVM, and the capillovirus ASGV as well as of ASPV and ACLSV, to those of tymoviral proteinases. The best score for the putative BNYVV proteinase (6.7SD) was obtained when its sequence was compared with the alignment of all the TYMV-like proteinase sequences. Such a score, although the worst one of the respective scores obtained for any of 9 other sequences, reflects a probable relationship.

Consensus sequence 1 derived for the tymoviral proteinases (Fig. 2B ) matches well the consensus of the alignment of the other 5 sequences (Consensus 2, Fig. 2C ). The resulting consensus (Fig. 2C) consists of the two motifs containing the essential Cys and His residues, respectively, and one more motif located between them, G*(s/x), where G is Gly, * is a hydrophobic residue, and s is Ser, which can be replaced by other residues (x). The conserved Gly is not present in the BNYVV sequence. Sequence database searches with the conserved motifs did not reveal additional relatives to the tymo-like proteinases.

In the sequences of polyproteins of plant Sindbis-like viruses we also observed some additional conserved short stretches that distantly resembled papain-like proteinases, but the similarities were below a statistical significance (not shown). However, taking into account the fact that the viral proteinases are extremely variable in their primary structure [4] , one can not exclude the possibility that some of the sequences mentioned belong to unusual proteinase domains only distantly resembling the papain-like domains. Simple deletion/mutagenesis experiments rather than more extensive computer analyses may help to clarify these uncertainties.

The deletion/mutagenesis experiments (Figs. 3A, B ) support our prediction of the papain-like thiol proteinase of TYMV with a putative catalytic dyad Cys-783 and His-869 ( Fig. 1; Figs. 2A, B) . Taking into account that, as mentioned above, there are no other obvious candidates for the role of catalytically active amino acids in the central region of the tymovirus polyproteins, it is likely that Cys783 and His869 play this role in the TYMV proteinase. Inhibition studies of TYMV RP autocleavage also favor our conclusions. Inhibitors of thiol proteinases such as ZnC12 abolish autoproteolysis of RP even at low concentrations, while inhibitors of serine proteinases affect proteolysis only at high concentration [13] . Failure of cystatin and E64 to influence hydrolysis of the RP [13] does not contradict our hypothesis, because not all genuine papain-like proteinases are sensitive to these inhibitors [44] . The question of whether the TYMV proteinase resembles papain in its 3D structure or at least in the configuration of its active site will be resolved only when X-ray data will become available. Figure 3C demonstrates that under our conditions the proteinase does not cleave the mutant polyprotein in trans, but acts in cis in agreement with previously reported data [ 13, 15] . Indeed, with the exception of the proteinases of alphaviruses [31] , most viral papain-like proteinases apparently function only in cis.

An additional indirect indication of the similarity of the tymovirus proteinase to other viral papain-like domains is the putative recognition/cleavage site. This site has been tentatively located by site-directed mutagenesis within a 9 amino acid stretch, between amino acids 1253 and 1261 [15] , at the border of the NTPase and polymerase domains. Alignment of five tymovirus sequences shows that this region of the tymoviral polyprotein is not highly conserved. But it does contain only small amino acid residues, Gly, Ala or Ser, in positions which correspond to amino acids 1258 and 1259 of the RP polyprotein of TYMV (M.N.R. unpubl.). Notably, a Gly residue is present in one of the corresponding positions in each sequence. Cleavage of the tymovirus polyproteins may take place at this site by analogy with those of other viral papain-like proteinases known to hydrolyze the peptide bond between two small amino acid residues (alphaviruses -nsP1/nsP2 and nsP2/nsP3 sites [45] , potyviruses [46] , HyAV [37, 38] , artheriviruses [35] ), or when at least P1 position (i.e. the position to the left of the scissile bond) is occupied by Gly or Ala (alphaviruses -nsP3/nsP4 site [451, coronaviruses-p28 site [47, 48] ). Similar potential cleavage/recognition sites are found at corresponding positions in polyproteins of other viruses containing the TYMV-like proteinase domain (M.N.R., unpubl.).

Does the putative NTPase/helicase domain influence (regulate) autoproteotysis of RP? Previous studies have not excluded this possibility [15] . Deletion of the putative TYMV NTPase domain or mutation of the invariant Lys982, an essential amino acid residue in the vast majority of NTP-binding enzymes [19, 49] did not affect autoproteolysis (Fig. 3A, B) . Consequently, the putative NTPase is not required for cleavage of RP in reticulocyte lysates.

Our results were recently confirmed by experiments [50] following our prediction of the TYMV papain-like proteinase domain and our mutagenesis data of the predicted catalytic Cys783 and His869 [51] . Extensive deletion analysis has shown that the proteinase can tolerate deletions upstream from residue 731 and downstream from residue 885 in vitro [50] .

Computer-assisted sequence analysis predicts that proteinase domains related to that of TYMV are encoded upstream from the putative NTPase gene of tymoviruses, carlaviruses, capilloviruses, ASPV and ACLSV. We suggest to name the new group the TYMO-LIKE PROTEINASE GROUP, since TYMV putative NTPase/ helicase; pol (putative) polymerase; P or P papain-tike proteinase predicted/characterized previously or in this study, respectively; r/t read-through; nsPl, nsP2, nsP3 and nsP4 are the alphavirus non-structural proteins; diamonds: stop codons; arrows: cleavage sites; SBWMV soilborne wheat mosaic furovirus is the prototype of this new group. In addition to the invariant papain-like Cys-His dyad, the tymo-like proteinases contain a third highly conserved amino acid residue, Gly (Gly821 in TYMV RP), followed by a hydrophobic residue. This motif is apparently missing from other papain-like proteinases and is thus an additional hallmark of this group. After the Cys-and His-containing motifs, the conserved Gly seems to be the third interesting target for mutagenesis experiments. The putative proteinase of BNYVV does not conserve the Gly residue, although it resembles the tymo-like proteinases at a statistically significant level. This is not surprising considering different location of this domain (between the putative NTPase and polymerase domains, Fig. 4 ) in the potyprotein and the fact that other BNYVV replicative proteins are not closely related to those of any plant RNA virus [11, 16] . Our attempts to produce significant sequence alignments between tymo-like and other proteinases were unsuccessful. We conclude that tymo-like proteinases constitute a distinct group (Fig. 2) within the majority of viral papain-like enzymes [4] , and speculate that the members of this group may share biochemical and biological features. Taking into account the indispensability of the essential papain-like Cys-His dyad for replication of TYMV [50] , (K. Sdron, G. Kadar6 [52] , it is tempting to speculate that all tymo-like proteinases cleave the large viral polyproteins yielding a functional "methyltransferase-helicase" block [11, 53] as an N-terminal product and an RNA polymerase domain as a C-terminal product. Notably, the alphavirus counterparts of the TYMV cleavage products, P123 and P34 are essential to form the viral replicative complex [54] . The other six groups of Sindbis-viruses shown on Fig. 4 do not seem to utilize the strategy of proteolysis of their proteins. Nevertheless 5 of them, i.e. tobamo-, tobra-, furo (SBWMV), tricorna-and hordeiviruses, do produce the "methyltransferase-helicase" block as a separate protein, but use two different mechanisms [5] . Thus, the strategy of processing of the large nonstructural viral polyprotein by the papain-like domain seems to be one of several possible solutions to the problem of organizing and regulating production of the viral replicase complex.

This study shows (see also Fig. 4 ) that the large replicase polyproteins of not only all animal Sindbis-like viruses but also of many of their plant counterparts such as tymo-, carlo-, capilloviruses, ACLSV, ASPV, and BNYVV possess (putative) papain-like proteinase domains in addition to conserved (putative) methyltransferase, helicase and polymerase domains [11] . We believe that this reflects similar functional and/or evolutionary constraints on the organization of the replicative machineries of animal and plant viruses. However, precise roles of the TYMV-like proteinases in virus replication and other potential functions remain to be elucidated.

Cysteine proteases of positive strand RNA viruses and chymotrypsin-like serine proteases: a distinct protein superfamily with a common structural fold

Viral cysteine proteases are homologous to the trypsin-like family of serine proteases: structural and functional implications

Processing of nonstructural polyproteins of Sindbis virus: nonstructural proteinase is in the C-terminal half of nsP2 and functions both in cis and in trans

Putative papain-related thiol proteases of positive-strand RNA viruses: identification of rubi-and aphthovirus proteases and delineation of a novel conserved domain associated with proteases of ruN-, ~-and coronaviruses

Genome similarities between plant and animal RNA viruses

Plant viral proteinases

Overlapping open reading frames revealed by complete nucleotide sequencing of turnip yellow mosaic virus genomic RNA

Cis-preferential replication of the turnip yellow mosaic virus RNA genome

Expression of ORF-69 of turnip yellow mosaic virus is necessary for viral spread in plants

Infectious TYMV RNA from cloned cDNA: effects in vitro and in vivo of point substitutions in the initiation codons of two extensively overlapping ORFs

Conservation of the putative methyltransferase domain: a hallmark of the 'Sindbis-like' supergroup of positive-strand RNA viruses

Unexpected close relationship between the large nonvirion proteins of filamentous potexviruses and spherical tymoviruses

Proteolytic origin of the 150-kilodalton protein encoded by turnip yellow mosaic virus genomic RNA

Comparison of the strategies of expression of five tymovirus RNAs by in vitro translation studies

Proteolytic maturation of the 206-kDa nonstructural protein encoded by turnip yellow mosaic virus RNA

Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses

Generation of a complete local similarity map for two biopolymer sequences (DOTHELIX program of the GENEBEE package)

Multiple sequence alignment with hierarchical clustering

An NTP-binding motif is the most conserved sequence in a highly diverged monophytetic group of proteins involved in positive strand RNA viral replication

The nucleotide sequence of the genomic RNA of kennedya yellow mosaic tymovirus-Jervis Bay isolate: relationships with potex-and carlaviruses

Nucleotide sequence of the ononis yellow mosaic tymovirus genome

Nucleotide sequence of the genome of eggplant mosaic tymovirus

Comparisons of the genomic sequences of erysimum latent virus and other tymoviruses: a search for the molecular basis of their host specificities

The genome organization of potato virus M RNA

Nucleotide sequences of apple stem pitting virus and of the coat protein gene of a similar virus from pear associated with vein yellows disease and their relationship to potex-and carlaviruses

Nucleotide sequence and genomic organization of apple chlorotic leaf spot closterovirus

I992) The nucleotide sequence of apple stem grooving capillovirus genome

Nucleotide sequence of beet necrotic yellow vein virus RNA-1

Haenni A-I, (1993) in vitro transcripts of turnip yellow mosaic virus encompassing a long 3' extention or produced from a full-length cDNA clone harbouring a 2kb-long PCR-amplified segment are infectious

Cleavage of structural proteins during the assembly of the head of bacteriophage T4

Alphavirus proteinases

Expression of the rubella virus nonstructural protein ORF and demonstration of proteolytic processing

Identification of the catalytic sites of a papain-like cysteine proteinase of murine coronavirus

The 5' end of the equine arteritis virus replicase gene encodes a papainlike cysteine protease

Identification of essential residues in potyvirus proteinase HC-Pro by site-directed mutagenesis

The autocatalytic protease p29 encoded by a hypovirutence-associated virus of the chestnut blight fungus resembles the potyvirusencoded protease HC-Pro

Gene expression by a hypovirulence-associated virus of the chestnut blight fungus involves two papain-like proteinase activities

Papain-like proteinase of turnip yellow mosaic virus residues and cleavage site requirements for p48 autoproteolysis

Antiviral effects of a thiol protease inhibitor on foot-and-mouth disease virus

The nucleotide sequence of RNA 2 of barley yellow mosaic virus

Nucleotide sequence of barley yellow mosaic virus RNA 2

Processing of pestivirus polyprotein: cleavage site between autoprotease and nucleocapsid protein of classical swine fever virus

Evolution and taxonomy of positive-strand RNA viruses: implications of comparative analysis of amino acid sequences

Stem bromelain: amino acid sequence and implications for weak binding of cystatin

Cleavage between nsP1 and nsP2 initiates the processing pathway of sindbis virus nonstructural polyprotein P123

Characterization of the potyviral HC-Pro autoproteolytic cleavage site

Analysis of MHV-A59 p28 cleavage site

Identification of the cleavage site of the first papain-like cysteine proteinase of routine coronavirus

Binding ofnucleotides by proteins

Identification of the essential cysteine and histidine residues of the turnip yellow mosaic virus protease

Papain-related proteinase of turnip yellow mosaic virus. NATO/EEC Course "Regulation of gene expression by animal viruses

The structure and expression of the genome of carlaviruses

Identification of the domains required for direct interaction of the helicase-like and polymeras-like RNA replication proteins of brome mosaic virus

Assembly of functional Sindbis virus RNA replication complexes: requirement for coexpression of P123 and P34

We thank Drs, A. Gorbalenya 

Received June 28, 1994