Characterization of the protease domain of Rice tungro bacilliform virus responsible for the processing of the capsid protein from the polyprotein

Abstract

Background: Rice tungro bacilliform virus (RTBV) is a pararetrovirus, and a member of the family Caulimoviridae in the genus Badnavirus. RTBV has a long open reading frame that encodes a large polyprotein (P3). Pararetroviruses show similarities with retroviruses in molecular organization and replication. P3 contains a putative movement protein (MP), the capsid protein (CP), the aspartate protease (PR) and the reverse transcriptase (RT) with a ribonuclease H activity. PR is a member of the cluster of retroviral proteases and serves to proteolytically process P3. Previous work established the N- and C-terminal amino acid sequences of CP and RT, processing of RT by PR, and estimated the molecular mass of PR by western blot assays.

Results: A molecular mass of a protein that was associated with virions was determined by in-line HPLC electrospray ionization mass spectral analysis. Comparison with retroviral proteases amino acid sequences allowed the characterization of a putative protease domain in this protein. Structural modelling revealed strong resemblance with retroviral proteases, with overall folds surrounding the active site being well conserved. Expression in E. coli of putative domain was affected by the presence or absence of the active site in the construct. Analysis of processing of CP by PR, using pulse chase labelling experiments, demonstrated that the 37 kDa capsid protein was dependent on the presence of the protease in the constructs.

Conclusion: The findings suggest the characterization of the RTBV protease domain. Sequence analysis, structural modelling, in vitro expression studies are evidence to consider the putative domain as being the protease domain. Analysis of expression of different peptides corresponding to various domains of P3 suggests a processing of CP by PR. This work clarifies the organization of the RTBV polyprotein, and its processing by the RTBV protease.

Figure 1

Mass spectrometry analysis performed on RTBV virions. In-line HPLC electrospray ionization mass spectrometry analysis performed on RTBV virions. Virus sample was denatured with guanidium hydrochloride 4.8 M prior, to injection onto the column. Analysis of peptide which coeluted from the in-line HPLC column in various charge states gave a molecular mass of 13,794 ± 4 Da.

Figure 2

Putative protease domains for RTBV. Position of five peptides that include the active protease domains with molecular mass that correspond to the mass spectrometry analysis. Peptide A has a predicted molecular mass of 13,790.71 kDa; peptide B of 13,790.71 kDa; peptide C of 13,796.70 kDa; peptide D of 13,793.60 kDa; peptide E of 13,795.70 kDa. Underlined sequences represent the active site of the protease. The grey box indicates a conserved motif among retroviral proteases. Numbers above arrows indicate position of amino acids in P3.

Figure 3

Structural sequence alignment of the RTBV protease with other retroviral proteases. Sequence alignment of the RTBV protease amino acid sequence with proteases of Rous sarcoma virus (RSV), Equine infectious anemia virus (EIAV) and Human immunodeficiency virus (HIV). The color scheme corresponds to percentage of similarity (based on physico-chemical properties). Black background and white foreground indicate 100%, grey background and white foreground indicate 80%, grey background and black foreground indicate 60%. Lower similarity values are not shown. Numbers over the alignment indicate the alignment length. Secondary structure elements from the RSV sequence are represented over the alignment. The numbering of the elements follows the RSV numbering based on structure. Boxes indicate beta strand elements assigned as β. The helix is represented as a cylinder and indicated as α. Thick lines connecting the elements are loops and dashed lines indicate a break in the sequence. The black triangle indicates the location of the active site.

Figure 4

Structural modelling of the RTBV protease. Structural modelling of the RTBV protease (A), and Rous sarcoma virus (RSV) protease used as template (B). The sphere indicates the N-terminal end, aspartic acid of active site is shown in the stick model. In red is the RSV protease inhibitor 39 coupled to the active site. The first residues of RTBV PR could not be modelled. Conservation of the active site and overall fold recognition analyses with modelling building show that the RTBV sequence resembles greatly a protease fold.

Figure 5

Polyprotein P3 peptide domains cloned in different constructs. Visualization of P3 peptide domains cloned in different constructs. Parts in grey are sequence derived from the full-length RTBV clone pBSR63A 11. Parts in black are sequence imported from plasmid pBS-mp/RT 19, containing the protease mutated active site. Parts in white are sequences from vectors. Underlined restriction enzymes are sites that are present in ORF3.

Figure 6

Induction of the putative protease domain. Expression of peptides in E.coli. Numbers on the left are estimated sizes in kDa of the molecular weight marker. (A) Coomassie blue-stained gel of induced peptides in E.coli. Lane 1: pTr-PR; Lane 2: pTr-mPR. (B) Western blot performed on induced peptides using antibodies raised against RTBV (Ab-RTBV). Lane 3: pTr-PR; Lane 4: pTr-mPR. (C) Western blot performed on induced peptides using antibodies raised against PR domain (Ab-PR). Lane 5: pTr-mPR. Peptide PR could not be induced from pTr-PR. pTr-mPR induced a specific peptide of about 14 kDa, corresponding to the protease domain (with mutation), and recognized by Ab-PR.

Figure 7

In vitro releasing of the coat protein from the polyprotein P3. Autoradiography of an SDS-PAGE of induced peptides from different pET-vectors induced in E. coli. ³⁵S radiolabelled methionine was added for 5 minutes after 60 minutes of induction with IPTG. Numbers (in kDa) on the left indicate mobility of the molecular weight markers. Lane 1: pET(no insert); Lane 2: pET-MP; Lane 3: pET-MP-PR; Lane 4: pET-MP-mPR; Lane 5: pET-P3; Lane 6: pET-mP3. Arrow shows the presence of a peptide (estimated molecular mass of 37 kDa) that is present only for constructs that code a peptide that contains the coat protein and the protease (pET-MP-PR; pET-P3).

Figure 8

Protease cleavage site sequences in the RTBV polyprotein P3. Protease cleavage site sequences in the RTBV polyprotein P3. The designation of amino acid residues spanning the cleavage site is according to [40]. MP: movement protein; IR: intervening region; CP: capsid protein; PR: Protease; RT: Reverse transcriptase ; Rnase H: Ribonuclease H. Cleavage site sequences MP/IR has not been determined yet. A lack of significant sequence similarities is observed, a characteristic of aspartate proteases.