English   español  
Please use this identifier to cite or link to this item: http://hdl.handle.net/10261/29172
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Exportar a otros formatos:


Implication of the C terminus of the Prunus necrotic ringspot virus movement

AuthorsAparicio Herrero, Frederic ; Pallás Benet, Vicente ; Sánchez-Navarro, J. A.
33 amino-acids
Issue Date10-Mar-2010
PublisherSociety for General Microbiology
CitationJournal of General Virology, 91: 1865-70 (2010)
AbstractAbstract 24 The movement protein (MP) of Prunus necrotic ringspot virus (PNRSV) is 25 required for viral transport. Previous analysis with MPs of others members of the 26 family Bromoviridae have shown that the C-terminal part of these MPs plays a critical 27 role in the interaction with the cognate coat protein (CP) and cell-to-cell transport. 28 Bimolecular fluorescence complementation and overlay analysis confirm an interaction 29 between the C-terminal 38 amino acids (aa) of PNRSV MP and its cognate CP. 30 Mutational analysis of the C-terminal region of the PNRSV MP revealed that its C31 terminal 38 aa are dispensable for virus transport, however, the 4 aa preceding the 32 dispensable C-terminus are necessary to target the MP to the plasmodesmata and for the 33 functionality of the protein. The capacity of the PNRSV MP to use either a CP34 dependent or a CP-independent cell-to-cell transport is discussed.Prunus necrotic ringspot virus (PNRSV) belongs 36 to the genus Ilarvirus (family 37 Bromoviridae). The PNRSV genome consists of three, single-stranded, plus sense 38 RNAs. RNAs 1 and 2 encode P1 and P2 polymerase protein subunits, respectively. 39 RNA 3 encodes directly the MP whereas the CP is synthesized via a subgenomic RNA 40 4 (Sánchez-Navarro and Pallás 1997). 41 The movement protein (MP) of PNRSV belongs to the 30K superfamily 42 (Melcher, 2000). The C-terminal region of the 30K MPs has a predicted random coiled 43 secondary structure (Melcher, 2000). The Alfalfa mosaic virus (AMV) MP specifically 44 interacts with its cognate CP through its C-terminal region and in Cowpea mosaic virus 45 (CPMV) MP this region is the virion-binding domain (Sánchez-Navarro et al., 2006; 46 Carvalho et al., 2003). In the case of the MP of the Cauliflower mosaic virus (CaMV), 47 the C terminus interacts indirectly to the cognate CP via the virus associated protein 48 VAP (Stavolone et al., 2005). The C-terminal region of the corresponding MPs of 49 Cucumber mosaic virus (CMV) and Brome mosaic virus (BMV) controls the 50 requirement of the CP for the intercellular movement (Nagano et al, 2001; Sánchez- 51 Navarro and Bol, 2001; Takeda et al., 2004). Deletion of the C-terminal 55 aa of 52 Tobacco mosaic virus (TMV) MP does not affect the cell-to-cell transport of the TMV 53 nor the cell wall localization of the protein (Berna et al., 1991). In the present work, we 54 have analysed the role in cell-to-cell movement of the PNRSV MP C-terminal region by 55 mutational analysis. We also analyzed the putative MP-CP interaction by bimolecular 56 fluorescence complementation (BiFC) and overlay assays, respectively. Finally, we 57 present data indicating that the capacity of the PNRSV MP to mediate virus transport is 58 independent of the MP-CP interaction. 59 Due to the lack of PNRSV infectious clones, a chimera cDNA3 construct 60 between AMV and PNRSV that expresses the green fluorescent protein (GFP), the wild
type PNRSV MP fused in frame to the C-terminal 61 44 aa of AMV MP (A44) and the 62 AMV CP was used (clone pGFP/MP:A44/CP in Sánchez-Navarro et al., 2006; hereafter 63 pMPP:A44). The A44 is required for a specific interaction with the AMV CP to render a 64 functional chimeric AMV/PNRSV RNA 3. The effect of different C-terminal deletions 65 of the PNRSV MP on the functionality of the protein in vivo was examined. MP genes 66 lacking the C-terminal 12, 38 or 42 aa were introduced in the plasmid pMPP:A44 to 67 generate pMPP271:A44, pMPP245:A44 and pMPP241:A44 constructs, respectively. 68 Derived transcripts were inoculated onto leaves of transgenic Nicotiana tabacum plants 69 expressing the AMV P1 and P2 proteins (P12 plants). Deletion of the C-terminal 12 or 70 38 aa did not block the virus transport meanwhile deletion of the C-terminal 42 residues 71 rendered an infection limited to single cells (Fig. 1). Moreover, the size of the foci was 72 reduced in the C-terminal 12 and 38 mutants (Supplementary Table 1). These results 73 indicate that the C-terminal 38 aa are dispensable for virus transport but also that 74 residues at position 242-245 are critical for the functionality of the MP. In fact, when 75 the 4 aa (242-245) were deleted in the wt MP (Fig. 1, pMPPΔ4:A44 construct) only 76 single infected cells were observed. Previous results obtained with MPs of the 30K 77 family have shown that the dispensable C-terminal portion of the MP plays a role in 78 determining the specificity of the interaction between the MP and CP (Sánchez-Navarro 79 et al., 2006; Nagano et al., 1997). In the case of the chimera AMV constructs this 80 problem was overcome by fusing the A44 at the C terminus of the heterologous MP. To 81 determine if the minimal PNRSV MP C-terminal mutant that is functional in vivo 82 required the presence of the A44, we inserted a stop codon after the 245 residue to 83 generate the plasmid pMPP245stop. This construct rendered similar infection foci than 84 the pMPP245:A44 chimera (Fig. 1) indicating that the PNRSV MP lacking the C85 terminal 38 aa is not inhibited by the presence of the heterologous AMV CP. To analyze
the effect of the deleted RNA coding sequences 86 in pMPP245stop:A44 and 87 pMPP241:A44 constructs, the corresponding stop codons were inserted between 245- 88 246 and 241-242 codons in the wt PNRSV MP gene, respectively. Similar results were 89 obtained with these new mutants indicating that the effects observed are exclusively 90 related to the deleted aa sequences. 91 To understand why some of the truncated PNRSV MP mutants were defective in 92 virus transport we analyzed their subcellular localization. All truncated PNRSV MP 93 genes were fused to the GFP by exchanging the PNRSV MP gene in the binary vector 94 pMPP:GFP (Herranz et al., 2005) to generate pMPP271:GFP, pMPP245:GFP, 95 pMPP241:GFP and pMPPΔ4:GFP plasmids (Fig. 2). The transiently expressed PNRSV 96 MP:GFP fusion protein (MPP:GFP) in N. benthamiana plants was located in punctate 97 structures at the cell wall as reported previously (Fig. 2a; Herranz et al., 2005). The 98 MPP:GFP co-localized with the well established plasmodesmata marker TMV MP 99 (MPT:DsRed in Fig. 2a). Transient expression of the MP C-terminal mutants supporting 100 virus transport presented the punctated localization in the cell wall (Fig. 2b, constructs 101 pMPP:GFP, pMPP271:GFP, pMPP245:GFP). However, MP mutants defective in virus 102 transport showed a diffuse fluorescence around the cell wall (Fig. 2b, constructs 103 pMPP241:GFP, pMPPΔ4:GFP,) suggesting that these mutants have altered its capacity 104 to interact with the plasmodesmata. In this sense, AMV mutants defective in 105 plasmodesmata localization were also affected in virus transport (Huang et al., 2001). It 106 is worth to note that the C-terminal 32 residues are the most variable part of the PNRSV 107 MP (Aparicio and Pallas, 2002) whereas, the region just upstream is highly conserved 108 and contains five sites under negative evolutive selection (V231, D236, R238, T242 and 109 P248) which could be an indication of its importance to maintain the functionality of the 110 protein (Fiore et al., 2008). Our results showed that the small region between residues
242-245 (mutant pMPΔ4:GFP) is critical for both the function 111 of the protein and its 112 subcellular localization. Similar results have been reported for the MP of TMV in which 113 the 19 aa preceding the dispensable C-terminal 55 residues are essential for both the 114 localization of the MP to the cell wall and the functionality of the protein (Berna et al., 115 1991). Thus, this small region could play a role maintaining the three-dimensional 116 structure of the protein or alternatively, represents a potential signal addressed to target 117 the MP to the plasmodesmata. However, we did not find a consensus sequence between 118 the small region of 4 aa and the rest of MPs of the 30K family. 119 Data obtained with AMV indicate that the virus could traffic through 120 plasmodesmata either as mature virions or by RNA-CP complexes where the C 121 terminus of AMV MP, although dispensable, could confer specificity to the transport 122 process via an interaction with the cognate CP (Sánchez-Navarro and Bol, 2001; 123 Sánchez-Navarro et al. 2006). Co-variation analysis performed between the MP and CP 124 genes of PNRSV suggested a putative interaction between the C-terminal region of the 125 MP and the N-terminal part of the CP (Codoñer et al., 2006). We analyzed the 126 postulated PRNSV MP-CP interaction by in vitro assays and in planta by BiFC. To 127 perform the BiFC assay, the C-terminal fragment of the yellow fluorescent protein 128 (YFP) was fused to the carboxy-terminal part of the full-length and MP mutants 129 whereas the N-terminal YFP fragment was fused to the PNRSV or AMV CPs (Aparicio 130 et al., 2006). Specific pairs of fusion proteins were transiently co-expressed in 131 Nicotiana benthamiana leaves by agroinfiltration. Reconstitution of fluorescence was 132 observed with the combination MPP-CYFP or MPP271-CYFP plus NYFP-CP (Fig. 3a: 133 panels 1-2). However, inconclusive results were obtained with the MPP245-CYFP 134 construct (data not shown). To resolve this problem the possible interaction between 135 MPP245 and the CP was investigated using the overlay assay. The full-length MP and
the MPP245 and MPPΔ4 mutants, all carrying the HA 136 epitope at its C terminus, were 137 cloned in the protein expression plasmid pDUET (Novagen). The expression of the 138 fusion proteins in BL21 cells (Novagen) was confirmed by Western blot analysis using 139 an anti-HA antibody (Fig. 3c). Total protein extracts were electroblotted in duplicate to 140 PVDF membranes and then subjected to overlay assays as described previously (Chen 141 et al., 2000). The membranes were incubated with either a purified 6 histidine-tagged 142 PNRSV CP (Fig. 3c, 6hisCP) or a protein extract derived from non-expressing bacteria, 143 as negative control (Fig. 3c, BL21 total). The 6hisCP specifically interacted with full144 length MP and MPΔ4 mutant but not with truncated MP245 (Fig. 3c). No interaction 145 was detected when the membrane was incubated with the negative control. This result 146 confirms the capacity of the PNRSV CP to associate to its cognate MP and suggests that 147 this interaction is mediated exclusively by the C-terminal 38 aa of the MP. To confirm 148 this hypothesis we performed the BiFC assay using only this C-terminal 38 aa 149 (ct38MPp-CYFP). Reconstitution of the fluorescence was observed when ct38MPp- 150 CYFP was co-expressed with NYFP-CP (Fig. 3a, panel 3). No signal was detected with 151 the negative control (Fig. 3a panel 4: NYFP + ct38MPp-CYFP). Similar negative results 152 were obtained with the pairs MPP-CYFP or MPP271-CYFP plus NYFP or the AMV CP 153 (not shown). The expression of all fusion proteins used in the BiFC assay was checked 154 by western blot (Figure 3b). Our results indicate that the MP-CP interaction is mediated 155 through the C-terminal 38 aa of the MP supporting the previous in silico analysis 156 determining that the PNRSV MP residues interacting with the CP are at position 253, 157 256, 257 and 261 (Codoñer et al., 2006). These results, together with previous analysis 158 of MPs assigned to the 30K family, point out an MP scheme in which the majority of 159 the protein (N terminus) is required to allow virus transport, meanwhile the C terminus 160 contains the specific CP interacting determinants. If both regions are independent, MPs
lacking the C-terminal region necessary for the CP interaction be 161 came nonspecific MPs 162 that have the capacity to transport any virus. With this idea we can expect as well, that 163 different MPs can be functional in a heterologous virus by fusing the specific C164 terminal region required for a compatible interaction with the corresponding CP. There 165 are examples for both situations. The MPs of six different viruses are functional in the 166 AMV virus when the A44 is fused at the C terminus of the heterologous MPs (Sánchez- 167 Navarro et al., 2006). The idea of a nonspecific MP is supported by the results 168 presented herein. Thus, the PNRSV MP lacking the C-terminal 38 aa allows the 169 transport of the heterologous AMV. Similar observation has been reported for other 170 Bromoviridae viruses: CMV or BMV. In addition, in those viruses it has been reported 171 that the C terminus of the MP could modulate the requirement (CP-dependent) or not 172 (CP-independent) of the CP for virus transport (Nagano et al., 2001; Takeda et al., 173 2005; Akamatsu et al., 2007). Probably, similar CP-independent virus transport scenario 174 mediated by the C terminus of the MP could occur in Ilarviruses and /or Alfamoviruses. 175 However, this aspect can not be analyzed since the CP is also required for other viral 176 functions (e.g. replication and/or translation; Bol, 2005). In the case of the CMV MP, 177 the deletion of the dispensable C terminus region induced an increasing of the MP RNA 178 binding affinity indicating that perhaps both effects, the lack of the C terminus and the 179 increasing of the RNA affinity could be required to convert a CP-dependent protein into 180 a CP-independent one (Kim et al., 2004). In BMV, MP-CP interaction requires mature 181 virion formation although an isolate has been described, capable to infect dicot plants, 182 that is able to move between cells independently of CP (Takeda et al., 2004; 2005). In 183 this case, the authors reported that single aa differences at the C terminus of BMV MP 184 are sufficient to alter the requirement for CP in the movement. Apparently, not only the 185 deletion but also modification of the C terminus could be sufficient to alter the CP
requirement of MPs assigned to the 30K family. The open question 186 in this work is to 187 know if the BMV MP variant could be functional in a heterologous virus with an 188 unrelated CP. In the case of the TMV MP this question is already answered. The TMV 189 MP has the same scheme of the rest of the MPs in which the C-terminal 55 aa are not 190 necessary for the functionality of the protein. However and unlike the majority of MPs 191 of the 30K family, the TMV MP functions as an nonspecific MP with no other 192 modification of its C terminus (Cooper et al., 1996, Sánchez-Navarro et al., 2006). 193 Regarding the rest of MPs of the 30K family that have been analysed it can be 194 concluded that the C-terminal region of MPs is associated directly or indirectly with the 195 CP interaction and its deletion or modification is sufficient to generate an nonspecific 196 viral transporting protein. In this sense, it will be very interesting to analyze if the 197 deletion of the C-terminal portion responsible for the CP interaction modifies the CP 198 dependence in the group of viruses that are transported as virus particles e.g., CPMV or 199 CaMV (Carvalho et al., 2003; Stavolone, et al., 2005).
Publisher version (URL)http://dx.doi:10.1099/vir.0.019950-0
Appears in Collections:(IBMCP) Artículos
Files in This Item:
File Description SizeFormat 
JGV20101865Pallas.pdf446,53 kBAdobe PDFThumbnail
Show full item record
Review this work

WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.