Reverse genetics system for Chandipura virus: Tagging the viral matrix protein with green fluorescent protein

Reverse genetics system for Chandipura virus: Tagging the viral matrix protein with green fluorescent protein

Virus Research 160 (2011) 166–172 Contents lists available at ScienceDirect Virus Research journal homepage: www.elsevier.com/locate/virusres Rever...

357KB Sizes 1 Downloads 87 Views

Virus Research 160 (2011) 166–172

Contents lists available at ScienceDirect

Virus Research journal homepage: www.elsevier.com/locate/virusres

Reverse genetics system for Chandipura virus: Tagging the viral matrix protein with green fluorescent protein Anthony C. Marriott ∗ , Crystal A. Hornsey Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK

a r t i c l e

i n f o

Article history: Received 9 March 2011 Received in revised form 7 June 2011 Accepted 9 June 2011 Available online 16 June 2011 Keywords: Chandipura Rhabdovirus Reverse genetics GFP Matrix Deletion mutant

a b s t r a c t Chandipura virus (CV; genus Vesiculovirus, family Rhabdoviridae) is an emerging arbovirus, responsible for a number of outbreaks of severe viral encephalitis affecting children in India. A reverse genetics system has been constructed which allows recovery of infectious recombinant CV (rCV) entirely from cDNA. This system was used to construct a virus, rCVE, which has an additional transcription unit encoding green fluorescent protein (EGFP) between the 3rd and 4th CV genes. This virus grew to titres comparable to the parental rCV and stably expressed EGFP in infected cells for at least 4 passages. A second virus, rCVGM, was constructed in which the CV matrix (M) coding region was replaced with the coding region for an EGFP-M fusion protein. Compared to rCV and rCVE, rCVGM was attenuated, giving a small plaque phenotype and lower yield, although it did express the EGFP-M protein in infected cells. Passage of rCVGM resulted in viruses with a standard plaque phenotype which no longer expressed EGFP. Analysis of two of these viruses showed that most or all of the EGFP ORF was deleted. The EGFP-M fusion protein showed cleavage to EGFP-sized and M-sized products, both in rCVGM-infected cells and when expressed from a plasmid. The EGFP-M fusion protein was not detected in virus particles, suggesting it was incompatible with virus assembly, and particles of rCVGM likely contained the M-sized cleavage product in its place. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Chandipura virus (CV) is an emerging pathogen, having been implicated in 6 outbreaks of severe encephalitis of children in India since 2003 (Rao et al., 2004, 2007; Tandale et al., 2008; Chadha et al., 2005; Gurav et al., 2010). The virus was first isolated in Nagpur, India, during an outbreak of febrile illness (Bhatt and Rodrigues, 1967). Serological studies suggested a widespread distribution in humans and animals in India (Tandale et al., 2008; Joshi et al., 2005), but little disease was noted prior to 2003. Isolation of CV from sandflies and its ability to infect a range of haematophagous insects in the laboratory suggests CV is an arbovirus (Geevarghese et al., 2005; Mavale et al., 2007; Ilkal et al., 1991), and it has been classified in the genus Vesiculovirus, family Rhabdoviridae (Tesh et al., 1983). The non-segmented, negative-sense RNA genome has been sequenced, and reveals that CV is phylogenetically closest to Isfahan and Piry viruses, and more distantly related to the vesicular

Abbreviations: CV, Chandipura virus; EGFP, enhanced green fluorescent protein; EMCV, encephalomyocarditis virus; IRES, internal ribosome entry site; nt, nucleotides; ORF, open reading frame; pfu, plaque forming unit; RFP, red fluorescent protein; RNP, ribonucleoprotein; VLP, virus-like particle; VSV, vesicular stomatitis virus. ∗ Corresponding author. Present address: 3 Southbank Court, Kenilworth, CV8 1LB, UK. Tel.: +44 01926 864560. E-mail address: [email protected] (A.C. Marriott). 0168-1702/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2011.06.007

stomatitis viruses (VSV) (Marriott, 2005). CV shares a number of properties with VSV, such as a small genome comprising only 5 genes, rapid growth to high titres in cultured mammalian cells, and the ability to infect a wide range of vertebrate and invertebrate cell lines (Marriott, 2005; Jadi et al., 2010). Unlike VSV, however, CV does not cause clinical disease in domestic animals (Wilks and House, 1986). On the basis of these properties, we reasoned that CV would make a good model virus for the study of virus–host interactions, and to that end we developed a minireplicon reverse genetics system for CV (Marriott and Hornsey, manuscript in preparation). As with other rhabdoviruses, the minimal unit required for viral transcription and replication is the genomic RNA (or a minireplicon analogue), and the nucleocapsid (N), phospho(P) and polymerase (L) proteins which form the viral ribonucleoprotein (RNP). Each virus protein is expressed from a separate transcription unit (gene), which is bounded by conserved gene start and gene end signals which direct the initiation and termination/polyadenylation of the mRNA. Formation of virus particles also requires the matrix (M) protein, which links the RNP to the viral envelope, and the glycoprotein (G) which is embedded in the envelope and acts as the attachment and fusion protein (Basak et al., 2007; Rose and Whitt, 2001). Reverse genetics can be used to construct recombinant viruses expressing proteins genetically tagged with a fluorescent protein, which can then be used to probe infection of cells in real time; examples include measles virus, VSV and rabies virus (Duprex et al., 2002; Finke et al., 2004; Das et al., 2006;

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

Klingen et al., 2008; Ruedas and Perrault, 2009). To enable investigation of CV, an infectious clone was produced expressing a protein tagged with enhanced green fluorescent protein (EGFP). It has been reported previously that the M protein of CV is able to inhibit host gene expression in a similar manner to the M protein of VSV (Taylor et al., 1999), and that the CV M protein tagged with EGFP retains this activity and localises to the nuclear periphery in transfected cells (Petersen et al., 2001). In this paper we report development of the CV reverse genetics system, and use of this system to replace the wild-type M gene of CV with a gene encoding an EGFP-M fusion protein. 2. Materials and methods 2.1. Matrix-EGFP fusion plasmids The coding region for EGFP-M fusion protein was amplified from pEGFP-CV-M (Petersen et al., 2001) by PCR and inserted into plasmid pT7.2 to give plasmid pT7EGFPM, such that transcription by T7 RNA polymerase yields an mRNA with the EMCV internal ribosome entry site (IRES) at the 5 end, thus directing cap-independent translation of the ORF. The equivalent plasmid pT7EGFP, which expresses only EGFP, was used as a control. The alternate fusion protein, MEGFP, which has the CV M sequence at the amino-terminus, was constructed by cloning the M ORF, amplified by PCR, upstream of the EGFP ORF in pT7EGFP to yield pT7MEGFP. 2.2. Minireplicon packaging assays The open reading frames encoding the CV N, P, L, M and G proteins were amplified by reverse transcription-PCR (RT-PCR) and cloned into vectors under the control of a T7 promoter and EMCV IRES sequence to produce plasmids pT7N, pT7P, pT7L, pT7M and pT7G. Elongase enzyme (Invitrogen) was used for most PCRs, in some cases KOD Hot-start polymerase (Merck Novagen) was used. The L gene was cloned into pTM1 (Moss et al., 1990) from 3 overlapping PCR products. The other 4 genes were cloned into pT7.2, which is a derivative of pTM1 lacking the vaccinia flanking sequences. The plasmid sequences were confirmed by dideoxy sequencing. Packaging assays comprised plasmids pT7N, pT7P, pT7L, pT7G and pT7M along with minireplicon mGFPLX, which were co-transfected into BSR-T7/5 cells (Buchholz et al., 1999). Plasmid mGFPLX is transcribed by T7 RNA polymerase to a negative-sense RNA containing the CV leader and trailer sequences, between which the EGFP coding region is flanked by the N gene start and L gene end sequences. Co-transfection of pT7lacZ was used to normalise for transfection efficiency. After 48 h, the supernatant was clarified and used to infect fresh cells in the presence of 3 pfu/cell of wild-type CV. After a further 24 h, cells were scored for EGFP fluorescence by epifluorescence microscopy. The pT7M plasmid was replaced by pT7EGFPM or pT7MEGFP where indicated.

167

for EGFP which had been amplified by PCR from the plasmid pEGFP-C1 (Clontech), yielding plasmid pT7CVE. Alternatively the wild-type M coding region of plasmid pT7CV was replaced by the coding region for the EGFP-M fusion protein which was amplified from pEGFP-CV-M, to yield pT7CVGM. Details of primers, cloning steps used and plasmid sequences are available from the authors. All plasmid constructions were confirmed by DNA sequencing. 2.4. Rescue of infectious recombinant viruses The genomic plasmid was transfected into BSR-T7/5 cells using TransIT-LT1 reagent (Mirus) along with the helper plasmids in the proportion of 0.8 ␮g genome plasmid to 0.9 ␮g pT7N, 150 ng pT7P and 150 ng pT7L. After 48–72 h at 37 ◦ C, the supernatant was used to infect fresh cells, either BSR-T7/5 or BS-C-1, and virus was harvested when the cytopathic effect was maximal. Virus titres were determined by plaque assay on BS-C-1 or Vero cells, using a 1% (w/v) carboxymethylcellulose overlay. Cells were stained with 0.04%, w/v Crystal Violet after fixation with 2% glutaraldehyde. 2.5. Fluorescence microscopy Cells were examined using a Nikon Diaphot 300 fluorescence microscope, using a filter set optimised for GFP. Images were captured with a QImaging digital camera and analysed using QImaging software (Media Cybernetics UK). 2.6. Protein analysis Cells were lysed in 1% (v/v) Triton X-100, separated by SDS-polyacrylamide gel electrophoresis, and transferred to nitrocellulose membrane (Hybond-ECL, GE Healthcare). Proteins were detected using a sheep polyclonal anti-CV IgG, or rabbit anti-GFP polyclonal IgG (Ab290, Abcam). 2.7. Sequence analysis of revertant viruses RNA was extracted from virus supernatant using Trizol LS reagent (Invitrogen) and reverse transcribed using random hexamer primers. The cDNA was then amplified in a PCR reaction using primers M2 (5 -ACTCTCTTAAGGAACTGGTCATTG-3 ) and RE1 (5 -TGGAGTAAATTCCATCATATCATC-3 ) (corresponding to nt 2150–2173 and 2465–2488, respectively of the wild-type CV genome sequence). PCR products were gel-purified (Qiaquick kit, Qiagen) and sequenced using primer M2. 3. Results 3.1. Expression of matrix-EGFP fusion proteins

2.3. Construction of CV genomic plasmids A full-length cDNA corresponding to the 11,119 nucleotide (nt) genome of the prototype 653514 strain of CV was cloned between a T7 promoter and the hepatitis delta virus ribozyme such that transcription by T7 RNA polymerase produced a positive-sense transcript, essentially as described previously for VSV (Lawson et al., 1995; Whelan et al., 1995). The wild-type plasmid, pT7CV was then modified in two ways. Firstly, an additional transcription unit was inserted into the NdeI site in the 3 untranslated region of the M gene, at position 3012 (numbering as in Marriott, 2005), comprising a gene junction sequence and the coding region

Matrix fusion proteins were expressed from T7 promoters in cells constitutively expressing T7 RNA polymerase to avoid the selfinhibition of M gene expression seen when the M gene is expressed from a polymerase II promoter (Taylor et al., 1999). EGFP was fused either to the amino-terminus (EGFP-M) or carboxyl-terminus (MEGFP) of the CV M protein. In each case, bright fluorescence was observed in transfected cells: the distribution of fluorescence was similar between EGFP-M and M-EGFP, with bright foci located in the cytoplasm close to the nucleus. This differs from the non-fused EGFP, which was evenly distributed throughout the cytoplasm and nucleus of transfected cells (data not shown).

168

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

the EGFP gene had been demonstrated, we replaced the M coding region with that encoding the EGFP-M fusion protein. EGFP-M was chosen rather than M-EGFP for two reasons: (i) as described above, EGFP-M protein showed a much greater ability than M-EGFP to function in the VLP packaging assay; (ii) it was reasoned that a virus encoding M-EGFP would be able to revert to an EGFP-negative phenotype by means of a single nonsense mutation at or near the end of the M coding sequence, whereas nonsense mutations in the EGFP-M gene would lead to a matrix-less virus which would be unable to propagate further.

10 6

VLP /0.1 ml

10 5

10 4

10 3

3.4. Rescue of Chandipura virus expressing EGFP-M fusion protein

10 2 M

EGFPM

MEGFP

no M

plasmid Fig. 1. Packaging of minireplicon into VLPs. In addition to plasmids pT7N, pT7P, pT7L, pT7G and mGFPLX, transfection mixes contained 0.3 ␮g of either pT7M, pT7EGFPM, pT7MEGFP, or empty vector pT7.2 (no M). Graph shows yield of VLP from transfected cells, determined by transduction of fluorescence onto fresh cells infected with CV. Bars show mean and standard deviation of 3 replicates.

3.2. Ability of matrix-EGFP fusion proteins to package a CV minireplicon RNA Since the EGFP tag approximately doubles the size of the M protein in the fusions, it may affect the ability of the protein to function in its essential role in packaging and budding of virus particles. To investigate this we used a newly developed minireplicon assay system for CV which involves co-transfection of plasmids expressing all 5 virus proteins, in addition to a minireplicon RNA which can express EGFP under the control of CV gene start and gene end signals. If all the components are functional, the transfected cells produce virus-like particles (VLPs) which contain the minireplicon RNA, and which can infect fresh cells. In the presence of super-infecting CV the minireplicon is transcribed to produce an EGFP mRNA. Thus EGFP fluorescence serves as an indicator that the minireplicon has been packaged into infectious VLPs (Marriott and Hornsey, unpublished). The packaging ability of the EGFP-M and M-EGFP fusion proteins was compared to wild-type M protein in the assay (Fig. 1). Both fusion proteins are clearly poorer than wild-type M protein in generating infectious VLPs, with EGFP-M producing a VLP titre 50-fold less, and M-EGFP 1100-fold less, than wild-type M. However both fusion proteins were able to induce the formation of significantly more VLPs than the control in which the M plasmid was replaced with empty vector (t-test, p < 0.05). 3.3. Rescue of infectious Chandipura virus from cDNA clones Genomic plasmids pT7CV and pT7CVE were transfected separately into cells expressing T7 RNA polymerase, along with pT7L, pT7P and pT7N which encode the minimum requirements for CV replication and transcription. Recombinant viruses designated rCV and rCVE were recovered and characterised. The genome of rCVE differs from rCV in having a sixth gene, encoding EGFP, located between the M and G genes (Fig. 2, panels a and b). Transfected cells produced up to 2 × 106 pfu of rCV per well by 6 days posttransfection. When used to infect fresh cell monolayers, both viruses produced plaques indistinguishable from those produced by non-recombinant prototype CV. When visualised under UV illumination, cells infected with rCVE showed bright green fluorescence in the infected cells (see Fig. 4h). The fluorescence was evenly distributed through the cytoplasm and nucleus, as seen for cells transfected with EGFP plasmid, and was also present in blebs present on the surface of infected cells (see Section 3.6). The expression of EGFP from rCVE was stable for at least 4 consecutive passages. Since the capacity of the CV genome to accommodate

The M gene of pT7CV was replaced with EGFP-M to give plasmid pT7CVGM, which was then recovered as infectious virus as described above (Fig. 2c). The recombinant virus rCVGM expressed green fluorescence in a pattern resembling that seen with the pT7EGFPM plasmid (see Fig. 4i). However rCVGM showed reduced growth in cell culture compared to rCV and rCVE, growing to 100-fold lower titres in BS-C-1 cells (Fig. 2d), and showed a smallplaque phenotype (Fig. 2f). The rCVGM virus grew equally well in interferon-competent (BS-C-1) and interferon-null (Vero) cells, and did not display a thermosensitive phenotype (Fig. 2g). Upon passage of this virus, it was often noted that larger plaque viruses would appear, which were EGFP-negative when examined by fluorescence microscopy. Four of 10 stocks of rCVGM were found to contain such viruses, which we shall refer to as revertants (as they have recovered a plaque phenotype similar to wild-type virus). The mean plaque sizes of two such revertants is shown in Fig. 2f, in comparison to the parental rCVGM and to rCV and rCVE. 3.5. Revertant viruses contain deletions in the EGFP coding region One stock which contained a high level of revertant plaques was analysed by PCR. RNA was extracted from clarified virus supernatant and amplified by PCR using primers designed to span the EGFP-insertion site (see Section 2.7). Two discrete products were obtained, of 350 and 490 bp, compared to the expected size of 1117 bp which was produced from the pT7CVGM template (not shown). These products were purified and sequenced, and revealed large deletions compared to the expected sequence of rCVGM (Fig. 3). In deletion-633, 633 nt were deleted from the EGFP coding region, such that translation from the first AUG codon would produce a peptide of only 14 amino acids. Translation from the next AUG downstream, in reading frame −1 relative to the mini-ORF, was predicted to give a product containing all of the M coding region, with 27 additional amino acids at the N-terminus. It is likely that this AUG was used, as it is in a strong translational context, GGCATGG. Deletion-768 showed a larger deletion of 768 nt, from within the 5 untranslated region of the gene to just upstream of the M coding region. Translation from the first AUG of this mutant would produce wild-type M protein (Fig. 3). The translational context for this start codon was GGAATGC, similar to the wild-type M start codon AAGATGC. 3.6. Cleavage of the EGFP-M fusion protein during virus infection The low growth rate of rCVGM, and its tendency to undergo deletion of the majority of the EGFP sequence (which was not observed for virus rCVE), suggested that the EGFP-M protein was defective in its role in virus propagation. In order to determine if the fusion protein could be incorporated into virus particles, we compared the production of EGFP-M and M-EGFP proteins in transfected cells by immunoblotting (Fig. 4, panels a and b). Although both plasmids expressed a full-length fusion protein, which was

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

169

Fig. 2. Recombinant viruses containing the EGFP coding region. (a) rCV, (b) rCVE and (c) rCVGM gene order, shown as mRNA-sense and to scale. Scale-bar below is in nucleotides; arrows represent ORFs. (d and e) Comparison of virus titres from multiple-step growth (d) or single-step growth (e) in BS-C-1 cells; bars represent mean and standard deviation of 3–4 replicates. (f) Comparison of plaque sizes between rCV, rCVE, rCVGM and 2 standard-plaque revertants derived from rCVGM (columns 1 and 2). Plaque diameter was determined by measuring enlarged micrographs of stained monolayers; scale is in arbitrary units; bars show mean and standard deviation of 10–20 plaques. (g) Growth of rCVGM in interferon-competent (BS-C-1) or interferon-null (Vero) cells at 31 and 37 ◦ C. Peak titres are shown.

easily detectable by both anti-EGFP and anti-CV IgG by 16–24 h post-transfection, significantly greater degradation was observed for the EGFP-M fusion than for the M-EGFP fusion protein (Fig. 4, panels a and b). Two breakdown products were observed, one of ∼30 kDa stained with anti-EGFP IgG (Fig. 4a), and one of ∼26 kDa stained with anti-CV IgG (Fig. 4b), suggesting that the EGFP-M fusion protein was being cleaved near to the junction between the EGFP and M polypeptides. Immunoblotting of rCVGM-infected cells showed that by 48 h post-infection, the majority of the EGFPM fusion protein had been cleaved to smaller products (Fig. 4c, lane 7). We were unable to detect any EGFP signal, either fused or monomeric, in virus particles of rCVGM or rCVE (Fig. 4c, lanes 8 and 9). Inspection of virus-infected cells for EGFP fluorescence showed a clear difference between cells infected with rCVE, which showed EGFP throughout the cell (Fig. 4h), and cells infected with rCVGM which showed punctate fluorescence in the cytoplasm, accompanied by a fainter diffuse staining (Fig. 4i). These patterns of

fluorescence resemble those described above for cells transfected with plasmids pT7EGFP and pT7EGFPM, respectively. In addition, the cytopathic effect induced by rCVGM was less severe than that induced by rCV and rCVE at the same time post-infection, compare Fig. 4f to d and e. Blebbing, presumably a sign of apoptosis, can be seen on the surface of cells infected with rCV and rCVE, but not those infected with rCVGM. In the case of rCVE, these surface blebs are seen to contain fluorescent material (Fig. 4h). 4. Discussion Recombinant viruses in which one protein is tagged with EGFP, constructed using reverse genetics techniques, have been used to study infection dynamics for the rhabdoviruses, rabies virus (Finke et al., 2004) and VSV (Das et al., 2006). In each case, the P protein was fused to EGFP, at the N-terminus or internally at the so-called hinge region, respectively. In this report, we have demonstrated a viable reverse genetics system for another rhabdovirus, CV, and

170

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

#

#

# Fig. 3. Matrix-fusion gene of rCVGM and mutants, shown schematically in mRNA-sense. Black indicates 5 and 3 untranslated regions, wider box represents ORF (pale grey: EGFP coding region, darker grey: M coding region). Dotted lines in mutants represent deletions. * Represents the position of the first AUG that could be used to initiate the M reading frame. Numbers above the bars refer to nucleotide numbers in the undeleted mRNA, while those below the bars indicate the residues flanking the deletions. The hatched box in deletion-633 shows the location of the first ORF, followed by the next downstream AUG at position 743.

used this system to generate a recombinant virus in which the M protein was tagged at its amino terminus with EGFP. Infectious virus was recovered directly from cDNA clones containing a copy of the antigenome under control of a T7 promoter, co-transfected into BSR-T7/5 cells along with plasmids which express the CV N, P and L proteins. Use of these cells avoids the necessity for a vaccinia virus which was used to supply T7 RNA polymerase in the original VSV reverse genetics protocols (Lawson et al., 1995; Whelan et al., 1995). As proof of principle, a virus rCVE was generated which stably expressed EGFP from an additional transcriptional unit positioned between the CV M and G genes. This virus showed very similar growth to the unmodified rCV (Fig. 2), although a single-step growth comparison showed approximately 3-fold lower titres reached by rCVE compared to rCV (Fig. 2e). This was unsurprising, as insertion of an additional transcriptional unit would be expected to reduce the expression of the downstream genes, G and L, due to the gradient of transcription observed in rhabdoviruses (Rose and Whitt, 2001), which could negatively impact the rate of virus growth. In this study the tag was attached to the M protein of CV, since this protein is known to be multifunctional in other negative-strand RNA viruses: as well as having a key role in mediating interaction of ribonucleoprotein complexes with glycoprotein at the cell surface and driving budding of virions (Jayakar et al., 2004), the M protein also has a key role in down-regulating host gene expression, inducing apoptosis and antagonising the interferon response (Kopecky et al., 2001; Kopecky and Lyles, 2003; Lichty et al., 2004; Stojdl et al., 2003; Taylor et al., 1999). The VSV M protein has been shown to retain its function in inhibiting host nuclear export when tagged with EGFP at either terminus (von Kobbe et al., 2000; Petersen et al., 2001). Furthermore, the CV EGFP-M fusion protein showed a greater inhibitory activity than VSV EGFP-M (Petersen et al., 2001). In the present study, CV M proteins tagged at either the amino terminus or the carboxyl terminus were initially expressed from plasmids in the absence of other virus proteins. Both CV EGFP-M and M-EGFP proteins showed reduced function in VLP formation using the minireplicon packaging assay (Fig. 1), but since EGFP-M retained the most activity that fusion protein was used to replace the M protein of CV in rCVGM. The virus was successfully rescued but was strongly attenuated for growth in cell culture compared to rCVE, indicating that the defect lay in the EGFP-M gene, rather than simply being due to the addition of extra EGFP sequence to the CV genome. The genomes of rCVE and rCVGM were very similar in size, 11,879 nt and 11,897 nt respectively, making each ∼7%

longer than the wild-type genome. The appearance of faster growing variants in which most or all of the EGFP sequence had been deleted suggested a strong selection pressure against retention of the intact EGFP-M gene. We have no evidence for the mechanism of deletion, but it is likely that mutants are generated by viral polymerase “skipping” in an analogous manner to the generation of deleted defective interfering RNA. Deletions resulting in an M protein which was fully functional in viral morphogenesis would then have a very strong growth advantage over the parental rCVGM. During the course of this research, a study was published showing that attempts to recover VSV in which the M protein was replaced by an RFP-M fusion protein were unsuccessful (Das et al., 2009). Furthermore, the authors suggested that neither RFP-M nor M-RFP fusion proteins were incorporated into virus particles. This agrees with our failure to detect EGFP-M in particles of rCVGM, although our study was hampered by our inability to grow rCVGM to high titres without generating deletion mutants. Both our observations and those of Das et al. (2009) may be explained by the recently published structural model for the VSV virion (Ge et al., 2010), which shows a regular lattice of M proteins interacting on the inner surface with N subunits of the RNP, and on the outer surface with the G protein tails protruding through the viral envelope. There is little space in this structure to accommodate an additional protein the size of EGFP or RFP, which are each roughly equal in size to M protein (26 kDa). The structure of the CV particle is presumably similar to that of VSV, based on the similarity of appearance in electron micrographs (Rao et al., 2004). It was shown that VSV M protein with a 12 residue C-terminal tag could be incorporated into the virus particle (Das et al., 2009), and sequence of deletion633 suggests that CV M protein with a 27 residue extension at the N-terminus is incorporated into the CV virus particle, although this has not been directly demonstrated. Since approximately Msized breakdown products were observed in cells transfected with pT7EGFPM or infected with rCVGM, it is tempting to speculate that a fraction of these breakdown products were able to participate in virus formation at the levels observed for rCVGM. It is likely that virus formation and/or budding were the major defects responsible for the attenuation of rCVGM, as both ourselves (ACM, unpublished data) and others (Petersen et al., 2001) have shown that CV EGFP-M retains its activity in host shut-off, and rCVGM did not show any enhancement of growth in interferon-negative Vero cells (Fig. 2g), whereas VSV mutants which are defective in host shut-off also show a strong preference for growth in interferon-negative cells (Stojdl et al., 2003).

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

171

Fig. 4. Expression of EGFP fusion proteins from plasmids and virus. (a and b) Time-course of expression from plasmids pT7EGFPM and pT7MEGFP, probed for (a) EGFP or (b) CV proteins. (a) Lanes 1–5, pT7EGFPM at 6, 16, 24, 30 and 48 h; lane 6, mock; lanes 7–11, pT7MEGFP at 6, 16, 24, 30 and 48 h; lane 12, pT7EGFP at 48 h. (b) Lanes 1–5, pT7EGFPM at 6, 16, 24, 30 and 48 h; lane 6, mock; lane 7, pT7M; lanes 8–12, pT7MEGFP at 6, 16, 24, 30 and 48 h; lane 13, mock; lane 14, pT7N + pT7M; lane 15, pT7N + pT7G. (c) Cell lysates (lanes 1–7) or virus (lanes 8–9) probed for EGFP. Lanes 1 and 5, mock; lane 2, pT7M; lane 3, pT7EGFP; lane 4, pT7EGFPM; lanes 6 and 8, rCVE; lanes 7 and 9, rCVGM. (d–i) BS-C-1 cells infected with (d and g) rCV; (e and h) rCVE; (f and i) rCVGM. Cells were fixed with 4% formaldehyde at 24 h post-infection and photographed under phase contrast (d–f) or UV illumination (g–i).

The punctate fluorescence pattern observed in transfected cells and in cells infected with rCVGM has not been previously reported. VSV M protein shows a fairly uniform cytoplasmic distribution, with some protein observed in the nucleus and a proportion at the cell membrane (Jayakar et al., 2004). Further research would be required to determine the significance of this observation. In light of the observed properties of rCVGM, it appears that it will not be possible to replace the CV M protein with an EGFP-tagged version in a stable manner. Interpretation of further experiments using rCVGM would be complicated by the presence of EGFP-M breakdown products and the appearance of deletion

mutants. It may be possible to grow the virus to high titres by supplying CV M protein, either in cis as an extra gene, or in trans from a plasmid, as has been done for rabies virus which expresses both N-EGFP and N proteins (Koser et al., 2004), but again this would complicate interpretation of experiments in which only the EGFPM could be observed. In summary, a new reverse genetics system for CV has been used to construct CV expressing either EGFP as an extra gene, or EGFPM in the place of the M gene. However, poor growth and genetic instability of the latter limit its usefulness in investigating the role of M protein in the CV life cycle. A more viable approach may be to

172

A.C. Marriott, C.A. Hornsey / Virus Research 160 (2011) 166–172

use a substantially smaller tag on the M protein, as has been used for VSV (Das et al., 2009). Acknowledgements We would like to thank Prof. C. Pringle, University of Warwick, for the anti-CV IgG, Prof. J. Dahlberg, University of Wisconsin, for the pEGFP-CV-M plasmid, and Prof A. Easton, University of Warwick, for his useful comments on the manuscript. A University of Warwick Research and Development Fund small grant supported CAH. References Basak, S., Mondal, A., Polley, S., Mukhopadhyay, S., Chattopadhyay, D., 2007. Reviewing Chandipura: a vesiculovirus in human epidemics. Biosci. Rep. 27, 275–298. Bhatt, P.N., Rodrigues, F., 1967. Chandipura: a new arbovirus isolated in India from patients with febrile illness. Ind. J. Med. Res. 55, 1295–1305. Buchholz, U.J., Finke, S., Conzelmann, K., 1999. Generation of bovine respiratory syncytial virus (BRSV) from cDNA: BRSV NS2 is not essential for virus replication in tissue culture, and the human RSV leader region acts as a functional BRSV genome promoter. J. Virol. 73, 251–259. Chadha, M.S., Arankalle, V.A., Jadi, R.S., Joshi, M.V., Thakare, J.P., Mahadev, P.V.M., Mishra, A.C., 2005. An outbreak of Chandipura virus encephalitis in the eastern districts of Gujarat state, India. Am. J. Trop. Med. Hyg. 73 (3), 566–570. Das, S.C., Nayak, D., Zhou, Y., Pattnaik, A.K., 2006. Visualisation of intracellular transport of vesicular stomatitis virus nucleocapsids in living cells. J. Virol. 80 (13), 6368–6377. Das, S.C., Panda, D., Nayak, D., Pattnaik, A.K., 2009. Biarsenical labeling of vesicular stomatitis virus encoding tetracysteine-tagged M protein allows dynamic imaging of M protein and virus uncoating in infected cells. J. Virol. 83 (6), 2611–2622. Duprex, W., Collins, F., Rima, B., 2002. Modulating the function of the measles virus RNA-dependent RNA polymerase by insertion of green fluorescent protein into the open reading frame. J. Virol. 76 (14), 7322–7328. Finke, S., Brzozka, K., Conzelmann, K., 2004. Tracking fluorescence-labeled rabies virus: enhanced green fluorescent protein-tagged phosphoprotein P supports virus gene expression and formation of infectious particles. J. Virol. 78 (22), 12333–12343. Ge, P., Tsao, J., Schein, S., Green, T., Luo, M., Zhou, Z., 2010. Cryo-EM model of the bullet-shaped vesicular stomatitis virus. Science 327, 689–693. Geevarghese, G., Arankalle, V., Jadi, R., Kanojia, P., Joshi, M., Mishra, A.C., 2005. Detection of Chandipura virus from sand flies in the genus Sergentomyia (Diptera: Phlebotomidae) at Karimnagar district, Andhra Pradesh, India. J. Med. Entomol. 42 (3), 495–496. Gurav, Y., Tandale, B., Jadi, R.S., Gunjikar, R., Tikute, S., Jamgaonkar, A., Khadse, R., Jalgaonkar, S., Arankalle, V., Mishra, A., 2010. Chandipura virus encephalitis outbreak among children in Nagpur division, Maharashtra, 2007. Indian J. Med. Res. 132, 395–399. Ilkal, M.A., Goverdhan, M.K., Shetty, P.S., Tupe, C.D., Mavale, M.S., Dhanda, V., 1991. Susceptibility of four species of mosquitoes to Chandipura virus and its detection by immunofluorescence. Acta Virol. 35, 27–32. Jadi, R.S., Sudeep, A.B., Kumar, S., Arankelle, V.A., Mishra, A.C., 2010. Chandipura virus growth kinetics in vertebrate cell lines, insect cell lines and embryonated eggs. Indian J. Med. Res. 132, 155–159. Jayakar, H.R., Jeetendra, E., Whitt, M.A., 2004. Rhabdovirus assembly and budding. Virus Res. 106, 117–132. Joshi, M., Patil, D., Tupe, C., Umarani, U., Ayachit, V., Geevarghese, G., Mishra, A., 2005. Incidence of neutralizing antibodies to Chandipura virus in domestic animals

from Karimnagar and Warangal Districts of Andhra Pradesh, India. Acta Virol. 49, 69–71. Klingen, Y., Conzelmann, K., Finke, S., 2008. Double-labeled rabies virus: live tracking of enveloped virus transport. J. Virol. 82 (1), 237–245. Kopecky, S., Lyles, D., 2003. The cell-rounding activity of the vesicular stomatitis virus matrix protein is due to the induction of cell death. J. Virol. 77 (9), 5524–5528. Kopecky, S., Willingham, M., Lyles, D., 2001. Matrix protein and another component contribute to induction of apoptosis in cells infected with vesicular stomatitis virus. J. Virol. 75, 12169–12181. Koser, M., McGettigan, J., Tan, G., Smith, M., Koprowski, H., Dietzschold, B., Schnell, M., 2004. Rabies virus nucleoprotein as a carrier for foreign antigens. Proc. Natl. Acad. Sci. U.S.A. 101 (25), 9405–9410. Lawson, N.D., Stillman, E.A., Whitt, M.A., Rose, J.K., 1995. Recombinant vesicular stomatitis viruses from DNA. Proc. Natl. Acad. Sci. U.S.A. 92, 4477–4481. Lichty, B.D., Power, A.T., Stojdl, D.F., Bell, J.C., 2004. Vesicular stomatitis virus: reinventing the bullet. Trends Mol. Med. 10 (5), 210–216. Marriott, A.C., 2005. Complete genome sequences of Chandipura and Isfahan vesiculoviruses. Arch. Virol. 150 (4), 671–680. Mavale, M., Fulmali, P., Ghodke, Y., Mishra, A.C., Kanojia, P., Geevarghese, G., 2007. Experimental transmission of Chandipura virus by Phlebotomus argentipes (Diptera: Psychodidae). Am. J. Trop. Med. Hyg. 76 (2), 307–309. Moss, B., Elroy-Stein, O., Mizukami, T., Alexander, W., Fuerst, T., 1990. New mammalian expression vectors. Nature 348 (6296), 91–92. Petersen, J.M., Her, L., Dahlberg, J.E., 2001. Multiple vesiculoviral matrix proteins inhibit both nuclear export and import. Proc. Natl. Acad. Sci. U.S.A. 98 (15), 8590–8595. Rao, B., Basu, A., Wairagkar, N., Gore, M., Arankalle, V., Thakare, J., Jadi, R., Rao, K., Mishra, A., 2004. A large outbreak of acute encephalitis with high fatality rate in children in Andhra Pradesh, India, in 2003, associated with Chandipura virus. Lancet 364, 869–874. Rao, S.N., Wairagkar, N.S., Mohan, V.M., Khetan, M., Somarathi, S., 2007. Brainstem encephalitis associated with Chandipura in Andhra Pradesh outbreak. J. Trop. Pediatr. 54 (1), 25–30. Rose, J.K., Whitt, M.A., 2001. Rhabdoviridae: the viruses and their replication. In: Fields Virology. Lippincott Williams and Wilkins, Philadelphia. Ruedas, J.B., Perrault, J., 2009. Insertion of enhanced green fluorescent protein in a hinge region of vesicular stomatitis virus L polymerase protein creates a temperature-sensitive virus that displays no virion-associated polymerase activity in vitro. J. Virol. 83 (23), 12241–12252. Stojdl, D.F., Lichty, B.D., tenOever, B.R., Paterson, J.M., Power, A.T., Knowles, S., Marius, R., Reynard, J., Poliquin, L., Atkins, H., Brown, E.G., Durbin, R.K., Durbin, J.E., Hiscott, J., Bell, J.C., 2003. VSV strains with defects in their ability to shutdown innate immunity are potent systemic anti-cancer agents. Cancer Cell 4 (4), 263–275. Tandale, B., Tikute, S., Arankalle, V., Sathe, P., Joshi, M., Ranadive, S., Kanojia, P., Eshwarachary, D., Kumarswamy, M., Mishra, A., 2008. Chandipura virus: a major cause of acute encephalitis in children in North Telangana, Andhra Pradesh, India. J. Med. Virol. 80, 118–124. Taylor, A., Easton, A.J., Marriott, A.C., 1999. Matrix protein of Chandipura virus inhibits transcription from an RNA polymerase II promoter. Virus Genes 19, 223–228. Tesh, R.B., Travassos da Rosa, A.P.A., Travassos da Rosa, J.S., 1983. Antigenic relationship among rhabdoviruses infecting terrestrial vertebrates. J. Gen. Virol. 64, 169–176. von Kobbe, C., van Deursen, J., Rodrigues, J., Sitterlin, D., Bachi, A., Wu, X., Wilm, M., Carmo-Fonseca, M., Izaurralde, E., 2000. Vesicular stomatitis virus matrix protein inhibits host cell gene expression by targeting the nucleoporin Nup98. Mol. Cell 6, 1243–1252. Whelan, S.P.J., Ball, L.A., Barr, J.N., Wertz, G.T.W., 1995. Efficient recovery of infectious vesicular stomatitis virus entirely from cDNA clones. Proc. Natl. Acad. Sci. U.S.A. 92, 8388–8392. Wilks, C., House, J., 1986. Susceptibility of various animals to the vesiculoviruses Isfahan and Chandipura. J. Hyg. Camb. 97, 359–368.