|Home | About | Journals | Submit | Contact Us | Français|
Borna disease virus (BDV), the prototypic member of the Bornaviridae family, within the order Mononegavirales, is highly neurotropic and constitutes an important model system for the study of viral persistence in the central nervous system (CNS) and associated disorders. The virus surface glycoprotein (G) has been shown to direct BDV cell entry via receptor-mediated endocytosis, but the mechanisms governing cell tropism and propagation of BDV within the CNS are unknown. We developed a small interfering RNA (siRNA)-based screening to identify cellular genes and pathways that specifically contribute to BDV G-mediated cell entry. Our screen relied on silencing-mediated increased survival of cells infected with rVSVΔG*/BDVG, a cytolytic recombinant vesicular stomatitis virus expressing BDV G that mimics the cell tropism and entry pathway of bona fide BDV. We identified 24 cellular genes involved in BDV G-mediated cell entry. Identified genes are known to participate in a broad range of distinct cellular functions, revealing a complex process associated with BDV cell entry. The siRNA-based screening strategy we have developed should be applicable to identify cellular genes contributing to cell entry mediated by surface G proteins of other viruses.
Borna disease virus (BDV) is an enveloped virus with a nonsegmented negative-strand (NNS) RNA genome whose organization (3′-N-p10/P-M-G-l-5′) is characteristic of mononegaviruses (MNV) (12, 49, 51). However, based on its unique genetics and biological features, BDV is considered to be the prototypic member of a new virus family, Bornaviridae, within the order Mononegavirales (13, 49). BDV is highly neurotropic and causes central nervous system (CNS) disease in a variety of vertebrate species that is frequently manifested by behavioral abnormalities (25, 53). Both host and viral factors contribute to a variable period of incubation and heterogeneity in the symptoms and pathology associated with BDV infection (18, 20, 32, 44, 46). BDV provides an important model for the investigation of both immune-mediated pathological events associated with virus-induced neurological disease and mechanisms whereby viruses induce neurodevelopmental and behavioral disturbances in the absence of the hallmarks of cytolysis and inflammation (19, 24, 42).
Serological data and molecular epidemiological studies suggest that BDV, or a BDV-like virus, can infect humans, and it might be associated with certain neuropsychiatric disorders (23, 29). Moreover, the recent identification of an avian bornavirus (22, 45) underscores the prospect of finding other BDV-like emerging viruses of potential relevance to human health. The identification of host cell factors involved in BDV cell entry will contribute to a better understanding of BDV cell tropism and propagation within the CNS, a first and necessary step to elucidate the mechanisms by which BDV disrupts normal brain functions. Likewise, the development of effective antiviral strategies to combat bornavirus infections may be facilitated by targeting host cell factors that participate in BDV cell entry.
As with other MNV, the BDV surface glycoprotein (G) plays a key role in virus cell entry, which uses a receptor-mediated endocytosis pathway (3, 18, 40). BDV G gene directs the synthesis of a precursor glycoprotein (GPC) with a predicted molecular mass of 56 kDa but, due to extensive glycosylation, GPC migrates with a molecular mass of 84 to 94 kDa. GPC is posttranslationally cleaved by the cellular protease furin into GP1 (GPN) and GP2 (GPC) that correspond to the N- and C-terminal regions, respectively, of GPC (17, 43). GP1 is sufficient for virus receptor recognition and cell entry (40), whereas GP2 has been implicated in the pH-dependent fusion event between viral and endosome membranes (17). This fusion event delivers the virus ribonucleoprotein (RNP) core to the cytoplasm, which is followed by its import into the nucleus, where both RNA replication and transcription take place (12). In addition to cell surface molecules that act as viral receptors and coreceptors, many other host cell proteins can influence virus cell entry (11, 21, 35). The emergence of RNA interference (RNAi) as a tool for gene expression inhibition has enabled a forward genetic approach to facilitate the identification of host cell factors contributing to different steps of virus multiplication. This approach has been successfully used to identify components of the kinome involved in VSV cell entry (38), as well as host proteins required for virus infection, including HIV-1 (6, 36) and West Nile virus (WNV) (28). The use of RNAi-based genetics approaches to study BDV-host cell interactions involved in virus cell entry are complicated by the inability to grow cell-free infectious BDV to high titers and the absence of cytopathic effect (CPE) associated with BDV infection. To overcome these limitations we took advantage of a replication-competent recombinant vesicular stomatitis virus (VSV) expressing the BDV G (rVSVΔG*/BDVG), which was shown to recreate the cell tropism and entry pathway of bona fide BDV (39), but retained the cytolytic phenotype of VSV and produced high titers (107 PFU/ml) of cell-free infectious progeny. The use of rVSVΔG*/BDVG allowed us to develop a small interfering RNA (siRNA)-based high-throughput screen (HTS) to identify cellular genes required for BDV G-mediated cell entry. We identified 24 host cell genes involved in BDV G-mediated cell entry. Identified genes are expressed in different subcellular locations and known to be involved in a wide range of different cellular physiological processes. These findings indicate that cell entry of BDV involves a complex array of interactions between the virus G and cellular proteins. This study provides also a novel and general approach to identifying host cellular genes involved in cell entry mediated by the surface glycoproteins of other viruses for which the corresponding rVSVΔG*/G could be generated.
Ol, an established human oligodendroglial cell line, was maintained in Dulbecco modified Eagle medium (DMEM) supplemented with 2 mM glutamine, 10 mM HEPES, and 10% heat-inactivated fetal bovine serum (FBS). The recombinant VSVs rVSVΔG*/BDVG and rVSVΔG*/VSVG have been described (39). The BDV, VSV, and lymphocytic choriomeningitis virus (LCMV) strains utilized were He80 (50), Indiana (26), and Armstrong (14), respectively.
The screen was performed essentially as described previously (2). Ambion Silencer Human Druggable Genome siRNA Library V2 was delivered into Ol cells by reverse transfection by mixing 1 μl of individual siRNA (0.5 μM) and 10 μl of diluted (1:100 in Optimem) siRNA MAX (Invitrogen) per well of a 384-well plate, followed by incubation for 15 min at room temperature. Ol cells (4,000 per well) diluted in 40 μl of DMEM supplemented with 10% FBS were added to the transfection mix and incubated for 36 h at 37°C 5% CO2, followed by the addition of 10 μl of medium containing 2.5 × 104 PFU of either rVSVΔG*/BDVG or rVSVΔG*/VSVG. Infected cells were incubated at 37°C for 48 h for rVSVΔG*/BDVG or 24 h in case of rVSVΔG*/VSVG, and cell viability was determined by addition of 10 μl of ATPlite (Perkin-Elmer) and measuring the levels of ATP by luminometry using an Envision 2103 Multilabel plate reader (Perkin-Elmer).
Ol cells were reverse transfected with different siRNA, and 36 h later the total cellular RNA was isolated by using TRI Reagent (Sigma Aldrich). Equal amounts of RNA were DNase treated and subjected to RT using random hexamers (Invitrogen) and SuperScript III reverse transcriptase (Invitrogen). Samples were treated with RNase and cDNAs amplified by PCR using the indicated primers (Table (Table1).1). PCR was done using Taq polymerase (5 Prime, Germany) and the following conditions: 94°C for 5 min; 30 cycles of 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min; and 72°C for 10 min and 4°C for infinity. Equal volumes of PCRs were subjected to 2% agarose gel electrophoresis analysis, and PCR products were visualized by ethidium bromide staining.
Ol cells were pretreated for 1 h prior infection with the indicated concentrations of the following inhibitors: the furin inhibitor hexa-d-arginine (Calbiochem), the ADAM17-selective inhibitor SP26 (Schering Plough), and the cathepsin L inhibitor Z-Phe-Tyr(tBu)-diazomethylketone (CatL; Axxora LLC). The cells were infected with 200 PFU of either rVSVΔG*/BDVG or rVSVΔG*/VSVG, BDV (multiplicity of infection [MOI] = 0.1), or LCMV (MOI = 0.01) in the presence of the drugs. Infected cells were fixed (4% paraformaldehyde in phosphate-buffered saline) at 7 h postinfection (p.i.) (infections with rVSV) or 48 h p.i. (infections with BDV and LCMV), permeabilized (0.1% saponin), and examined by immunofluorescence assay (IFA). Cells infected with rVSV were identified by using a mouse monoclonal antibody to VSVN (10G4; Ab-VSV N). Cells infected with LCMV or BDV were identified by using mouse monoclonal antibody 1.1.3 to LCMV NP or a rabbit antiserum to BDV N, respectively.
The viability of siRNA-transfected and rVSVG*/BDVG-infected cell populations was normalized to positive and negative controls on each plate during the primary screen. The average score of duplicates for each siRNA was treated as the siRNA score. A “survival score” was determined based on the median score of the six wells with siRNA targeting each gene. Quantile-quantile plots of each siRNA score and each “survival score” against normal distributions of the same size show a good fit (R = 0.992). Genes with two or more siRNAs inducing 2-fold activation or greater were considered hits. P values were calculated for each gene by using a Student t test on all six well scores per gene versus scores for all nonspecific control wells.
To identify candidate host cell genes involved in BDV entry, we designed a siRNA-based HTS that used as readout the survival of Ol cells after infection with rVSVΔG*/BDVG. We selected the Ol cell line for this HTS because of its high susceptibility to BDV infection (7, 37) and great siRNA-mediated silencing efficiency in the absence of significant toxicity (Fig. (Fig.1A).1A). Moreover, Ol cells transfected with control siRNAs remained highly susceptible to both VSV and rVSVΔG*/BDVG, as determined by cell survival measurements using an ATPlite assay (41). This assay had an excellent dynamic range when mock and virus infection conditions were compared (Fig. (Fig.1B1B).
The siRNA-based HTS consisted of three distinct steps (Fig. (Fig.2).2). The first step involved a reverse transfection protocol to deliver the Ambion Silencer Human Druggable Genome siRNA Library V2 into Ol cells, followed 36 h later by infection of transfected cells with rVSVΔG*/BDVG. Initial hits were selected based on increased cell resistance to CPE associated with rVSVΔG*/BDVG infection. In a second step we compared the susceptibility to rVSVΔG*/BDVG and rVSVΔG*/VSVG of cells transfected with hits selected in the primary screen of the screen, which allowed us to identify siRNAs that specifically interfered with BDV G-mediated cell entry. The third and final step involved the validation of selected hits in the context of infection with bona fide BDV.
The siRNA library used in our screening targeted 5,516 different genes, with each gene being covered by three independent siRNAs. Ol cells were reverse transfected in duplicate with each siRNA, and 36 h later the cells were infected with rVSVΔG*/BDVG (MOI = 5). To assess the inhibitory effect of transfected siRNAs on rVSVΔG*/BDVG infection, we measured cell viability at 48 h p.i. using a luminometric assay based on the determination of cellular concentrations of ATP (41). The data collected were inspected for plate patterns and other artifacts for quality assessment, and suspected wells were eliminated prior to subsequent statistical analysis (Table (Table22).
Curated data were normalized, and duplicate measurements for each siRNA were averaged to a single value to be used as siRNA score. Using this information, we calculated P values for each siRNA as described in Materials and Methods. We next calculated the median of the scores of the three siRNAs directed against each gene target (“survival score”), and ranked genes represented in the library according to this value. In this way high medians (and low P values) would correspond to genes whose knockdown via siRNA led to increased cell survival upon virus infection. The distribution of survival scores fitted well a normal distribution as assessed by quantile-quantile plot analysis (not shown). The histogram representing the distribution of survival scores calculated for each gene (Fig. (Fig.3)3) was used to identify genes with two or more siRNAs inducing twofold activation or greater. Subsequent P value calculations were made under the assumption of a normal distribution (Table (Table33).
Since our assay was based on the cytolytic effect of rVSVΔG*/BDVG, high scoring targets were likely to represent cellular genes necessary for either viral infection or viral induced cell death. Accordingly, several top ranked targets were well known proapoptotic genes likely involved in the cytopathic activity of VSV (data not shown), and were no longer pursued. A caveat associated with siRNA-based screens relates to the frequency of false positives resulting from off-target effects. This concern is minimized when different siRNAs targeting the same mRNA species produce the same phenotype. Consequently, for further analysis we selected only candidate genes that had a score higher than two and at least two siRNA with a P value of <0.05. Based on these criteria we selected 123 distinct siRNAs (P ≤ 0.05) representing 58 different candidate gene targets (Table (Table33).
We next examined which of the 123 siRNA hits selected by the primary screen inhibited consistently and specifically infection with rVSVΔG*/BDVG. We first assessed the effect of the transfected gene-specific siRNAs on cell viability. For this, we compared the magnitudes of the ATP signals generated by cells transfected with each one of the 123 gene-specific siRNA candidates and those obtained from cells transfected with each of the 48 different nonspecific (NS) control siRNAs. We used the data from this comparison to normalize readouts from cells that were siRNA transfected and subsequently infected with either rVSVΔG*/BDVG or rVSVΔG*/VSVG.
We next compared the 123 gene-specific selected siRNAs with NS control siRNAs with respect to their ability to increase cell survival upon infection with rVSVΔG*/BDVG. The data for this analysis were obtained from four independent replicates for each siRNA. This allowed us to select 47 siRNAs, among the 123 gene-specific siRNAs, that consistently inhibited infection with rVSVΔG*/BDVG. To determine which of these selected 47 siRNAs exhibited their effects by specifically targeting BDV G-mediated cell entry, we compared their effects on cell survival upon infection with either rVSVΔG*/BDVG or rVSVΔG*/VSVG (Fig. (Fig.4).4). This approach led us to classify the 58 initially identified candidate genes into three groups. The first two groups included (i) genes (n = 21) whose siRNA-mediated knockdown resulted in specific increased resistance to rVSVΔG*/BDVG compared to rVSVΔG*/VSVG infection and (ii) genes (n = 15) whose siRNA-mediated knockdown conferred similar levels of resistance to both rVSVΔG*/BDVG and rVSVΔG*/VSVG. Genes belonging to this latter group likely represented genes that played a role in any of the multiple steps involved in RNA replication or gene expression of VSV. However, it should be noted that some genes within this group might play a common role in entry mediated by the G protein of both BDV and VSV. (iii) The third group included genes (n = 22) whose knockdown by any of the gene-specific siRNA selected for this analysis did not result in a statistically significant increase in cell survival upon infection with rVSVΔG*/BDVG compared to cells transfected with NS siRNA prior infection with rVSVΔG*/BDVG. Candidate genes within this group were considered to be false positives and thus were not examined further.
To assess the biological significance of our siRNA-based HTS results, we examined whether siRNA-mediated knockdown of genes that resulted in specific inhibition of rVSVΔG*/BDVG infection, resulted also in inhibition of infection with bona fide BDV. For this, we conducted six independent experiments where Ol cells were transfected with each of the 28 siRNAs identified as capable of inhibiting specifically BDV G-mediated cell entry (Fig. (Fig.4)4) and 36 h later infected them with BDV (MOI = 2). At 48 h p.i., we determined the number of BDV-infected cells by IFA using a rabbit serum to BDV N. As a control, Ol cells transfected with nonspecific siRNA were also infected with BDV. All but two 28 siRNA assayed inhibited to different degrees BDV infection (Fig. (Fig.5).5). Statistical analysis based on P values obtained for each siRNA indicated that of the 21 initially selected candidate genes, 15 could be considered bona fide host cell factors specifically required for BDV G-mediated cell entry (Table (Table4).4). We used the same approach to test the set of 18 siRNAs, representing 15 different genes, that inhibited the infection of both rVSVΔG*/BDVG and rVSVΔG*/VSVG (Fig. (Fig.4).4). The results from this analysis identified nine additional candidate genes involved in BDV cell entry (Fig. (Fig.66).
To assess further the specificity of our findings and rule out possible nonspecific off target effects, we examined the effects of additional distinct specific siRNAs targeting a subset of five candidate genes among the 24 validated as contributing to BDV G-mediated cell entry. Cells transfected with these additional siRNAs exhibited levels of increased resistance to infection with rVSVΔG*/BDVG similar to that provided by the corresponding gene-specific siRNA used during the primary screen (compared Fig. Fig.44 and Fig. Fig.7A).7A). We used RT-PCR assays (Fig. (Fig.7B)7B) to confirm the expression in Ol cells of all candidate genes that our HTS identified as involved in BDV G-mediated entry. Likewise, we used a selected group of candidate genes to confirm that transfection with gene-specific siRNAs resulted in decreased levels of the corresponding gene mRNA (Fig. (Fig.7C7C).
Our siRNA-based HTS identified furin, as well as metalloprotease ADAM17, and the lysosomal cysteine protease cathepsin L as host cell factors contributing to BDV cell entry.
We have documented that furin plays a key role in the maturation of BDV G and the production of infectious progeny (3, 9, 17, 43). Therefore, our finding that furin was also playing a role in BDV cell entry was rather intriguing. To further validate this observation, we examined the effect of pharmacological inhibition of furin on BDV G-mediated cell entry. For this analysis, we treated Ol cells with increasing concentrations of the furin inhibitor hexa-d-arginine (H-d-R) starting 1 h prior to infection (MOI = 0.1) with rVSVΔG*/BDVG, bona fide BDV, or (as a control) rVSVΔG*/VSVG. Treatment with the furin inhibitor exerted a dose-dependent inhibition of infection with rVSVΔG*/BDVG, but not with rVSVΔG*/VSVG (Fig. (Fig.8A).8A). Treatment with H-d-R also inhibited infection with bona fide BDV, whereas infection with the prototypic arenavirus LCMV was not affected (Fig. (Fig.8B).8B). We used LCMV, instead of VSV, as a control for this experiment because BDV antigen in infected cells can be readily detected only after 36 to 48 h p.i., a time at which infection with VSV had caused a massive CPE that impeded a quantitative assessment of a potential drug effect on virus multiplication. When treatment with H-d-R was initiated at 1 h p.i., we did not observe a significant reduction in the numbers of BDV-infected cells at 48 h p.i. (data not shown).
We also sought to gain further evidence for a role of ADAM17 and cathepsin L in cell entry of BDV by examining the effects of pharmacological inhibition of these proteases. For this, we used SP26 and Z-Phe-Tyr(tBu)-diazomethylketone (CatL), which are well-established inhibitors of ADAM17 (30) and cathepsin L (47), respectively. Treatment with both SP26 and CatL resulted in a drug dose-dependent significant reduction of the numbers of BDV-infected but not LCMV-infected cells (Fig. (Fig.99).
We have reported that BDV enters cells via receptor-mediated endocytosis (17). This process requires the initial participation of lipid rafts (10) and is completed in the endosomal compartment of the cell via a pH-dependent fusion event between viral and cellular membranes required to liberate the virus RNP core into the cell environment (17). As with other MNVs, the G surface glycoprotein of BDV mediates receptor recognition and directs the pH-dependent fusion event necessary for a productive infection (17, 40). However, many of the specific host cell factors and pathways involved in the BDV cell entry process remain unknown. To fill this gap of knowledge about cell entry of BDV, we utilized an siRNA-based screen aimed at identifying host cellular genes whose expression is required for virus cell entry mediated by BDV G.
BDV is characterized by its noncytolytic multiplication in all cell types thus far examined, which together with the inability to produce high titers of cell-free infectious either BDV or viral particles pseudotyped with BDV G, posed significant obstacles for the development of an HTS to identify host cell factors involved in BDV entry. We overcame this problem by using a previously described recombinant VSV (rVSVΔG*/BDVG) wherein VSVG was replaced by BDV G. rVSVΔG*/BDVG recreated the cell tropism and entry pathway of bona fide BDV (39) but retained the strong cytolytic phenotype characteristically associated with VSV infection, which facilitated the use of cell survival as a readout of the assay.
Our screen identified 24 candidate genes whose silencing resulted in the inhibition of BDV infection. It should be noted that several reasons precluded our screen to identify all possible candidate cellular genes that might contribute to BDV cell entry. Three key limitations associated with our siRNA-based HTS were as follows: (i) the library utilized for this HTS did not cover all of the transcriptome; (ii) in some cases, the siRNAs present in our library may not efficiently knock down their respective mRNA targets; and (iii) there was a potentially long half-life stability for polypeptides encoded by mRNAs targeted via siRNA.
Previous reports have documented the use of siRNA-based functional genetic screens to examine the role of host cellular genes during infection with a variety of viruses, including HIV (6, 36), RSV (27), and WNV (28). However, a unique feature of our HTS design was that it allowed us to identify candidate genes that specifically contribute to BDV G-mediated cell entry rather than to any of the multiple steps involved in virus RNA replication and gene expression. This was achieved by comparing the susceptibility of siRNA-transfected cells to infection with either rVSVΔG*/BDVG or rVSVΔG*/VSVG. Likewise, this approach identified host candidate genes involved in cell entry steps common to both VSV and BDV. Among the 58 candidate genes identified by the initial screen, 24 of them were validated as host cell factors required for infection with bona fide BDV. Of these 24 genes, 15 corresponded to cellular factors that contributed specifically to BDV G-mediated cell entry, whereas the other 9 appeared to contribute to cell entry or gene expression processes, or both, that were common to both VSV and BDV.
The host cell proteins identified by our HTS as required for BDV G-mediated cell entry are known to participate in a broad range of cellular functions and to display different subcellular localization patterns. Cellular receptors for BDV remain to be identified, and any candidate gene uncovered by our HTS that encodes a protein expressed at the cell surface could be considered a potential receptor, coreceptor, or attachment factor for BDV. Some of the targets identified are predominantly CNS molecules, including the subunit alpha-3 of the GABAA receptor (GABRA3) or the serotonin receptor 1F (HTR1F), but whether these molecules contribute to the neurotropism or neuropathogenesis, or both, of BDV remains to be determined. Cortical shrinkage associated with neonatal BDV infection is associated with a decreaseγ in parvalbumin-positive large neurons (a diameter of >100 μm), suggesting that GABA-ergic neurons may be particularly susceptible to neonatal BDV infection (32, 42).
Several cell surface proteins identified by our HTS as host factors involved in BDVG-mediated entry, including ADAM17 (55), endothelin receptor B (EDNRB) (15), and GABA receptors bearing the subunit GABRA3 (31), have been shown to be located in lipid rafts, a finding potentially related with our observation that lipid rafts are involved in BDV cell entry (10).
Our HTS identified several cellular proteases as host factors contributing to BDV cell entry. One of these proteases, furin, has been shown to play a critical role in BDV G maturation and production of infectious BDV progeny (3, 9, 43), as well as in the processing of several other viral glycoproteins (4). Accordingly, furin inhibitors have been shown to block infection by influenza virus (52) and HIV (36). However, before this work there was no evidence for a role of furin in BDV cell entry, which might be related to the role played by furin in the activation of ADAM17 and MMP21 (33), two different metalloproteases identified in our HTS as host factors involved in BDV cell entry. Notably, ADAM17 is located in lipid rafts (55) and may be part of the selected group of proteins recruited in this compartment to facilitate efficient BDV cell entry. Pharmacological inhibition of ADAM17 resulted in an ~60% reduction in the numbers of BDV-infected cells at 48 h p.i. In contrast, the numbers of LCMV-infected were minimally (<15% reduction) affected by treatment with SP26 and only at the highest drug concentration used, a finding which may indicate nonspecific effects on virus growth.
Our HTS identified also cathepsin L2, also known as cathepsin V, as a host cell factor involved in BDV cell entry. Notably, Ebola virus G protein requires the participation of two different endosomal cathepsins to generate the G conformation competent in fusion (8). Whether BDV requires a similar additional processing remains to be determined. It is, however, equally plausible that cathepsin L2-mediated processing of a host cell protein is required for cell entry of BDV. Cathepsin L2 was reported to be expressed at high levels in the thymus and testis, whereas CNS appears to express very low levels (48). However, RT-PCR of Ol cells readily detected the presence of cathepsin L2 mRNA in these cells (Fig. (Fig.77).
Our HTS identified also LDLR and DNCL2B (Table (Table2)2) as cellular factors required for BDV cell entry. LDLR has been shown to function as a receptor for rhinoviruses (34, 56) and potentially also HCV (16); whether LDLR can serve a similar role for BDV remains to be investigated. In addition, LDLR can trigger the clathrin-mediated endocytotic pathway (5), which we have found to direct the entry of BDV into cells (submitted for publication). We have also obtained evidence that microtubules play a key role in the trafficking of BDV-containing endosomes to the location where viral and cell membranes fuse (1, 54), a step required for the release of the viral RNP into the cell environment. This finding is consistent with the inhibition of BDV cell entry observed in cells subjected to siRNA-mediated knockdown of DNCL2B, a member of the complex dynactin that interacts with endosomes and shuttle these compartments through the cytoplasm using the microtubule network (21).
Further studies aimed at elucidating the roles of each of the BDV cell entry factors identified by our siRNA-based HTS would provide us with a detailed picture of this first and critical step in the virus life cycle. Moreover, it is worth noting that the design of our screen could be applied to identifying host cellular genes contributing to cell entry mediated by the surface glycoprotein of other viruses for which the corresponding rVSVΔG*/G can be generated.
We thank Apple Cortes and Eigo Suyama for advice and support in the screening and G. Sobko for statistics support.
This study was supported by a fellowship from the Ministerio de Educacion y Ciencia of Spain to R.C. and NIH grant R21 AI064820 to J.C.D.L.T.
This is publication 19493 from MIND.
Published ahead of print on 13 January 2010.