Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Virology. Author manuscript; available in PMC 2010 May 10.
Published in final edited form as:
PMCID: PMC2674145

The Use of Random Homozygous Gene Perturbation to Identify Novel Host-Oriented Targets for Influenza


Conventional approaches for therapeutic targeting of viral pathogens have consistently faced obstacles arising from the development of resistant strains and a lack of broad-spectrum application. Influenza represents a particularly problematic therapeutic challenge since high viral mutation rates have often confounded many conventional antivirals. Newly emerging or engineered strains of influenza represent an even greater threat as typified by recent interest in avian subtypes of influenza. Based on the limitations associated with targeting virally-encoded molecules, we have taken an orthogonal approach of targeting host pathways in a manner that prevents viral propagation or spares the host from virus-mediated pathogenicity. To this end, we report herein the application of an improved technology for target discovery, Random Homozygous Gene Perturbation (RHGP), to identify host-oriented targets that are well-tolerated in normal cells but that prevent influenza-mediated killing of host cells. Improvements in RHGP facilitated a thorough screening of the entire genome, both for overexpression or loss of expression, to identify targets that render host cells resistant to influenza infection. We identify a set of host-oriented targets that prevent influenza killing of host cells and validate these targets using multiple approaches. These studies provide further support for a new paradigm to combat viral disease and demonstrate the power of RHGP to identify novel targets and mechanisms.


Infection with influenza virus causes a highly contagious disease of the respiratory tract. In an average year, influenza infects millions worldwide and, in the United States, is responsible for hundreds of thousands of hospitalizations and 20,000–30,000 deaths (Thompson et al., 2003). The primary prevention modality for influenza consists of a vaccine against strains of influenza that are predicted to infect the population in the coming season and the primary antibody response of these vaccines is directed towards variability in the hemagglutinin (HA) and neuraminidase (NA) antigens. Since influenza virus can change within a single season by the processes of antigenic drift and shift, the efficacy of annual vaccination can be circumvented. Thus, the value of the annual vaccine fluctuates depending on the similarity of the vaccine strain and the circulating influenza strain. In a single-center analysis by the CDC, the estimated clinical effectiveness of the influenza vaccine during the 2007–2008 season was limited to 44% of those vaccinated (Belongia et al., 2008). Compounding this, recent surveys indicate that less than one third of American adults opt for seasonal immunization. Consequently, influenza is anticipated to remain a prominent and persistent cause of morbidity and mortality.

Once an individual is infected with influenza, therapeutic options are limited to supportive care or a handful of medications that target viral pathways, primarily neuraminidase (Luscher-Mattli, 2000; Ong and Hayden, 2007). However, the wide application of traditional antiviral medications has favored the emergence of drug-resistant viruses, thereby presenting a major new challenge to the control of infectious diseases. Compounding the increasing problems with seasonal influenza, much recent attention has focused on emerging pandemic strains of influenza and the potential that genetically-modified influenza could provide an effective tool for bioterrorism. The public health and economic impacts of such drug resistant or non-traditional influenza variants could be devastating, thus spurring the need for new paradigms to target influenza.

One new approach for addressing viral infection is to target host factors that are essential to the pathogenesis of viral disease (Fox, 2007; Tan et al., 2007). A prominent example from the war on HIV/AIDS is the development of antagonists of CCR5 (Reeves and Piefer, 2005; Wheeler et al., 2007). These antagonists seek to prevent the function of a surface-exposed co-receptor that is necessary for the HIV binding and internalization. Early findings suggest these agents may provide much-needed options for the treatment of the subset of patients infected with CCR5-trophic HIV-1 viruses.

The viral life cycle is governed by a dynamic interplay among viral and host factor proteins (Lamb and Takeda, 2001). Productive infection of the influenza virus requires the cooperation of host proteins for virtually every step of the life cycle, including entry and internalization, uncoating of nucleic acid, genome replication, protein translation, transportation, and processing, as well as virus assembly and late budding (Ahlquist et al., 2003; Chen and Lamb, 2008; Ludwig, Pleschka, and Wolff, 1999). Based on these findings, our laboratory and others have begun to consider host targets for antiviral therapy. Recent studies utilized small interference RNA (siRNA) approach to identify host factors involved in the early viral infection with influenza (Hao et al., 2008), HIV (Brass et al., 2008; Konig et al., 2008) or West Nile virus (Krishnan et al., 2008). Although intriguing, siRNA is intrinsically limited to known sequences and efficient siRNA delivery into the host cells, which often limits the efficiency of these approaches.

Our laboratories have been pioneering a new approach to sample any gene in a eukaryotic cell that can control cellular processes of interest to the investigator. Originally known as Random Homozygous Knockout (RHKO), the initial approach was developed to overcome barriers arising from the fact that eukaryotic cells generally have two sets of chromosomes (Li and Cohen, 1996). For example, the diploid nature of mammalian cells had precluded simple knock-out based evaluation of target genes that have proven to be so powerful in our understanding of bacterial pathways. The improvements provided by RHGP include the ability to simultaneously knockdown both copies of any target gene, independent of any prior knowledge or annotation of that gene. Herein, we report on the application of an improved form of RHGP, which can also cause up-regulation of any target in the cell, including whole genes or individual domains. Consequently, RHGP provides a means to interrogate the entire genome for any genetic change that is causative of the phenotype under investigation. We apply RHGP to identify and validate a set of host-oriented targets that are required for influenza infection.


Chemicals and cell culture

MDCK cells, human 293HEK cells and MEM medium were purchased from the American Type Culture Collection (ATCC). Influenza virus A/Udorn/72 was purchased from Charles River Lab. All siRNA oligonucleotides were synthesized by Dharmacon Research.

Construction of pTet-Off transactivator vector and the MDCK/Tet-off stable cell lines

The pTet-Off transactivator vector was constructed using pIREShyg3 (Clontech) as a backbone. A PCR-generated tTA fragment was inserted into the BamHI and SofI sites of the pIREShyg3 vector to generate the pTet-Off transactivator vector, pIREShyg3-tTA. To create MDCK/Tet-Off cell lines, 2 ×106 MDCK cells were cultured in 10 cm plates and transfected with 10 μg pIREShyg3-tTA vector using Fugene 6 (Roche). Twenty-four hours post-transfection, log-phase cultures of MDCK cells selected using 200 μg/mL hygromycin and the medium was changed every 3 days until all controls cells were killed. Hygromycin resistant clones were screened for the Doxycylcine (Dox; Invitrogen) induction activity using pTRE-luc (Clontech) vector and the Bright-Glo luciferase assay kit (Promega). Stable clones with at least 40-fold induction were used to create RHGP libraries.

Construction of the RHGP gene search vector, pRHGP22

The RHGP gene search vector, pRHGP22 was constructedusing the pLEST vector as a backbone (provided by Dr. Stanley Cohen, Stanford) (Lu et al., 2004). The pTRE-Tight-CMV-BlasticidinR was substituted as the promoter. The 3′ UTR WPRE element was removed and the p15A ori-Chloramphenicol acetyltransferase (CAT) resistant gene was inserted at the same site. Two LoxP sites were inserted downstream from the BlasticidinR gene and upstream from the p15A ori-CAT, respectively.

Production of RHGP virus and Construction of RHGP libraries

RHGP lentiviruses were produced using ViroPower Expression System (Invitrogen). HEK293FT cells were plated in 10 cm plates at 106 cells per plate. After 24 h incubation, the cells were transfected with 3 μg RHGP22 and 9 μg ViroPower Packaging Mix using Lipofectimine 2000 (Invitrogen). The medium was changed after 5h incubation at 37°C. After 48 h, viruses in the culture medium were filtrated through a 0.45 μm filter and tittered according to the manufacturer’s instruction.

To construct the RHGP library, MDCK/Tet-off cells were plated in 15 cm plates at 2×106 cells per plate. The next day, the samples were washed with PBS and transduced with RHGP viruses in 6μg/mL polybrene. The samples were incubated at 37°C for 2h with mixing prior to replacement with fresh media. Forty eight hours post-transduction, the cells were washed with PBS and infected with influenza virus A/Udorn/72 at a multiplicity of infection (MOI) of 0.1 for 2h in DMEM medium containing 0.2% BSA, 200 μg/mL hygromycin, 2 μg/mL bovine trypsin and 6 μg/mL Blasticidin with gentle mixing every 15 min. The media was then replenished with MEM. After two days, the cells were washed with PBS and infected again with influenza virus as indicated above. After the second round of infection, the samples were washed with PBS and incubated in conditioning media containing 200 μg/mL hygromycin and 6 μg/mL blasticidin. The medium was changed every 3 days until all non-transduced cells were killed and the virus resistant colonies had formed. Two weeks after second round of infection, the visible individual colonies were isolated and subcultured in 24-well plates.

Reversibility assay for influenza virus resistant cell clones

To verify reversibility of RHGP event, the MDCK/Tet-Off cells were cultured in absence or presence of Dox for 3 days. The cells were plated in 10 cm plates at 2×106 cells per plate and cultured with or without Dox. After 24 h, the samples were infected with influenza virus A/Udorn/72 at MOI of 1×10−2 and 1×10−5, respectively. After 3 days, surviving cells were stained with crystal violet to visualize the remaining cells.

Identification of the candidate genes from influenza virus resistant clones

1×106 cells from each clone were suspended in lysis buffer containing 0.32 M Sucrose, 10 mM Tris pH 7.5, 5 mM MgCl2 and 1% Triton X-100. The lysate was centrifuged at 1500 × g for 15 min and the pellet was treated with 100μg of proteinase K (Sigma) in proteinase K buffer containing 25mM EDTA, 150mM NaCl and 40mM Tris, 0.5% SDS, pH 7.5. The mixture was incubated overnight at 37°C, treated with 50μg RNase A (Sigma) and incubated at 37°C for 2h. Genomic DNA was extracted with phenol/isoamyl alcohol/chloroform followed by precipitation with isopropanol. The DNA pellet was washed with 70% ethanol and dissolved in TE buffer (pH 7.5).

Genomic DNA (10 μg) was digested with restriction enzyme BamHI or XbaI, extracted with phenol/isoamyl alcohol/chloroform, precipitated with ethanol and dissolved in TE buffer. Digested DNA (2.5μg) was self-ligated overnight using T4 ligase (Invitrogen) at 16 °C. The ligated DNA was precipitated with ethanol and washed with 70% ethanol. The DNA pellet was dissolved in TE buffer and electroporated into DH10B ElectroMax competent cells (Invitrogen). Multiple colonies were isolated for plasmid DNA preparation and restriction enzyme digestion. The plasmid DNA was further used to identify the target genes by DNA sequencing and genome mapping.

Validation of host target genes with siRNA

The human duplex siRNA homologues for candidate gene MRPL42, COX5A, SLC25A25 and TAPT1 were prepared as recommended by the manufacturer. The siRNA NP-1496 sequence (GGAUCUUAUUUCUUCGGAGUU), which targets the nucleocapsid (NP) gene of influenza virus, provided a positive control (Ge et al., 2003). Non-targeting siRNA, siCONTROL1 provided a negative control. HEK293 cells were plated in 24-well plates at 1×105 cells per well, respectively. After 24h incubation, the cells were transfected with 20 nM of siRNA and TransIT-TKO, according to the manufacturer’s instruction (Mirus). The cells were re-transfected with siRNA after 24h incubation. Twenty-four hours after second round of transfection, the samples were washed with MEM followed by infection with influenza virus A/Udorn/72 (MOI 1). The cells were incubated for 1h with gentle rocking every 15 min. The culture medium from each well was collected 48 h post-transfection and progeny viruses in the medium were titrated using standard plaque assays.


Construction of the RHGP cellular library

The central feature of RHGP is a unique lentiviral-based genetic element, known as a gene search vector (GSV), which was designed to interrogate the entire genome and identify targets that allow host cells to resist or survive infection with influenza virus. Our experimental strategy centered upon integration of the GSV at a single site in the genome, where it regulated expression of the target gene via an inducible promoter (Figure 1A). The vector encoded for a self-inactivating lentiviral LTR, which prevented the GSV from re-emerging from a transduced cell. The RSV E/P promoter was used only for production of the GSV and was eliminated by the integration of the GSV into the host genome.

Figure 1
Schematic overview of the RHGP Gene Search Vector

The GSV could integrate in either a sense or an antisense orientation. In the antisense configuration, the integration event itself inactivated one allele and facilitated expression of an antisense construct, which knocked down genes encoded on the other allele (Figure 1B, left panels). In this way, RHGP facilitated homozygous perturbation of both gene copies in diploid cells. In the opposite (sense) orientation, RHGP facilitated overexpression of the target gene (Figure 1B, right panels). This outcome could extend beyond simple overexpression of an entire gene (e.g., insertion upstream of the start site) since integration downstream of the start site could trigger overexpression of domains, which could produce a dominant-negative inhibitor of wild-type gene function. As such, RHGP allowed us to interrogate the entire cell genome to identify different types of events that allow host cells to resist or survive influenza infection.

To facilitate RHGP, a library of MDCK cells was transduced using a lentiviral-based GSV. As the average gene size in the genome is estimated to be 27kb (Gupta and Varshney, 2004), we calculated that 105 independent integration events would ensure random coverage of the entire genome (Li and Cohen, 1996). As outlined in the Materials and Methods, pRHGP22 was constructed using an expression cassette consisting of a pTRE-Tight-CMV promoter driving a blasticidin resistance gene, which provided a selectable marker under strict inducible control. The library of transduced host cells was then selected using blasticidin and surviving cells were screened for influenza resistance as detailed below.

Screening the influenza virus resistant phenotypes

To identify targets that render host cells resistant to influenza, an RHGP library was generated in MDCK cells and influenza challenge performed as outlined in Figure 2. This particular cell model was selected based on its well-established responsiveness to influenza infection and because the canine genome has been annotated and can be compared with human homologues. We conducted preliminary studies to ensure that MDCK cells were efficiently and entirely killed by A/Udorn/72 under the conditions utilized for selection. This outcome was essential to ensure that surviving cells arose as a result of the RHGP perturbation and not as an artifact of spontaneous resistance to influenza. Following library construction with the pRHGP22 GSV, a library consisting of at least 107 independent MDCK clones was created to ensure 100-fold coverage of the genome. A low MOI (0.1) was employed during the library creation to minimize the transduction of any cell by more than one different GSV. As an additional means of confirming GSV integration, the MDCK library was incubated with a lethal concentration of blasticidin (administered 48 hours post-infection).

Figure 2
Schematic overview of the strategy to identify host genes necessary for influenza infection

The library was challenged by infection with influenza A/Udorn/72 to select for influenza-resistant cells. We had previously established that infection with A/Udorn/72 (MOI of 10−1) reproducibly killed all MDCK cells within 48 hours (not shown). As a control, parallel cultures of mock-transduced cells were treated identically and no survivors were observed after 48 hours. We further sought to minimize artifacts by subjecting surviving cells to multiple rounds of lethal challenge. Thus, the samples were subjected to at least three rounds of influenza challenge, at which time no surviving cells were detected in any of the matched controls (not shown).

In the course of these studies, we considered that challenge at a high relative MOI of influenza might bias for certain types of targets. For example, survival at a high MOI might unintentionally favor targets associated with early stages of infection or bias against targets associated with later stage events. To preclude such bias, a parallel series of studies were conducted to compare survival in the context of viral challenge at low titer. For this, the clones were subjected to infection at low initial MOI (MOI of 10−5). The cultures were then evaluated over time and using multiple rounds of infection to ensure that all cells in the control samples had been killed by influenza (see Figure 3 for an example).

Figure 3
Validation of influenza virus resistant subclones with reversibility assay

After selecting for RHGP-transduced cells that survived challenge with influenza, clones were isolated by single cell cloning and expanded, yielding an average of 3000 clones per experiment. Given the abundance of clones, a subset of 303 different influenza-resistant clones (10% of the total population) was selected at random and subjected to a second round of single cell cloning to ensure clonality (Table I). These subclones were then subjected to a battery of studies to confirm the phenotype and identify the target genes. Each subclone was tested again for the ability to survive a lethal challenge with influenza. Parental MDCK host cells that had been transduced with an inactive Tet-off GSV provided a matched negative control. 129 different subclones (43% of the initial population) were isolated based on their ability to survive lethal challenge with influenza (See the “-Dox” samples in Figure 3A for representative examples of influenza-resistant clones). Most of the down-selection at this stage resulted from colonies that failed to survive the single cell cloning process.

Table I
Summary of RHGP-Based Discovery of Gene Perturbations that Render Host Cells Resistant to a Lethal Challenge with Influenza Virus

Reversibility assay of the virus resistant clones

A key feature of the RHGP technology is the ability to validate candidate targets via regulation by an inducible promoter. This allowed us to eliminate candidates that might have become resistant to influenza as a result of spontaneous mutation or other artifacts not related to RHGP. Since the promoter for the RHGP vector was under control of a Tet-off system, we compared influenza-mediated killing of the candidates in the presence or absence of Doxycycline. This study confirmed that 111 of the 129 surviving clones (86%) demonstrated reversible resistance to influenza challenge (See Figure 3A and Figure 3B for representative findings of a reversible phenotype). The remaining 18 clones (14%) were de-prioritized to ensure that RHGP, and not unrelated artifacts, was responsible for the influenza-resistant phenotype.

Identification of the host gene by genomic DNA cloning

The RHGP gene search vector was designed to efficiently locate target genes and determine the orientation (sense or antisense) of the integration event. Specifically, the gene search vector encodes for an Ori-CAT reporter gene, which can be rescued by restriction enzyme-based genomic DNA cloning. Genomic DNA was isolated from the 111 clones that had demonstrated reversible resistance to influenza, yielding a total of 110 target genes (Table I). The resulting genomic DNA sequences flanking the RHGP vector insertion sites were subjected to genome mapping against the canine genome using the UCSC Genome Browser. Consistent with the experimental design, most targets consisted of a single integration site per clone. We were unable to isolate candidate genes from a small subset of candidates due to technical limitations. In addition, a minority of samples contained multiple integration events, which could have arisen as a result of multiple RHGP integration sites or if the subclone had not been clonal. Due to this ambiguity, the genes associated with multiple integration events were de-prioritized for further validation.

The site and orientation of integration offered by RHGP provided insight into the types of perturbations that allowed host cells to survive challenge with influenza (Figure 4). Specifically, the RHGP perturbations were broadly divided into three groups: 1) “Antisense”: Antisense integration events that facilitated disruption of one allele and antisense inhibition of the other allele; 2) “Sense Downstream”: Integration in a sense orientation, which would be predicted to facilitate production of one or more domains, which could act as a dominant-negative inhibitor of the full-length gene product; and 3) “Sense-Upstream”: Integration in a sense orientation upstream of the start site, which could facilitate overexpression of the entire target gene. Of the 99 known targets identified using RHGP, 56 targets (56% of total targets) represented “Antisense” knockdown of target expression. Another 35% of the targets represented “Sense-Downstream” events, likely conveying overexpression of dominant-negative inhibitors of wild-type gene expression. Consistent with this interpretation, we noted that one particular target, PTBP1, was independently identified as a target in two separate clones; one clone in an “Antisense” orientation and another in a “Sense-Downstream” orientation. The remaining 9% represented “Sense-Upstream” insertions. The orientation of 11 target genes (10% of the 110 targets sequenced herein) could not be ascertained because the gene itself had not been previously described and therefore assignment of orientation was not possible.

Figure 4
Gene Search Vector Integration and Resultant Perturbations of Host Targets

In a separate assessment, we asked if these 110 target genes had been linked previously with viral infection. Only four of the 110 targets identified herein (MDN1, GRK6, AKT1 and STXBP1) had been directly linked viral infection (Table II), suggesting novel information about 106 targets (96%) identified using RHGP. By extending our analysis to related proteins, we found that 23 of 110 targets (21%) identified herein shared homology with or was a co-member of a superfamily with, host targets that had been linked previously with viral infection. These findings suggest RHGP identified novel functions for the preponderance (75%) of the host targets. Two targets identified herein (C4orf32 and C21orf33) had simply been described as open reading frames (orfs), with no function ascribed. Altogether, these findings suggest that RHGP-based interrogation of the host genome had identified both novel targets and/or ascribed novel functions to known genes.

Table II
Comparison of Common Families of Host-Oriented Targets Identified Using RHGP Versus siRNA Approaches

Validation of target genes using naïve cells

RHGP identified a series of targets in MDCK cells that conferred resistance to influenza infection and the reversibility of the phenotype provided an initial validation for these targets. We then sought to verify these candidates using an independent experimental system to exclude outcomes that might arise as an artifact of the MDCK cells or RHGP technology. Since the targets genes had been isolated from canine-derived MDCK cells, we identified their human homologues and expressed siRNAs selective for these targets in human HEK-293 cells.

A subset of targets was selected for independent validation in HEK-293 cells. Due to the sheer volume and efficiency of targets identified in our RHGP-based screen, we recognized that it would be impossible to validate all the targets. Therefore, a subset of targets was selected to establish proof of concept. Duplex siRNAs were generated for MRPL42, COX5A, TAPT1 and SLC25A25 and these siRNAs were transfected into HEK293 cells (Figure 5). Non-targeting siRNAs provided a matched control for the transfection and a reference standard. NP-1496 duplex siRNAs were synthesized and provided a positive control. NP-1496 was selected based on a recent report that siRNA targeting of NP-1496 efficiently decreases influenza infection (Ge et al., 2003). Since influenza infection does not efficiently kill HEK293 cells, we modified our experimental protocol to measure viral titers (instead of host cell survival) as the primary endpoint for efficacy. The siRNA-treated HEK293 cells were infected with A/Udorn/72 for 48 hours and viral titers were measured by plaque assays. As indicated in Figure 5, MRPL42, COX5A, SLC25A35 or TAPT1 duplex siRNAs each decreased influenza virus production and at levels comparable to the positive control (NP-1496 siRNAs). After validating the findings using Udorn, we then extended the results to other influenza viruses. For example, these siRNAs specifically also blocked host cell killing by A/WSN/33, a different variant of influenza (data not shown). Altogether, these findings validate the application of RHGP to identify novel host-based targets and suggest the potential for broad-spectrum application to different forms of influenza.

Figure 5
Validation of representative RHGP targets via siRNA targeting of HEK293 cells


The major finding of our present study is the demonstration that RHGP can provide an efficient means to conduct genome-wide screening for host factors that are required for influenza virus infection. RHGP identified modifications that prevented influenza-mediated killing of host cells including targets that have been over-expressed or knocked-down. We were able to validate these targets using an inducible promoter within the RHGP vector to reverse the phenotype, thereby minimizing the potential for spontaneous artifacts that were not attributable to the RHGP strategy. This genome-wide screen allowed us to identify 110 novel targets that render host cells resistant to an otherwise lethal challenge with influenza virus. Of these targets, most (106 of 110) had not been previously linked with influenza. In addition, we ascribe novel functions to previously-unknown genes and orfs. As proof of concept, a subset of targets was further validated in independent studies using conventional siRNA approaches.

One unique aspect of our present study is the use of an improved form of RHGP. Previous studies with RHKO (Random Homozygous Knockout), a precursor technology to RHGP, had employed MMLV-based gene search vectors (GSVs) (Li and Cohen, 1996). The MMLV technologies were intrinsically limited by low titers of GSV production and this inefficiency limited the ability to conduct a thorough assessment of the genome. We demonstrate for the first time that a lentiviral system (RHGP) overcomes these prior limitations. A lentiviral vector also favors single-site insertion into actively-transcribed genes (Mitchell et al., 2004), which further increases the efficiency of the screening procedure.

RHGP is not biased by prior knowledge of the target. We analyzed the human homologues of the remaining targets using the PANTHER Classification System (Mi et al., 2005) to identify putative biological pathways. These targets could be broadly related to different biological processes or molecular functions. Of the 99 targets available for evaluation within the PANTHER system, 91 could be classified and revealed 7 common biological processes and 8 major molecular functions. Prominent among the biological processes were pathways associated with nucleic acid and protein metabolism and intracellular protein trafficking (Figure 6). The target functions included nucleic acid binding, enzymatic activities (membrane traffic proteins, kinases, ligases, hydrolases), cell signaling and intracellular transport. It is important to note that the assay utilized (cell survival following influenza challenge) does not facilitate assessment of all cellular genes. Specifically, this outcome precludes targets, where RHGP perturbation of normal expression or function causes toxicity to the host cell. Although this outcome limits the coverage of potential host targets, we believe such an outcome is nonetheless desirable since it might minimize potential targets that might be associated with toxic side effects.

Figure 6
Classification of the target genes by Panther Classification System

A recent series of siRNA-based studies identified host targets associated with infection by influenza, HIV, or West Nile virus (See Table II for summary) (Brass et al., 2008; Hao et al., 2008; Konig et al., 2008; Krishnan et al., 2008). These earlier studies utilized host cells that were derived from different species, which increases the complexities of comparisons with our present findings. Compounding this, siRNA is increasingly utilized for target discovery but is intrinsically limited by the need for robust and sustained over-expression of the siRNA. The outcomes of some siRNA findings have also been clouded by questions of whether the siRNAs might non-specifically alter host defense mechanisms (Kleinman et al., 2008), which could be particularly problematic for applications of siRNA technology to antiviral disease. RHGP might also induce potential antiviral host defense mechanisms. For example, an antisense orientation of the GSV is expected to induce double stranded RNA, which itself can trigger host antiviral responses. Such a mechanism could give rise to inherent resistance to influenza infection, particularly when studying targets, where high levels of double stranded RNA have accumulated. Such a response would not be expected to impact situations, where the GSV inserts in a “Sense” orientation. With this in mind, more than half of the influenza resistant clones (56%) were attributed to an antisense insertion of the GSV and we cannot exclude that some of these antisense events might be subject to this alternative explanation.

Most of the targets and pathways identified herein had not been previously linked with influenza infection. This outcome likely reflects the fact that RHGP can identify targets based on loss of expression, overexpression of entire genes or overexpression of gene domains. Some pathways identified herein had some relationship (in terms of sequence, protein families or mechanistic pathways) with host genes that had been linked with HIV or West Nile virus infection (Brass et al., 2008; Konig et al., 2008; Krishnan et al., 2008). These similarities may suggest different viral pathogens may share host pathways. The genes indentified herein are less similar to a recent siRNA-based influenza study (Hao et al., 2008) (Table II). We postulate this difference may reflect the use of a Drosophila host cell system in the earlier study as compared to a canine system herein.

Our siRNA studies indicate that decreased expression of MRPL42, COX5A, TAPT1 or SLC25A25 decreases susceptibility to influenza infection (Figure 5). MRPL42 is a mitochondrial ribosomal protein involved in mitochondrial protein synthesis. Although our current evidence linking MRPL42 with influenza is novel, a recent study indicated that introduction of SIV-Nef in CD4+ T cells stimulated expression of related mitochondrial ribosomal proteins, MRPL1, MRPL14 and MRPL19 (Ndolo et al., 2006). This outcome is intriguing and may suggest common host mechanisms that are shared by HIV and influenza. Likewise, COX5A is a component of the cytochrome C oxidase (COX) complex and a different subunit, COX6A1, was recently shown to regulate influenza replication (Hao et al., 2008). The transmembrane protein TAPT1, which normally governs axial skeletal patterning during development (Howell et al., 2007), can also function as a cellular receptor for the human cytomegalovirus gH receptor (Baldwin, Kleinberg, and Keay, 1996). The mechanistic basis of TAPT1 in the regulation of influenza infection is currently unknown and will require additional investigation. Finally, members of the same gene superfamily containing SLC25A25 reportedly regulate HIV (Brass et al., 2008; Konig et al., 2008) and West Nile virus (WNV) (Krishnan et al., 2008) infection.

Overexpression of some targets allowed MDCK cells to survive influenza infection. Prominent among these are the B-cell CLL/lymphoma 2 (BCL2) and plasminogen activator, urokinase (PLAU) genes. The proto-oncogene, BCL2, is a well-characterized regulator of cell survival and BCL2 regulates cell survival during infection by Sindbis (Levine et al., 1993), influenza (Hinshaw et al., 1994) or HIV-1 (Mishra, Mishra, and Kumar, 2007) viruses. Likewise, the amino-terminal fragment of PLAU was recently shown to suppress the assembly and budding of HIV-1 from infected cells (Wada et al., 2001). PLAU and its receptor have also been linked with the regulation of infection by human respiratory syncytial virus (Martinez et al., 2007).

Although these studies provide compelling new information, it is important to appreciate that cell culture-based information may not necessarily translate into new therapeutics. For example, transgenic mouse studies demonstrate that homozygous knock-out of TAPT1 results in embryonic lethality (Howell et al., 2007), which may indicate that such targets may not be suitable for safe therapeutic intervention. Future investigation will be necessary to extrapolate our present findings with cell-based studies to work with more complex systems. Nonetheless, the work described above may provide new insight into new opportunities for future host-based therapeutic targeting.


This work in this manuscript was supported by a grant from the National Institute of Allergy and Infectious Diseases (1R43 AI068217-01). The authors thank Dr. Stanley Cohen (Stanford University) for providing plasmid reagents and expert advice for RHKO technologies and Dr. Zhiping Ye (US Food and Drug Administration) for providing influenza A/WSN/33 virus.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Ahlquist P, Noueiry AO, Lee WM, Kushner DB, Dye BT. Host factors in positive-strand RNA virus genome replication. Journal of Virology. 2003;77(15):8181–8186. [PMC free article] [PubMed]
  • Baldwin BR, Kleinberg M, Keay S. Molecular cloning and expression of receptor peptides that block human cytomegalovirus cell fusion. Biochemical and Biophysical Research Communications. 1996;219(2):668–673. [PubMed]
  • Belongia E, Kieke B, Coleman L, Donahue J, Irving S, Meece J, Vandermause M, Shay D, Gargiullo P, Balish A, Foust A, Guo L, Lindstrom S, Xu X, Klimov A, Bresee J, Cox N. Interim within-season estimate of the effectiveness of trivalent inactivated influenza vaccine - Marshfield, Wisconsin, 2007–08 influenza season (Reprinted from vol 57, pg 393–398, 2008) Jama-Journal of the American Medical Association. 2008;299 (20):2381–2384.
  • Brass AL, Dykxhoorn DM, Benita Y, Yan N, Engelman A, Xavier RJ, Lieberman J, Elledge SJ. Identification of host proteins required for HIV infection through a functional genomic screen. Science. 2008;319 (5865):921–926. [PubMed]
  • Chen BJ, Lamb RA. Mechanisms for enveloped virus budding: Can some viruses do without an ESCRT? Virology. 2008;372(2):221–232. [PMC free article] [PubMed]
  • Fox JL. Antivirals become a broader enterprise. Nature Biotechnology. 2007;25 (12):1395–1402. [PubMed]
  • Ge Q, McManus MT, Nguyen T, Shen CH, Sharp PA, Eisen HN, Chen JZ. RNA interference of influenza virus production by directly targeting rnRNA for degradation and indirectly inhibiting all viral RNA transcription. Proceedings of the National Academy of Sciences of the United States of America. 2003;100(5):2718–2723. [PubMed]
  • Gupta PK, Varshney RK. Cereal genomics: An overview. Cereal Genomics. 2004:1–18.
  • Hao LH, Sakurai A, Watanabe T, Sorensen E, Nidom CA, Newton MA, Ahlquist P, Kawaoka Y. Drosophila RNAi screen identifies host genes important for influenza virus replication. Nature. 2008;454 (7206):890–U46. [PMC free article] [PubMed]
  • Hinshaw VS, Olsen CW, Dybdahlsissoko N, Evans D. APOPTOSIS - A MECHANISM OF CELL-KILLING BY INFLUENZA-A AND INFLUENZA-B VIRUSES. Journal of Virology. 1994;68 (6):3667–3673. [PMC free article] [PubMed]
  • Howell GR, Shindo M, Murray S, Gridley T, Wilson LA, Schimenti JC. Mutation of a ubiquitously expressed mouse transmembrane Protein (Tapt1) causes specific skeletal homeotic transformations. Genetics. 2007;175 (2):699–707. [PubMed]
  • Kleinman ME, Yamada K, Takeda A, Chandrasekaran V, Nozaki M, Baffi JZ, Albuquerque RJC, Yamasaki S, Itaya M, Pan YZ, Appukuttan B, Gibbs D, Yang ZL, Kariko K, Ambati BK, Wilgus TA, DiPietro LA, Sakurai E, Zhang K, Smith JR, Taylor EW, Ambati J. Sequence- and target-independent angiogenesis suppression by siRNA via TLR3. Nature. 2008;452 (7187):591–U1. [PMC free article] [PubMed]
  • Konig R, Zhou YY, Elleder D, Diamond TL, Bonamy GMC, Irelan JT, Chiang CY, Tu BP, De Jesus PD, Lilley CE, Seidel S, Opaluch AM, Caldwell JS, Weitzman MD, Kuhen KL, Bandyopadhyay S, Ideker T, Orth AP, Miraglia LJ, Bushman FD, Young JA, Chanda SK. Global analysis of host-pathogen interactions that regulate early-stage HIV-1 replication. Cell. 2008;135 (1):49–60. [PMC free article] [PubMed]
  • Krishnan MN, Ng A, Sukumaran B, Gilfoy FD, Uchil PD, Sultana H, Brass AL, Adametz R, Tsui M, Qian F, Montgomery RR, Lev S, Mason PW, Koski RA, Elledge SJ, Xavier RJ, Agaisse H, Fikrig E. RNA interference screen for human genes associated with West Nile virus infection. Nature. 2008;455 (7210):242–U67. [PMC free article] [PubMed]
  • Lamb RA, Takeda M. Death by influenza virus protein. Nature Medicine. 2001;7(12):1286–1288. [PubMed]
  • Levine B, Huang Q, Isaacs JT, Reed JC, Griffin DE, Hardwick JM. CONVERSION OF LYTIC TO PERSISTENT ALPHAVIRUS INFECTION BY THE BCL-2 CELLULAR ONCOGENE. Nature. 1993;361 (6414):739–742. [PubMed]
  • Li L, Cohen SN. Tsg101: a novel tumor susceptibility gene isolated by controlled homozygous functional knockout of allelic loci in mammalian cells. Cell. 1996;85 (3):319–29. [PubMed]
  • Lu Q, Wei WS, Kowalski PE, Chang ACY, Cohen SN. EST-based genome-wide gene inactivation identifies ARAP3 as a host protein affecting cellular susceptibility to anthrax toxin. Proceedings of the National Academy of Sciences of the United States of America. 2004;101(49):17246–17251. [PubMed]
  • Ludwig S, Pleschka S, Wolff T. A fatal relationship - Influenza virus interactions with the host cell. Viral Immunology. 1999;12(3):175–196. [PubMed]
  • Luscher-Mattli M. Influenza chemotherapy: a review of the present state of art and of new drugs in development. Archives of Virology. 2000;145(11):2233–2248. [PubMed]
  • Martinez I, Lombardia L, Garcia-Barreno B, Dominguez O, Melero JA. Distinct gene subsets are induced at different time points after human respiratory syncytial virus infection of A549 cells. Journal of General Virology. 2007;88:570–581. [PubMed]
  • Mi HY, Lazareva-Ulitsky B, Loo R, Kejariwal A, Vandergriff J, Rabkin S, Guo N, Muruganujan A, Doremieux O, Campbell MJ, Kitano H, Thomas PD. The PANTHER database of protein families, subfamilies, functions and pathways. Nucleic Acids Research. 2005;33:D284–D288. [PMC free article] [PubMed]
  • Mishra S, Mishra JP, Kumar A. Activation of JNK-dependent pathway is required for HIV viral protein R-induced apoptosis in human monocytic cells - Involvement of antiapoptotic BCL2 and c-IAP1 genes. Journal of Biological Chemistry. 2007;282 (7):4288–4300. [PubMed]
  • Mitchell RS, Beitzel BF, Schroder ARW, Shinn P, Chen HM, Berry CC, Ecker JR, Bushman FD. Retroviral DNA integration: ASLV, HIV, and MLV show distinct target site preferences. Plos Biology. 2004;2 (8):1127–1137. [PMC free article] [PubMed]
  • Ndolo T, George M, Nguyen H, Dandekar S. Expression of simian immunodeficiency virus Nef protein in CD4(+) T cells leads to a molecular profile of viral persistence and immune evasion. Virology. 2006;353(2):374–387. [PubMed]
  • Ong AK, Hayden FG. John F Enders lecture 2006: Antivirals for influenza. Journal of Infectious Diseases. 2007;196(2):181–190. [PubMed]
  • Reeves JD, Piefer AJ. Emerging drug targets for antiretroviral therapy. Drugs. 2005;65 (13):1747–66. [PubMed]
  • Tan SL, Ganji G, Paeper B, Proll S, Katze MG. Systems biology and the host response to viral infection. Nature Biotechnology. 2007;25 (12):1383–1389. [PubMed]
  • Thompson WW, Shay DK, Weintraub E, Brammer L, Cox N, Anderson LJ, Fukuda K. Mortality associated with influenza and respiratory syncytial virus in the United States. Jama-Journal of the American Medical Association. 2003;289 (2):179–186. [PubMed]
  • Wada M, Wada NA, Shirono H, Taniguchi K, Tsuchie H, Koga J. Amino-terminal fragment of urokinase-type plasminogen activator inhibits HIV-1 replication. Biochemical and Biophysical Research Communications. 2001;284 (2):346–351. [PubMed]
  • Wheeler J, McHale M, Jackson V, Penny M. Assessing theoretical risk and benefit suggested by genetic association studies of CCR5: experience in a drug development programme for maraviroc. Antiviral Therapy. 2007;12(2):233–245. [PubMed]