key: cord-0805344-rbk1ehz1
authors: Orr-Burks, Nichole; Murray, Jackelyn; Todd, Kyle V.; Bakre, Abhijeet; Tripp, Ralph A.
title: G-Protein-Coupled Receptor and Ion Channel Genes Used by Influenza Virus for Replication
date: 2021-04-12
journal: J Virol
DOI: 10.1128/jvi.02410-20
sha: b8f5a27342796111f5c0cea80a9741cd6377122d
doc_id: 805344
cord_uid: rbk1ehz1

Influenza virus causes epidemics and sporadic pandemics resulting in morbidity, mortality, and economic losses. Influenza viruses require host genes to replicate. RNA interference (RNAi) screens can identify host genes coopted by influenza virus for replication. Targeting these proinfluenza genes can provide therapeutic strategies to reduce virus replication. Nineteen proinfluenza G-protein-coupled receptor (GPCR) and 13 proinfluenza ion channel genes were identified in human lung (A549) cells by use of small interfering RNAs (siRNAs). These proinfluenza genes were authenticated by testing influenza virus A/WSN/33-, A/CA/04/09-, and B/Yamagata/16/1988-infected A549 cells, resulting in the validation of 16 proinfluenza GPCR and 5 proinfluenza ion channel genes. These findings showed that several GPCR and ion channel genes are needed for the production of infectious influenza virus. These data provide potential targets for the development of host-directed therapeutic strategies to impede the influenza virus productive cycle so as to limit infection. IMPORTANCE Influenza epidemics result in morbidity and mortality each year. Vaccines are the most effective preventive measure but require annual reformulation, since a mismatch of vaccine strains can result in vaccine failure. Antiviral measures are desirable particularly when vaccines fail. In this study, we used RNAi screening to identify several GPCR and ion channel genes needed for influenza virus replication. Understanding the host genes usurped by influenza virus during viral replication can help identify host genes that can be targeted for drug repurposing or for the development of antiviral drugs. The targeting of host genes is refractory to drug resistance generated by viral mutations, as well as providing a platform for the development of broad-spectrum antiviral drugs.

Viruses exploit host genes and their pathways to support entry, replication, and egress. Some of the most studied pathways exploited by influenza virus include the nuclear factor kappa B (NF-κB), phosphatidylinositol 3-kinase (PI3K), mitogen-activated protein kinase (MAPK), protein kinase C/protein kinase R (PKC/PKR), toll-like receptor (TLR), and retinoic acid-inducible gene 1 (RIG-I) pathways (14) (15) (16) (17) . Anti-influenza drugs typically target viral proteins, but often these drugs can have reduced efficacy due to drug resistance acquired through antigenic shift and drift (18) . For example, amantadine is no longer recommended for the treatment of influenza virus infection due to increased drug resistance, and the reduced efficacy observed for oseltamivir is linked to neuraminidase (NA) mutations (19) , creating inconsistencies among therapies (20) . In contrast, therapeutics targeting host genes necessary for virus replication could offer an approach refractory to drug resistance while providing broader-spectrum drug efficacy.

RNA interference (RNAi) is a conserved mechanism of posttranscriptional gene-specific regulation (21) . RNAi can probe the virus-host interface to identify host genes necessary for virus replication (22) (23) (24) (25) (26) . Genome-wide RNAi screening has uncovered key virus-host interactions, has helped identify drug targets for influenza viruses (27) , and has been used to validate host genes important for virus replication (28) (29) (30) (31) (32) . Small interfering RNAs (siRNAs) mediate posttranscriptional gene silencing via sequence-specific nucleolytic cleavage or translational inhibition upon interaction with their target mRNAs (29) . siRNAs are rationally designed to be specific for one mRNA target (33) .

G-protein-coupled receptors (GPCRs) are a family of seven-transmembrane cell surface receptor proteins that facilitate intracellular communication via activation of signal transduction pathways (34) . Viruses use GPCRs to facilitate attachment, entry, replication, and egress. For example, HIV tropism is associated with the CXCR4/CCR5 coreceptor and GPCR15 (35) (36) (37) . In addition, blocking of select GPCRs with drug antagonists obstructs Marburg virus and Ebola virus cell entry and replication (38) . The overarching influence of GPCRs on the cell makes drugs that target GPCRs amenable to disease intervention. Similarly, ion channels (ICs) are assemblages of integral protein domains that allow transmembrane passage between the extracellular and intracellular components of the cell (39) . ICs enable the influx/efflux of Na 1 , K 1 , Cl -, or Ca 21 ions, which regulate effector pathways. For example, inhibition of K 1 channels at the early stages of Bunyamwera virus infection hinders virus replication postentry (40) . In addition, Clchannels are important for herpes simplex virus 1 entry and virus-host fusion (41) . Further, the Na 1 channel opener SDZ-201106 can inhibit IAV replication via PKC pathway inhibition (42) , and modulation of Cl 2 or Na 1 secretion/absorption in the respiratory tract contributes to the regulation of respiratory disease (43) .

In this study, we used RNAi as a tool to survey the virus-host interface connected to GPCR and IC genes needed for influenza virus replication. Using siRNA pools to mediate RNAi, we examined GPCR and IC genes for their effects on influenza virus replication in A549 cells based on the following: (i) Z-score, (ii) Ingenuity Pathway Analysis software (2014) (IPA; Qiagen, Inc., Valencia, CA; Qiagen Knowledge Base; Qiagen.com; i.e., searching public databases and published texts), (iii) the availability of small-molecule inhibitors and antagonists, and (iv) targeting by microRNAs (miRs). The gene hits from the RNAi screen of A/WSN/33-infected A549 cells were validated following deconvolution using A/WSN/33. Confirmed hits were reexamined using A/CA/04/09-or B/ Yamagata/16/1988-infected A549 cells. The findings from this study provide a better understanding of the virus-host interface and host genes needed for influenza virus replication and provide drug target information for the development of new drugs, or for the repurposing of existing FDA-approved drugs, to combat influenza.

An RNAi screen identifies GPCR genes. GPCR genes permit intracellular communication via signal transduction following activation (34) and are involved in virus replication (38, (44) (45) (46) . We performed a genome-wide RNAi screen of GPCR genes required for influenza virus replication in A549 cells. Briefly, A549 cells were reverse transfected with siRNA SMARTpools, and 48 h posttransfection, the cells were infected (multiplicity of infection [MOI], 0.001) with A/WSN/33. The levels of virus replication were determined, and a Z-score was applied that showed the number of standard deviations by which the gene knockdown event differed from the mean. A negative Zscore (#21.0) indicated decreased virus replication, while a positive Z-score ($1.0) indicated increased viral replication. Our study focused on gene knockdown events that decreased influenza virus titers, since the goal was to determine strategies for host cell-targeted antiviral therapeutics.

We identified 185 GPCR genes whose knockdown resulted in Z-scores of #21.0. Further evaluation of these genes with IPA and Gene Ontology (GO) analyses, as well as the implementation of selection criteria, identified 19 critical GPCR genes: ADGRF1,  ADORA1, ADRB2, AGTR1, C5AR2, CCKBR, FFAR1, HCAR3, HCRTR2, HRH2, HTR1B, LGR4,  LPAR3, MTNR1B, NMUR2, OXGR1 , OXTR, P2RY12, and PRLHR (Table 1) . GPCRs are grouped into six classes (A to F) based on sequence homology and functional similarity (34) . Sixteen of 19 GPCR genes were identified as class A; ADGRF1 belongs to class B, C5AR2 is a nonclassical GPCR, and LGR4 is an orphan receptor. To limit off-target results, the 19 GPCR genes identified by SMARTpool screens were reexamined by deconvolution of the siRNA pools (24, 47) . Here, A549 cells were transfected with individual ON TARGETplus (OTP)-modified siRNAs from the SMARTpool. OTP-siRNAs have improved gene targeting due to a dual-strand modification that provides increased interaction with the RNA-induced silencing complex (RISC), decreasing off-target effects by antisense strands (48) .

OTP-siRNA-transfected A549 cells were infected (MOI, 0.01) with A/WSN/33, and after 48 h, the levels of infectious virus production were determined by a plaque assay. GPCR genes that were knocked down by OTP-siRNAs and had decreases in virus plaque titers for two or more individual OTP-siRNAs were further evaluated. For example, silencing of the MTNR1B gene by transfecting cells with siRNA 4 from the SMARTpool markedly reduced influenza virus titers, but transfection of siRNA 1, 2, or 3 had only a modest effect (Fig. 1A) ; thus, the MTNR1B gene was not considered further. Additionally, silencing of the NMUR2 or PRLHR gene had no substantial effect on viral titers (Fig. 1A) . In contrast, OTP-siRNA knockdown of the ADGRF1, ADORA1, ADRB2, AGTR1, C5AR2, CCKBR, FFAR1, HCAR3, HCRTR2, HRH2, HTR1B, LGR4, LPAR3, OXGR1, OXTR, or P2RY12 gene resulted in decreased virus titers (#21.0) for two or more siRNAs (Fig. 1A and B) , and knockdown of the C5AR2, CCKBR, OXTR, or P2RY12 gene gave the greatest reduction in virus titers for two or more siRNAs (Fig. 1B) . Knockdown of the ADGRF1, ADRB2, C5AR2, CCKBR, HCRTR2, LPAR3, OXTR, or P2RY12 gene yielded a greater reduction in infectious viral titers than knockdown of the mitogen-activated protein kinase kinase (MAP2K) gene (29.54-fold change), which is known to limit the replication of influenza virus and thus to reduce infectious viral Results are presented as heat maps depicting fold changes in influenza virus titers (in PFU per milliliter) from titers with a nontargeting control siRNA (siNTC). A positive fold change equates to an increase in PFU per milliliter over the control. A negative fold change equates to a decrease in PFU per milliliter over the control. A zero fold change equates to no change in PFU per milliliter over the control. Asterisks indicate significant differences from the control by two-way mixed analysis of variance with Dunnett's multiple-comparison test (*, P , 0.05; **, P , 0.001; ***, P , 0.0001; ****, P , 0.00001). siNTC results are corrected to zero to reflect the baseline change in replication (which is zero). Results are normalized to those for the siNTC control. titers ( Fig. 1A and B) (49, 50) . Thus, 16 GPCR genes (ADGRF1, ADORA1, ADRB2, AGTR1, C5AR2, CCKBR, FFAR1, HCAR3, HCRTR2, HRH2, HTR1B, LGR4, LPAR3, OXGR1, OXTR, and P2RY12) were further evaluated. An RNAi screen identifies IC genes. Ion channels (ICs) are membrane-spanning proteins that allow for ion flux across cellular membranes (51) , which affects signaling cascades and effector functions (52) , as well as the activity and stability of viral proteins (53) (54) (55) . Thus, ion channels affect influenza virus replication (40) (41) (42) , since influenza viruses attach to the cell membranes during infection and incorporate the membrane into an acidified endosome, triggering conformational changes in HA (56, 57) . We screened 352 IC genes for their importance in influenza virus replication and found Z-scores of #21.0 for 173 IC genes. These proviral genes were analyzed by IPA and GO analyses, yielding 13 IC genes (ASIC1, CACNA1C, CHRNA1, GABRA3, GRID2, GRIN3A, KCNA7, KCNE2, KCNIP2, KCNMB2, MCOLN2, SCN7A, and SCNN1D) ( Table 2) . OTP-siRNA SMARTpools were deconvoluted (1 siRNA pool per treatment; 4 siRNAs per target) and reverse transfected into A549 cells, and then the cells were infected (MOI, 0.01) with A/ WSN/33 after 48 h (24, 56) . Levels of infectious influenza virus were determined by a plaque assay. IC genes that showed decreased plaque titers for two or more individual OTP-siRNAs were further evaluated. Silencing the CACNA1C, CHRNA1, GRIN3A, KCNA7, KCNE2, KCNIP2, KCNMB2, or SCN7A gene did not detectably affect virus titers relative to those for nontargeting siRNA controls (siNTC) (Fig. 1C) ; however, silencing the ASIC1, GABRA3, GRID2, MCOLN2, or SCNN1D gene resulted in a ,21.0-fold change. Silencing ASIC1 led to a greater reduction in influenza virus titers than silencing the MAP2K gene (24.3-fold change). Silencing SCNN1D resulted in a small decrease in viral titers; however, since SCNN1D is targeted by the ion channel inhibitor triamterene, and thus, a potential repurposed drug, identified by IPA, was available, this gene was further evaluated (58, 60) . Thus, a total of five ion channel genes-ASIC1, GABRA3, GRID2, MCOLN2, and SCNN1D-were further evaluated.

Distinctive GPCR and IC genes are utilized for the replication of influenza virus strains and subtypes. To better understand GPCR and IC genes that have influenza virus strain and type differences, the GPCR and IC genes were evaluated following A/CA/ 04/2009 or B/Yamagata/16/1988 infection of A549 cells using a plaque assay and a 50% tissue culture infective dose (TCID 50 ) assay. Our initial RNAi screen investigated A/WSN/33 infection of A549 cells at a lower MOI (0.001). To corroborate earlier data, gene hits were confirmed using individual OTP-siRNAs and a higher MOI Forty-eight hours postinfection, the titer and 50% tissue culture infective dose were determined by a plaque assay and a TCID 50 HA assay, respectively. The results showed that silencing 16 GPCR and 5 IC proinfluenza genes individually was associated with a .2-fold decrease in influenza plaque formation in A549 cells infected with A/WSN/33 ( Fig. 2A Infectious viral titers (expressed as PFU per milliliter) and TCID 50 titers were determined by an MDCK plaque assay and sample titration on MDCK cells followed by an HA assay, respectively. Plaque assay data for GPCR (A) and ion channel (B) genes and TCID 50 data for GPCR (C) and ion channel (D) genes are presented as the inverse of the fold decrease from the level with nontargeting control siRNA (siNTC) for three independent experiments performed in triplicate. A positive increase in the fold change equates to a decrease in PFU per milliliter or TCID 50 per milliliter from that with siNTC. Data show means 6 standard errors of the means for three independent experiments performed in triplicate. Asterisks indicate significant differences from the control by ordinary one-way analysis of variance with Dunnett's multiple-comparison test (*, P , 0.05; **, P , 0.001; ***, P , 0.0001; ****, P , 0.00001). siNTC results are corrected to zero to reflect the baseline change in replication (which is zero). Results are normalized to those for siNTC. Numbers under graphs represent individual siRNAs from the SMARTpool (siRNA 1, 2, 3, or 4) targeting a particular gene.

GPCR and 5 IC genes on A/CA/04/2009 replication were also determined (Fig. 3) . The results show that siRNAs targeting the AGTR1, HCRTR2, P2RY12, or GRID2 gene substantially reduced A/CA/04/2009 replication ( Fig. 3A and B ). Silencing P2RY12 also showed a considerable reduction in virus titers from those with siMAP2K (6.84-fold reduction) (Fig. 3A) . The result of individually silencing 16 GPCR genes and 5 IC genes on B/ Yamagata/16/1988 replication was also determined (Fig. 4) . Importantly, silencing the HRH2 or GRID2 gene substantially reduced the B/Yamagata/16/1988 titer, and targeting HRH2 resulted in a reduction in the virus titer greater than that with siMAP2K (30-fold reduction) gene silencing ( Fig. 4A and B) . These results confirm earlier results from the A/WSN/33 screen and show that several GPCR and IC genes affect A/CA/04/09 and B/ Yamagata/16/1988 replication.

RNAi screens have aided in the discovery of essential features of the host-virus Infectious viral titers (expressed as PFU per milliliter) and TCID 50 titers were determined by an MDCK plaque assay and sample titration on MDCK cells followed by an HA assay, respectively. Plaque assay data for GPCR (A) and ion channel (B) genes and TCID 50 data for GPCR (C) and ion channel (D) genes are presented as the inverse of the fold decrease from the level with nontargeting control siRNA (siNTC) for three independent experiments performed in triplicate. A positive increase in the fold change equates to a decrease in PFU per milliliter or TCID 50 per milliliter from that with siNTC. Data show means 6 standard errors of the means from three independent experiments performed in triplicate. Asterisks indicate significant differences from the control by ordinary one-way analysis of variance with Dunnett's multiple-comparison test (*, P , 0.05; **, P , 0.001; ***, P , 0.0001; ****, P , 0.00001). siNTC results are corrected to zero to reflect the baseline change in replication (which is zero). Results are normalized to those with siNTC. Numbers under graphs represent individual siRNAs from the SMARTpool (siRNA 1, 2, 3, or 4) targeting a particular gene.

interface, specifically the host pathways used to facilitate virus replication (23, 54) , and have provided information used to develop disease intervention strategies (28, 29) . GPCRs and ICs are implicated in the replication mechanisms of several RNA viruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), Marburg virus, Ebola virus, and HIV, but have not been well described for influenza virus (38, 40, 41, 46, 55) . In this study, we identified GPCR and IC genes used by influenza virus for replication and determined influenza virus strain and type differences. We screened 390 GPCR and 349 IC genes, of which 19 GPCR and 13 IC genes were selected for validation studies. Secondary validation by siRNA pool deconvolution yielded 16 confirmed GPCR genes (ADGRF1, ADORA1, ADRB2, AGTR1, C5AR2, CCKBR, FFAR1, HCAR3, HCRTR2, HRH2, HTR1B, LGR4, LPAR3, OXGR1, OXTR, and P2RY12) and 5 IC genes (ASIC1, GABRA3, GRID2, MCOLN2, and SCNN1D) (Fig. 1) . The genes from the RNAi screen were validated by using two individual OTP-siRNAs and testing the effects on A/WSN/33 replication using a higher MOI (0.01) to ensure robust infection. These studies used two endpoints to evaluate the effects of knockdown on influenza virus replication: infectious virus titers (expressed in PFU per milliliter), quantitated by plaque assays, and the amount of virus required to infect 50% of cells (50% tissue culture infective dose [TCID 50 ]), measured by HA assays (Fig. 2 ). siRNA silencing of GPCR genes LGR4, LPAR3, and OXGR1, and silencing of IC genes ASIC1, GABRA3, and MCOLN2, in A549 cells yielded substantial decreases in A/WSN/33 titers, showing that these genes are needed for A/WSN/33 replication. Of note, the decreases in virus plaque numbers were greater than those with the control siRNA siMAP2K (4-fold decrease), which targets mitogenactivated protein kinase, shown to be required for influenza virus replication (49, 50) . To examine influenza virus strain differences, siRNA-transfected A549 cells were infected with A/CA/04/2009, a representative circulating strain of human influenza A virus, and levels of virus replication were determined by quantification of infectious virus (by plaque assay) and determination of the TCID 50 following transfection (Fig. 3) . Silencing of GPCR and IC genes gave results similar to those for A/WSN/33-infected A549 cells, where influenza virus titers linked to the GPCR genes AGTR1, HCRTR2, and P2RY12 and the IC gene GRID2 were considerably reduced. Of note, silencing P2RY12 reduced virus titers 6-fold more than the siMAP2K control. We also examined the potential for influenza virus type differences linked to GPCR and IC genes in A549 cells by evaluating the replication of B/Yamagata/16/1988 after siRNA transfection (Fig. 4) . siRNA silencing of GPCR and IC genes also yielded reduced B/Yamagata/16/1988 replication, but the reductions were statistically significant (P , 0.01) only for the HRH2 and GRID2 genes; targeting HRH2 yielded a reduction in virus titers greater than that with siMAP2K (30-fold change).

The results suggest that influenza virus strains and types coopt similar GPCR and IC genes as part of the replication process in A549 cells but have the ability to utilize different genes in similar pathways (54, 61) . It has been reported that the tempo of signal transduction and host gene expression is associated with viral replication and virus production dynamics (61) . It is possible that different host genes are used for influenza virus replication in other cell types, particularly since transformed cell lines can have distinct gene expression (62) . This is a caveat with A549 cells, since some host genes identified as important may not translate to primary cell cultures. Additionally, the findings in this study were limited to 48 h postinfection (p.i.) due to the high-throughput screening procedure, and the later phases of virus replication were not evaluated. Additionally, GPCR signaling is a complex network; each GPCR complex may have a number of isoforms and splice variants that create hundreds of combinations of G proteins. Thus, differences in cell signaling associated with the kinetics of infection and/or GPCR isoforms/splice variants can go unnoticed (34) . In addition, the configuration of the G protein affects not only which transmembrane receptor it can bind to but also which downstream target is affected (34, 63, 64) . GPCR Ga subunits are grouped into four families (Ga s , Ga i , Ga q , and Ga 12/13 ) based on sequence homology, consisting of approximately 20 distinct Ga subunit proteins due to splice variants (65) . The host genes ADORA1, AGTR1, HTR1B, and P2RY12 are coupled to Ga i (Fig. 5) , while the ADRB2, HCAR3, and HRH2 genes are coupled to Ga s (Fig. 6) . Ga i signaling inhibits adenyl cyclase, which decreases intracellular cAMP levels, while Ga s signaling stimulates adenyl cyclase, prompting the opposing effect. Modulation of cAMP levels regulates the duration and intensity of cAMP signaling via feedback mechanisms (66) . G proteins have been implicated in late stages of influenza virus infection, specifically virus budding (67) (68) (69) . The host genes AGTR1, CCKBR, FFAR1, HCRTR2, OXGR1, and OXTR were associated with Ga q signaling by IPA (Fig. 7) . Ga q signaling is associated with multiple downstream pathways, but the best characterized are those associated with phospholipase Cb (PLC) activation and phosphatidylinositol 3-kinase (PI3K) (70) . Alteration of this pathway has been shown to play a regulatory role in the clathrin-mediated and clathrin-independent endocytosis pathways utilized by influenza virus at entry (71). The host genes ADGRF1 and LGR4 are orphan receptors, with no identified endogenous ligand (72) (73) (74) . C5AR2 is a nonclassical GPCR, and although it is a seven-transmembrane receptor, it does not couple to a G protein and instead binds b-arrestins (75, 76) . In this study, we show that siRNA silencing of the GPCR genes AGTR1, CCKBR, FFAR1, HCRTR2, OXGR1, and OXTR inhibits A/WSN/33, A/CA/04/2009, and B/Yamagata/16/1988 replication in A549 cells.

IPA of the validated IC genes determined in this study suggested that several genes affected influenza virus replication. ASIC1 is an acid-sensing sodium channel gene whose regulation is controlled by activation of the PKC pathway (77); however, it remains unclear how ASIC1 is necessary for viral replication. Similarly, GRID2 (or GluRd 2) is an orphan glutamate receptor gene whose function is poorly understood (78) . SCNN1D (the delta subunit of the epithelial sodium channel [d ENaC]) is one of four subunits that compose the epithelial sodium channel located on the apical surfaces of polarized tissues, e.g., the lung. It is involved in Na 1 transport across the transepithelial surface during Na 1 reabsorption (60, 79) . In this study, silencing of SCNN1D reduced virus replication, suggesting a novel role for this subunit compared to its a, b, and g counterparts (80). GABRA3 has been shown to be expressed in the lung (81) , and its activation is linked to autophagy (81), a strategy used by influenza viruses to promote replication (82) . We show that GABRA3 silencing reduces viral replication. It has been shown that MCOLN2 is associated with improved influenza virus, dengue virus, yellow fever virus, and equine arteritis virus infectivity (83) , possibly by promoting virus trafficking between the early and late endosomes and releasing virus FIG 7 Ga q signaling pathway generated by IPA. The AGTR1, CCKBR, FFAR1, HCRTR2, OXGR1, and OXTR genes were associated with Ga q signaling. RGS, regulators of G protein signaling; c-RAF, RAF proto-oncogene serine/threonine-protein kinase; MEK 1/2, mitogen-activated kinases 1 and 2; ERK 1/2, extracellular signal-regulated kinase; PIP2, phosphatidylinositol biphosphate; IP3, inositol triphosphate; DAG, diacylglycerol; PKC, protein kinase C; PLD, phospholipase D; PA, phosphatidic acidic; PC, phosphatidylcholine; CALM, clathrin assembly lymphoid myeloid leukemia protein; NFATc, nuclear factoractivated T cells, cytoplasmic; PYK2, Tau tyrosine kinase; PI3K, phosphatidylinositol 3-kinase; AKT, protein kinase B; IKK, IκB kinase; NF-κB, nuclear factor kappa-light-chain enhancer of activated B cells; RhoGEF, Rho guanine nucleotide exchange factor; ROCK, Rho-associated protein kinase; CSK, tyrosine protein kinase; GSK3b, glycogen synthase kinase-3b; PLCb, phospholipase Cb; BTK, Bruton tyrosine kinase. into the cytosol independently of interferon (IFN) signaling (83) . Our findings concur, showing that siRNA silencing of MCOLN2 decreases influenza virus replication and that MCOLN2 is an important host factor not only for the replication of IAVs but also for that of IBVs, which was not previously known.

Understanding the host factors used by influenza virus during entry, replication, and egress can help identify targets for drug repurposing or for the development of novel antiviral drugs. Targeting of host factors is refractory to the development of drug resistance generated by viral mutations (18) . Here, we identify several GPCR and ion channel genes that can be targeted by FDA-approved drug antagonists and/or inhibitors (Table 3) . For example, P2RY12 (a GPCR gene) can be targeted by the drug clopidogrel bisulfate (Plavix), which is currently approved for the inhibition of platelet aggregation and the treatment of patients with acute coronary syndrome (84) (85) (86) . Interestingly, the AGTR1 gene (a GPCR gene) has been shown to be associated with the coronavirus infection pathway, which has a possible link between angiotensin-converting enzyme 2 (ACE2) and lung injury (127) (128) (129) . AGTR1 can be targeted by angiotensin receptor blockers (ARBs), including candesartan, which has been suggested as a treatment for coronavirus disease 2019 (COVID-19) (87) . ARBs have shown efficacy in decreasing lung injury in animal models of acute respiratory distress syndrome (ARDS), but not without potential side effects (88) . Further studies are needed to determine the importance of this association with COVID-19. The ion channels ASIC1 and SCNN1D can be inhibited by amiloride, which has been shown to suppress the replication of coxsackievirus B3 (CVB3) and foot-and-mouth disease virus (FMDV) (89, 90) . These examples show the therapeutic potential of drug repurposing to target host factors needed for virus replication.

To summarize, this study identified and evaluated GPCR and IC genes coopted by influenza viruses (A/WSN/33, CA/04/2009, B/Yamagata/16/1988) for replication and identified strain and type differences. Collectively, the identification of these GPCR and IC genes provides the opportunity to develop host-directed virus control strategies to limit influenza virus replication and disease using drug repurposing or the development of novel antivirals.

Cells and viruses. Type II human lung epithelial (A549) cells (ATCC CCL-185) were propagated in Dulbecco's modified Eagle's medium (DMEM; HyClone, Logan, UT) supplemented with 5% heat-inactivated fetal bovine serum (HI-FBS; Atlas Biologics Inc., Fort Collins, CO). Madin-Darby canine kidney (MDCK) cells (ATCC CCL-34) were propagated in DMEM supplemented with 5% HI-FBS. All experiments were performed using log-phase A549 or MDCK cells.

A/WSN/33 (H1N1; ATCC VR-825), which is lab adapted and trypsin independent (38, 39) , A/CA/04/ 2009 (H1N1; BEI Resources), and B/Yamagata/16/1988 (BEI Resources) were grown in 9-day-old embryonated chicken eggs as described previously (91) . The A/WSN/33 and A/CA/04/2009 viruses used in siRNA validation and miR studies were propagated in MDCK cells (91) . Viral titers were determined by plaque assays and were calculated using the Reed and Muench method (92) (93) (94) . siGENOME screen. siGENOME plates received from Dharmacon/Horizon Discovery were preloaded with 0.5 nmol of pooled, lyophilized siRNAs targeting 390 GPCR or 349 IC genes. siRNAs were designed to ensure $85% knockdown of target gene expression, and optimal antisense-strand RISC loading is guaranteed (95) . siRNA pools were resuspended in siRNA resuspension buffer to a concentration of 1 mM, aliquoted, and stored at 280°C until use. For the screen, A549 cells were reverse transfected with siRNA SMARTpools or siRNA controls (50 nM) and were incubated at 37°C under 5% CO 2 for 48 h to allow for silencing of the targeted gene prior to virus infection as described previously (24, 30) . Briefly, transfections were performed in a 96-well plate format in triplicate. The siRNA SMARTpools were diluted in Hanks' balanced salt solution (HBSS; GIBCO), added to the plate, and incubated at room temperature (RT) for 5 min. Following incubation, 0.4 ml of DharmaFECT 1 transfection reagent (Horizon Discovery) and 9.6 ml of HBSS per well were added, and the mixture was incubated for 20 min at RT. Lastly, 80 ml containing 1.5 Â 10 4 A549 cells in DMEM supplemented with 5% HI-FBS was added to each well, and the mixture was incubated at 37°C under 5% CO 2 for 48 h. After transfection, the cells were washed twice with phosphate-buffered saline (PBS), infected with A/WSN/33 at an MOI of 0.001 to reduce defective interfering particles, and incubated at 37°C under 5% CO 2 for 48 h. After infection, the supernatant was collected and analyzed by a TCID 50 assay for virus replication by HA titers as described previously (24) . HA titer results were normalized to those with siNTC. A primary screen was performed twice in two independent experiments. Results were pooled and analyzed. All RNA interference (RNAi) experiments were completed according to the Minimum Information about an RNAi Experiment (MIARE) guidelines (96) .

Host genes with a Z-score of #21.0 were considered proinfluenza because siRNA silencing reduced virus replication from that with nontargeting controls. A total of 185 GPCR genes and 173 IC genes were proinfluenza genes (94) . These genes were evaluated by Ingenuity Pathway Analysis (IPA; Ingenuity Systems, Inc., Redwood City, CA) and Gene Ontology (GO) analysis. Comprehensive gene interaction networks were determined by combining IPA and GO analysis to identify relationships, functions, mechanisms, and pathways. Following IPA and GO analysis, the GPCR and IC host genes identified were evaluated for their abilities to be targeted by miRs. These data were used to select 19 proinfluenza GPCR and 13 proinfluenza ion channel candidates for further examination.

siRNA pool deconvolution and validation. The four siRNAs per SMARTpool were individually examined in a deconvolution assay to eliminate false-positive results and to determine the most effective siRNAs for reducing influenza virus replication. Plates containing 0.5 nmol of individual lyophilized ON-TARGETplus (OTP) siRNAs (Horizon Discovery) against a single host gene target were tested. OTP-modified siRNAs contained a modification within seed regions to reduce off-target effects and to increase selectivity and effectiveness. siRNAs were suspended in siRNA buffer according to the manufacturer's recommendations to a concentration of 1 mM, aliquoted, and stored at -80°C until use.

A549 cells were reverse transfected with one of four OTP-siRNAs as described elsewhere (44) . Briefly, siRNAs targeting a given GPCR or IC gene (Table 4) , nontargeting control siRNA (siNTC), siMAP2K (siRNA targeting the mitogen-activated protein kinase 1 gene), or the RNAi transfection control siTOX was used at a final concentration of 50 nM, and transfected cells were incubated at 37°C under 5% CO 2 for 48 h to allow for gene silencing prior to infection. Transfections were performed in a 96-well plate in triplicate. Briefly, siRNA reverse transfection was done using 0.4% DharmaFECT 1 transfection reagent, where siRNA was preincubated with DharmaFECT 1 in serum-free DMEM at RT for 20 min. A549 cells were suspended in DMEM supplemented with 5% HI-FBS, and 1.5 Â 10 4 cells were added to each well. Transfection plates were incubated at 37°C under 5% CO 2 for 48 h. After transfection, the medium was decanted, and the cells were washed twice with PBS and then infected with A/WSN/33 (MOI, 0.001) diluted in infection medium (MEM plus 0.3% bovine serum albumin [BSA] plus 1 mg/ml L-(tosylamido-2- (45, 46) ; and siTOX was used to confirm siRNA transfection under transfection conditions. Following incubation, supernatants were collected and stored at 280°C until they were tested by plaque assays. For the selected gene targets, the two siRNAs that gave the greatest reduction in virus titers were used for all remaining studies. Validated hits. A549 cells were transfected with individual OTP-siRNAs (2 siRNAs/gene target) from the deconvolution screen or with a control siRNA (siNTC, siMAP2K, or siTOX) at a final concentration of 50 nM in triplicate. Following transfection, the cells were infected with either A/WSN/33 (MOI, 0.01), A/CA/04/2009 (MOI, 0.1), or B/Yamagata/16/1988 (MOI, 0.1). The MOIs mediated low or no cytopathic effect (CPE). Following incubation, supernatants were removed and stored at 280°C until they were tested by plaque assays and TCID 50 assays. Two independent experiments were performed.

Cytotoxicity assay. Any cytotoxic effects associated with siRNA silencing were determined using a ToxiLight BioAssay kit (Lonza, Rockland, ME). Results were normalized to those with the siTOX transfection control, which results in complete cell death 48 h posttransfection. SMARTpools were considered toxic if transfection resulted in luminescence equivalent to $20% of the luminescence of the siTOX control.

Plaque assay. Infectious virus titers were determined by plaque assays as described elsewhere (47, 48, 91) . Briefly, supernatants were serially diluted 10-fold in MEM with 1 mg/ml TPCK-trypsin and were inoculated onto 90% confluent MDCK cell monolayers in 12-well tissue culture plates (Corning Costar, Cambridge, MA). The virus was adsorbed for 1 h at 37°C under 5% CO 2 before the addition of 3 ml of overlay. The overlay medium contained 1 part liquid medium containing 10Â MEM supplemented with 200 mm L-glutamine (Gibco), HEPES solution (Gibco), 7.5% NaCHO 3 (Gibco), penicillinstreptomycin-amphotericin B solution (Gibco), and 1 part 2.4% Avicel (FMC BioPolymer, Philadelphia, PA) in water or 1 part 1% agarose in water. Samples from A/WSN/33 or A/CA/04/2009 wells were incubated at 37°C under 5% CO 2 for 3 days. B/Yamagata/16/1988 was incubated at 37°C under 5% CO 2 for 5 days to allow for better plaque formation. The overlays were removed, the plates were washed twice with PBS, and the cell monolayers were fixed with acetone-methanol (80:20) for 20 min at RT. Following fixation, the plates were stained with crystal violet as described previously, and viral titers were determined (92, 93) . TCID 50 assay. Endpoint titers were determined by a TCID 50 assay as described previously (22, 25, 92) . Briefly, supernatants collected from influenza virus-infected A549 cells were serially diluted 10-fold in triplicate on MDCK cells in 96-well plates. Influenza virus-infected MDCK plates were incubated 5 days using cell culture conditions described elsewhere (22, 25) . Following incubation, an HA test was performed using 50 ml of supernatant from infected MDCK cells and 50 ml of 1% turkey red blood cells (RBC) for a final concentration of 0.5% in a round-bottom plate (97) . The TCID 50 titers were calculated using the Reed and Muench method (92) .

Hemagglutination assay. Hemagglutination was used for viral diagnosis of influenza viruses (92, 97) . Briefly, 2-fold serial dilutions of virus in PBS were dispensed into individual wells of a 96-well roundbottom microtiter plate (Corning Costar, Cambridge, MA). Then aliquots of turkey RBC were added to each well to 0.5% of final volume. The highest dilution at which clumping was observed was regarded as the HA titer of the sample.

Statistics. HA assay results were normalized to results for siNTC-transfected controls. The nontargeting control was set to an arbitrary value of 1. Genes were specified a Z-score, calculated as Z ¼ ðx 2 mÞ=ðs= ffiffiffi n p Þ, where x is equal to the average HA value of each gene, m is equal to the population mean of the HA, s is equal to the standard deviation of each gene across the two independent experiments, and n is equal to the number of genes within the populations (16) . Genes in the primary screen that were ,1.5 standard deviations from the plate mean in both duplicates were considered primary hits. 

An overlapping protein-coding region in influenza A virus segment 3 modulates the host response

Evolutionary conservation of the PA-X open reading frame in segment 3 of influenza A virus

Overlapping signals for translational regulation and packaging of influenza A virus segment 2

A complicated message: identification of a novel PB1-related protein translated from influenza A virus segment 2 mRNA

Identification of a novel viral protein expressed from the PB2 segment of influenza A virus

Uncovering the potential pan proteomes encoded by genomic strand RNAs of influenza A viruses

Discordant antigenic drift of neuraminidase and hemagglutinin in H1N1 and H3N2 influenza viruses

Influenza vaccine: the challenge of antigenic drift

World Health Organization

Influenza (seasonal)

Estimating the burden of 2009 pandemic influenza A (H1N1) in the United States

Comparing influenza vaccine efficacy against mismatched and matched strains: a systematic review and meta-analysis

influenza vaccine effectiveness in the United States by vaccine type

A physical and regulatory map of host-influenza interactions reveals pathways in H1N1 infection

The Ras-PI3K signaling pathway is involved in clathrin-independent endocytosis and the internalization of influenza viruses

Bivalent role of the phosphatidylinositol-3-kinase (PI3K) during influenza virus infection and host cell defence

Development of cellular signaling pathway inhibitors as new antivirals against influenza

Reassortment and mutations associated with emergence and spread of oseltamivir-resistant seasonal influenza A/H1N1 viruses in 2005-2009

Surveillance for neuraminidase inhibitor resistance among human influenza A and B viruses circulating worldwide from

Antivirals in seasonal and pandemic influenza-future perspectives

RNA interference

Identification of host kinase genes required for influenza virus replication and the regulatory role of microRNAs

Host gene targets for novel influenza therapies elucidated by high-throughput RNA interference screens

MicroRNA regulation of human protease genes essential for influenza virus replication

Verdinexor, a novel selective inhibitor of nuclear export, reduces influenza A virus replication in vitro and in vivo

RNA interference inhibits respiratory syncytial virus replication and disease pathogenesis without inhibiting priming of the memory immune response

Genome-wide RNAi screen identifies human host factors crucial for influenza virus replication

siRNA genome screening approaches to therapeutic drug repositioning

Development of RNAi technology for targeted therapy -a track of siRNA based agents to RNAi therapeutics

A universal mammalian vaccine cell line substrate

Engineering enhanced vaccine cell lines to eradicate vaccine-preventable diseases: the polio end game

The miR-35-41 family of microRNAs regulates RNAi sensitivity in Caenorhabditis elegans

GPCR signalling complexes: synthesis, assembly, trafficking and specificity

Viral tropism, fitness and pathogenicity of HIV-1 subtype C

Use of alternate coreceptors on primary cells by two HIV-1 isolates

Primary human immunodeficiency virus type 2 (HIV-2) isolates, like HIV-1 isolates, frequently use CCR5 but show promiscuity in coreceptor usage

Inhibition of Ebola and Marburg virus entry by G protein-coupled receptor antagonists

Ion channels as therapeutic targets: a drug discovery perspective

Modulation of potassium channels inhibits bunyavirus infection

Inhibition of herpes simplex virus type 1 entry by chloride channel inhibitors tamoxifen and NPPB

Modulation of influenza virus replication by alteration of sodium ion transport and protein kinase C activity

Chloride and potassium channel function in alveolar epithelial cells

Biological phenotype of human immunodeficiency virus type 1 clones at different stages of infection: progression of disease is associated with a shift from monocytotropic to T-cell-tropic virus populations

Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply-exposed individuals to HIV-1 infection

Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene

A genome-wide siRNA screen reveals diverse cellular processes and pathways that mediate genome stability

Position-specific chemical modification of siRNAs reduces "off-target" transcript silencing

Influenza virus propagation is impaired by inhibition of the Raf

MEK inhibition impairs influenza B virus propagation without emergence of resistant variants

Quantifying mechanistic traits of influenza viral dynamics using in vitro data

One-step growth curves of various strains of influenza A and B viruses and their inhibition by inactivated virus of the homologous type

Uncovering the global host cell requirements for influenza virus replication via RNAi screening

Viral hijacking of G-proteincoupled-receptor signalling networks

Receptor binding and membrane fusion in virus entry: the influenza hemagglutinin

Uncoating of influenza virus in endosomes

Drugs, their targets and the nature and number of drug targets

Gene Summary: SCNN1D -sodium channel epithelial 1 delta subunit (human), on NCBI

Differential activation of host cell signalling pathways through infection with two variants of influenza A/Puerto Rico/8/34 (H1N1) in MDCK cells

Comparative proteomic phenotyping of cell lines and primary cells to assess preservation of cell type-specific functions

Reviews in molecular biology and biotechnology: transmembrane signaling by G protein-coupled receptors

G-protein-coupled receptor oligomerization and its potential for drug discovery

The G protein subunit gene families

Integrating signals between cAMP and the RAS/RAF/MEK/ERK signalling pathways. Based on the anniversary prize of the Gesellschaft für Biochemie und Molekularbiologie Lecture delivered on 5

Role of G protein and protein kinase signalling in influenza virus budding in MDCK cells

Receptor and protein kinase C-mediated regulation of ARF binding to the Golgi complex

Activators of protein kinase A stimulate apical but not basolateral transport in epithelial Madin-Darby canine kidney cells

Endothelin-1 and endothelin-3 stimulate calcium mobilization by different mechanisms in vascular smooth muscle

A Ca 21 -dependent signalling circuit regulates influenza A virus internalization and infection

Orphan receptor GPR110, an oncogene overexpressed in lung and prostate cancer

Novel human G protein-coupled receptors with long N-terminals containing GPS domains and Ser/Thr-rich regions

R-spondins function as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/betacatenin signaling

C5L2, a nonsignaling C5A binding protein

C5a-stimulated recruitment of beta-arrestin2 to the nonsignaling 7-transmembrane decoy receptor C5L2

Protein kinase C isoform antagonism controls BNaC2 (ASIC1) function

The d 2 glutamate receptor: a key molecule controlling synaptic plasticity and structure in Purkinje cells

Regulation of the epithelial sodium channel (ENaC) by membrane trafficking

SARS-CoV proteins decrease levels and activity of human ENaC via activation of distinct PKC isoforms

Gammaaminobutyric acid A receptor alpha 3 subunit is overexpressed in lung cancer

Cellular response to influenza virus infection: a potential role for autophagy in CXCL10 and interferon-alpha induction

Mucolipin-2 cation channel increases trafficking efficiency of endocytosed viruses

Clopidogrel: interactions with the P2Y12 receptor and clinical relevance

The active metabolite of clopidogrel disrupts P2Y12 receptor oligomers and partitions them out of lipid rafts

The P2Y12 receptor as a therapeutic target in cardiovascular disease

Candesartan could ameliorate the COVID-19 cytokine storm

Recombinant human ACE2: acing out angiotensin II in ARDS therapy

Antiviral effect of amiloride on replication of foot and mouth disease virus in cell culture

Acidinduced pain and its modulation in humans

Avian influenza virus isolation and propagation in chicken eggs

A simple method of estimating fifty percent endpoints

Influenza virus titration, antigenic characterization, and serological methods for antibody detection

Plaque formation by influenza viruses in the presence of trypsin

On the art of identifying effective and specific siRNAs

Increasing the robustness and validity of RNAi screens

The quantitative determination of influenza virus and antibodies by means of red cell agglutination

Xanthine derivatives as antagonists at A 1 and A 2 adenosine receptors

Adenosine receptors: development of selective agonists and antagonists

Immunomodulatory properties of pentoxifylline are mediated via adenosine-dependent pathways

Ability of the new AT1 receptor blocker azilsartan to block angiotensin II-induced AT1 receptor activation after wash-out

TTD: Therapeutic Target Database

Long-lasting angiotensin type 1 receptor binding and protection by candesartan: comparison with other biphenyl-tetrazole sartans

Comparison of the effects of AT1 receptor blockade and angiotensin converting enzyme inhibition on atherosclerosis

Irbesartan: review of pharmacology and comparative properties

Pharmacological profile of valsartan: a potent, orally active, nonpeptide antagonist of the angiotensin II AT

Asenapine: a novel psychopharmacologic agent with a unique human receptor signature

Efficacy and safety analysis of new P2Y12 inhibitors versus clopidogrel in patients with percutaneous coronary intervention: a meta-analysis

P2Y receptor antagonists in thrombosis

Molecular modeling of purinergic receptor P2Y12 and interaction with its antagonists

Pharmacokinetics of clopidogrel after administration of a high loading dose

The reversible P2Y antagonist cangrelor influences the ability of the active metabolites of clopidogrel and prasugrel to produce irreversible inhibition of platelet function

Ticagrelor: pharmacokinetic, pharmacodynamic and pharmacogenetic profile: an update

Beta 2-adrenoceptor-mediated intrinsic sympathomimetic activity of carteolol: an in vivo study

Binding of beta-adrenoceptor antagonists to rat and rabbit lung: special reference to levobunolol

How many drug targets are there?

Binding of (-)-[3H]-CGP12177 at two sites in recombinant human beta 1-adrenoceptors and interaction with beta-blockers

Comparisons of beta-adrenergic blocking properties of S-and R-timolol in humans

Lafutidine, a newly developed antiulcer drug, elevates postprandial intragastric pH and increases plasma calcitonin gene-related peptide and somatostatin concentrations in humans: comparisons with famotidine

Clinical pharmacology of famotidine: a summary

Nonsteroid antiinflammatory drugs inhibit both the activity and the inflammationinduced expression of acid-sensing ion channels in nociceptors

Transforming growth factor-b downregulates sGC subunit expression in pulmonary artery smooth muscle cells via MEK and ERK signaling

The oxytocin receptor antagonist atosiban inhibits cell growth via a "biased agonist" mechanism

The oxytocin receptor antagonist, Atosiban, activates pro-inflammatory pathways in human amnion via G ai signalling

Evans blue is a specific antagonist of the human epithelial Na 1 channel delta-subunit

Effects of the serine/threonine kinase SGK1 on the epithelial Na 1 channel (ENaC) and CFTR: implications for cystic fibrosis

Candesartan could ameliorate the COVID-19 cytokine storm

Targeting the renin-angiotensin signaling pathway in COVID-19: unanswered questions, opportunities, and challenges

A hypothesis for pathobiology and treatment of COVID-19: the centrality of ACE1/ACE2 imbalance

We would like to acknowledge funding in part from NIAID CEIRS contract numbers HHSN266200700006C and HHSN2722014000004C to R.T. and from the Georgia Research Alliance to R.T.