Peptide-Mediated Interference of PB2-eIF4G1 Interaction Inhibits

Jun 7, 2016 - Influenza viruses are obligate parasites that hijack the host cellular system. Previous results have shown that the influenza virus PB2 ...
0 downloads 7 Views 5MB Size
Download PDF
Letter pubs.acs.org/journal/aidcbc

Peptide-Mediated Interference of PB2-eIF4G1 Interaction Inhibits Influenza A Viruses’ Replication in Vitro and in Vivo Shuofeng Yuan,† Hin Chu,† Jiahui Ye,† Meng Hu,† Kailash Singh,‡ Billy K. C. Chow,‡ Jie Zhou,† and Bo-Jian Zheng*,† †

Department of Microbiology, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China School of Biological Sciences, Faculty of Science, The University of Hong Kong, Hong Kong SAR, China

‡

S Supporting Information *

ABSTRACT: Influenza viruses are obligate parasites that hijack the host cellular system. Previous results have shown that the influenza virus PB2 subunit confers a dependence of host eukaryotic translation initiation factor 4-γ 1 (eIF4G1) for viral mRNA translation. Here, we demonstrated that peptidemediated interference of the PB2-eIF4G1 interaction inhibited virus replication in vitro and in vivo. Remarkably, intranasal administration of the peptide provided 100% protection against lethal challenges of influenza A viruses in BALB/c mice, including H1N1, H5N1, and H7N9 influenza virus subtypes. Mapping of the PB2 protein indicated that the eIF4G1 binding sites resided within the PB2 cap-binding domain. Virtual docking analysis suggested that the inhibitory peptide associated with the conserved amino acid residues that were essential to PB2 cap-binding activity. Overall, our results identified the PB2-eIF4G1 interactive site as a druggable target for influenza therapeutics. KEYWORDS: influenza virus, antiviral peptide, host−virus interaction, eIF4G1, PB2, Tat

T

On the other side, the mammalian eIF4G1 protein plays a pivotal role in the initiation of cellular translation, which mediates the cap-dependent assembly of the preinitiation complex.6 Influenza viruses are obligate parasites that hijack the host system and reconstruct cellular networks for their own propagation. Previous investigations have revealed that the influenza virus nonstructural protein NS1 recruited eIF4G1 specifically to allow the preferential translation of the viral mRNAs7 and that the NS1 eIF4G1 binding-domain-truncated mutants were attenuated with respect to viral replication in vitro and in vivo.8 Subsequently, the association between virus PB2 and host eIF4G1 during active viral replication dynamics has been reported by two independent groups,9,10 in which such an interaction has also been demonstrated to contribute to the optimal synthesis of viral proteins during the virus life cycle. We proposed that the peptides derived from the eIF4G1 interaction sites would compete for the PB2-eIF4G1 interaction and thus disrupt viral mRNA translation. To facilitate cell internalization, additional C-terminal amino acids (YGRKKRRQRRRPP) from the human immunodeficiency virus Tat protein were fused to the peptides.11 Our results demonstrated that one of the peptides, designated Mouse-eIF4G1-B-Tat, exhibited potent antiviral efficacy in vitro and in vivo. We further determined that the PB2 residues that mediated the association with eIF4G1 resided within the PB2 cap-binding domain (PB2cap).

he increasing use of licensed antivirals has resulted in the global emergence of amantadine- and/or oseltamivirresistant strains of influenza virus and the occasional isolations of peramivir- or zanamivir-resistant influenza viruses, which are exemplified by the worldwide spread of adamantine-resistant A(H3N2) viruses since 2003, oseltamivir-resistant seasonal A(H1N1) viruses since 2007, adamantane-resistant pandemic A(H1N1) viruses in 2009, and peramivir-resistant A(H7N9) viruses in 2013.1,2 Given that the current anti-influenza drugs act directly on inhibiting the biological function of viral proteins, which involves mostly enzymatic activities, host−virus interactions that involve the viral replication machinery are attractive targets for antiviral therapy.3 The progress in our understanding of influenza virus−host interactomes over the past decade has inspired alternative strategies and novel targets of treating viral infection. Importantly, it is a promising approach that is highlighted by the advantages of (i) delayed emergence of drug resistance and (ii) potentially broadspectrum antiviral effects.4 However, the potential side effects of host-targeting antivirals remain a major concern, especially for those host targets crucial to cellular viability. In this regard, we conceived that blocking the interactive surface of viral proteins would minimally evoke cytotoxicity while remaining capable of disturbing virus−host interactions and subsequently suppressing virus propagation. To verify the hypothesis, we examined the interplay between the influenza A PB2 protein and host eukaryotic translation initiation factor 4-γ 1 (eIF4G1). The PB2 subunit of influenza virus polymerase is responsible for cap binding, nuclear localization, and host adaptation.5 © XXXX American Chemical Society

Received: April 21, 2016

A

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

Letter

Overall, the findings hinted at a new avenue for the development of novel anti-influenza agents. Two regions on the eIF4G1 protein have been identified previously as the PB2 interaction sites,9 termed A and B. Sequence alignments between human and mouse eIF4G1 protein region A and region B reveal that there are three and five substitutions in each region, respectively (Figure 1a). To promote cell entry, a

dose-dependent manner, and Human B peptide exhibited a significantly stronger antiviral efficacy (Figure 2a). We then explored whether Human B could provide cross-protection against infections of different subtypes of influenza virus. The result showed that Human B inhibited the viral replication of all tested influenza virus subtypes, including H3N2, H5N1, H7N7, H7N9, and H9N2, in a dose-dependent manner (Figure 2b) in which its half maximal effective concentration (EC50) ranges from 29.7 to 72.5 μM (Supporting Information Table 1). In addition, low cytotoxicity was detected when A549 cells were treated with Human B at concentrations of as high as 1 mM (Supporting Information Figure S1), which revealed its potential as an anti-influenza agent. To test if the antiviral peptides were host-selective, four peptides originating from either human or mouse (Figure 1a) were compared for their antiviral efficacies in mouse LA-4 cells. As shown in Figure 2c, Mouse-eIF4G1-A-Tat (Mouse A) and Mouse-eIF4G1-B-Tat (Mouse B) exhibited stronger antiviral efficacy than did Human A and Human B in LA-4 cells (p < 0.01). The result suggested that the antiviral peptides were selective toward host specificity. Thus, we selected Mouse B peptide to evaluate its in vivo antiviral efficacy in a lethal mouse model. The cross-subtype antiviral effect of Mouse B against infections of different subtypes of influenza virus was evaluated in a murine lethal-infection model using H1N1, H5N1, and H7N9 viruses. BALB/c mice (14 mice/group) were intranasally (i.n.) inoculated with five lethal dose 50% (LD50) of the viruses, followed by i.n. delivery with Mouse B peptide at 6 h postinfection. In parallel, data from a group of mice qtreated with a similar concentration (7.5 mM) of zanamivir was collected as a positive control. As shown in Figure 3a,b, when the mice were challenged with a lethal dose of H1N1 or H5N1 influenza viruses, all mice treated with Mouse B or zanamivir survived (survival rate = 100%), whereas none was alive in the PBS-treated group (survival rate = 0%). Notably, when the mice were challenged with H7N9 influenza virus, the peptide-treated group showed higher survival (100%) and less body weight loss than did the zanamivir-treated group (60%) (p < 0.05), while all of the PBS-treated mice died (survival = 0%) (Figure 3c). We also measured the viral loads and examined the lung pathology of infected mice on day 5 postinfection (Figure 4). As shown in the bar charts, significantly reduced viral loads were detected in Mouse B-treated mouse lungs (p < 0.01 for H1N1 and H5N1 groups, p < 0.05 for the H7N9 group). In the mice treated with zanamivir, two to three logs of reduction in viral load were detected in the H1N1 and H5N1 groups, whereas a nonsignificant difference was found in H7N9-infected mice. Furthermore, pathological examinations revealed that Mouse B administration mitigated lung lesions in virus-infected mice, whereas the untreated mice showed apparent inflammatory infiltration and bronchopneumonia (right panel, Figure 4). Taken together, the results supported that the Mouse B peptide effectively inhibited influenza virus replication in vivo. Together with the previous findings by Yángüez and his colleagues,9 our study strongly proposed the important role of PB2-eIF4G1 association in the influenza virus life cycle. Thus, the PB2-eIF4G1 interaction might be an antiviral target for the development of agents that can block such contact. However, bacterially expressed PB2 is insoluble, and the yield of active enzyme produced in eukaryotic expression systems is prohibitively low, which restricts high-throughput screenings by

Figure 1. eIF4G1-derived peptides disrupted eIF4G1-PB2 interaction in vitro. (a) Alignment of the selected eIF4G1-PB2 interacting peptides from human and mouse hosts. Identical residues are in black, substitutions are in red, and cell-penetrating peptide Tat is labeled in green. (b) Immunoprecipitation (IP) for the detection of the eIF4G1-PB2 interaction. PB2-transfected HEK293T cells were cultured in the absence or presence of Human A or Human B peptide. Cell lysate was prepared 24 h post-transfection for the detection of endogenous eIF4G1 by anti-eIF4G1 (α eIF4G1) antibody, and overexpressed PB2, by anti-PB2 (α PB2) antibody. The lysates were further subjected to IP using a different anti-eIF4G1 antibody, followed by the detection of PB2 or eIF4G1 in the precipitated materials. Lane A, Human A-treated sample; lane B, Human B-treated sample; lane Input, transfected cell extract without peptide treatment; lane Control, nontransfected cell extract. Western blotting images are shown.

13-amino acid peptide, Tat, with a cell-penetrating function was linked to the C-terminus of each peptide. To confirm that these peptides could indeed interrupt PB2-eIF4G1 association, an immunoprecipitation assay was performed in which we transiently expressed viral PB2 protein in HEK293T cells, followed by treatment with either Human-eIF4G1-A-Tat (Human A) or Human-eIF4G1-B-Tat (Human B) peptide. As shown in Figure 1b, in the absence of the peptides (i.e., the lane of input) PB2 was overexpressed properly and immunoprecipitated PB2-eIF4G1 complex formed, whereas neither PB2 nor the PB2-eIF4G1 complex was detectable in the control panel. In the presence of the peptides, eIF4G1 protein could be detected in the precipitated materials. However, PB2 could not be coimmunoprecipitated with eIF4G1 protein even though each of them alone was expressed successfully. This result suggested that the examined peptides were able to enter the cell and block the PB2-eIF4G1 interaction. To investigate whether eIF4G1-derived peptides suppress influenza virus replication, we evaluated the antiviral effect of Human A and Human B peptides in A549 cells. Indeed, both peptides inhibited influenza virus H1N1 replication in a B

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

Letter

Figure 2. eIF4G1-derived peptides inhibited influenza A virus replication in vitro. (a) Antiviral effects of Human A and B were evaluated in A549 cells using multicycle virus growth assays. Cells were infected with H1N1 influenza virus at an MOI of 0.1. After virus internalization, the inoculum was removed, washed, and replaced with culture medium in the presence or absence of the indicated peptides. Virus titers in the cell supernatant were collected 24 h postinfection and determined by RT-qPCR methods. (b) The cross-subtype antiviral effect of Human B against infections of different subtypes of influenza viruses was evaluated with the same method as described in (a). The data were evaluated for statistical significance using one-way ANOVA. (c) The same procedures were used to examine the antiviral efficacy of the indicated peptides in mouse LA-4 cells, whereas virus titers in the cell supernatant were determined by plaque assay. The data were evaluated for statistical significance using the student’s t test. * p < 0.05 and ** p < 0.01. Shown are the means value of two independent experiments ± SD.

demonstrated that PB2cap, rather than the other two functional domains of the PB2 subunit, was key to PB2-eIF4G1 association. Surprisingly, the result was in disagreement with a previous finding that reported that amino acids 538−693 of PB2 within the 627-NLS domain mediated its binding with eIF4G1.9 The apparent discrepancy might be due to the conformational variations of target proteins between the two studies. For example, a representative peptide (biotinylated Human B) instead of the intact eIF4G1 protein was utilized as a detection probe in our study. In this regard, structural and functional analyses of the PB2-eIF4G1 co-crystal might be essential in future studies. Since prokaryotic expressed PB2cap is functional and proteolytically stable,13 high-throughput screening of novel PB2eIF4G1 inhibitors that target PB2cap becomes feasible. For example, the competitive binding assay, as shown in Figure 5b, could be further optimized to search small-molecule compounds that competitively displace the binding of biotin-Human B from PB2cap. Next, docking analysis was carried out to predict the interaction sites between Human B peptide and PB2cap. As shown in Figure 6, the peptide bound to key residues such as Ser 322, Phe 323, Ser 324, Lys 339, Arg 427, Asn 429, and His 432 that were located on PB2cap, in which strong hydrogen bonds were formed between Thr 7 (Human B) and Ser 324 (PB2cap) or between Gln 15 (Human B) and Arg 427 (PB2cap) (Figure 6b). PB2cap harbors a cap-binding activity pocket coordinated

using full-length PB2 as one of the binding partners. To address this issue, we investigated the minimal essential domain of PB2 that was responsible for PB2-eIF4G1 association. Influenza virus PB2 involves the cap snatching of viral polymerase and can be divided into three major functional domains. Residues 1−37 of PB2 contain an interface for PB1 binding,12 and residues 318−483 (i.e., the cap-binding domain (PB2cap)) are crucial to the recognition of cellular mRNAs via a 5′ cap structure.13 Remaining residues 538−760 (627-NLS domain) carry host- and pathogenicity-determinant lysine 62714 as well as the motif that enables PB2 nuclear importation.15 To determine which domain is essential for the PB2-eIF4G1 interaction, ELISA was carried out to measure the binding intensity, in which a biotinylated Human B peptide was used as “bait” to represent eIF4G1. As shown in Figure 5a, both PB1-binding and 627-NLS domains exhibited a negligible binding affinity for the bait, which appeared to be at the same level as that of negative controls. In contrast, PB2cap showed significant binding to biotin-Human B peptide in a dose-dependent manner. To validate the binding specificity, absorbance values were recorded in the presence or absence of unlabeled Human B peptide as a positive control inhibitor. At a fixed concentration of 1 μM PB2cap, dose-dependent inhibition was detected in which higher concentrations of Human B peptide resulted in lower binding signals (Figure 5b). The result suggested that the interaction between PB2cap and biotin-Human B was specifically inhibited. Taken together, we C

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

Letter

Figure 3. Efficacy of Mouse B peptide against different subtypes of influenza viruses in mice. Survival rates and body weight changes of mice infected by (a) H1N1, (b) H5N1, and (c) H7N9 influenza viruses are shown. Mice (10 per group) were infected with 5 LD50 of each virus and were treated with 20 μL of 25 mg/mL (∼7.5 mM) Mouse B peptide, 2.5 mg/mL (∼7.5 mM) zanamivir, or PBS by intranasal administration. Treatments started 6 h after virus challenge and continued for 5 doses in 2.5 days (2 doses/day). Survival rate differences between groups were compared using the log-rank (MantelCox) test; * indicates p < 0.05. Body weight changes and sick signals of each mouse were monitored daily for 14 days or until death. The body weight values are shown as means ± SD for the mice that were alive at each time point.

Figure 4. Viral loads and pathological findings in the lungs of infected mice. Mice were treated as described in the legend of Figure 3. (a) H1N1, (b) H5N1, and (c) H7N9. Four mice from each group were euthanized on day 5 postinfection, and lungs were collected for the detection of viral loads by plaque assay. The results are presented as bar charts with mean values + SD. Differences between groups were compared using the one-way ANOVA test. * indicates p < 0.05 and ** indicates p < 0.01 as compared to the PBS-treated group. Pathological findings in the lungs of infected mice were observed by H&E staining. Representative histologic sections of the lung tissues from the infected and uninfected mice (normal) are shown. In either influenza virus subtype, fewer inflammatory infiltrates or lesions of bronchopneumonia are observed in samples from mice treated with Mouse B peptide or zanamivir as compared to that from PBS-treated mice (magnification 200×). D

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

■

Letter

METHODS

Cells and Viruses. Human lung carcinoma A549 cells, Madin-Darby canine kidney (MDCK) cells, human embryonic kidney 293T (HEK293T) cells, and mouse lung adenoma LA-4 cells were used in the study. Upon virus infection, the infected cells were maintained in medium supplemented with 0.1 μg/mL TPCK trypsin and 0.25% FBS. A total of 7 strains/ 6 subtypes of influenza A virus [A/Hong Kong/415742/ 2009(H1N1), A/Hong Kong/1/1968 (H3N2), A/Vietnam/ 1194/2004(H5N1_VN), A/Hong Kong/156/97(H5N1_HK), A/Netherlands/219/2003(H7N7), A/Anhui/1/2013(H7N9), and A/HK/1073/1999 (H9N2)] were cultured in MDCK cells and used for in vitro antiviral testing. Tree mouse-adapted influenza virus strains A/Hong Kong/415742Md/2009 (H1N1), A/Vietnam/1194Md/2004(H5N1), and A/Anhui/1Md/2013(H7N9) were propagated in chick embryos and used for in vivo antiviral testing. All experiments with live viruses were conducted using biosafety level 2 or 3 facilities as described previously.18 Details of the cell culture and other biological materials are described in the Supporting Information. Immunoprecipitation Assay. The inhibitory effect of the Human A and Human B peptides to PB2-eIF4G1 binding was tested in the plasmid pCMV3-His-PB2 transfected cells using an immunoprecipitation assay. The protocol is outlined in the Supporting Information. Multicycle Virus Growth Assay. Multicycle virus growth assays were carried out to compare the antiviral efficacy of selected peptides in A549 or LA-4 cells and to evaluate the crosssubtype antiviral activity of Human B peptide. Experimental details are presented in the Supporting Information. Mouse Study. All experimental protocols were approved by the Animal Ethics Committee at the University of Hong Kong and were performed in compliance with the standard operating procedures of biosafety level-3 animal facilities. BALB/c female mice, 6−8 weeks old, were kept in biosafety level-3 housing and given access to standard pellet feed and water ad libitum. The regimen of the antiviral study wis fully specified in the Supporting Information. Mapping of PB2 Binding Site. Three functional domains of PB2 were prepared. The binding affinity of an individual domain for the representative Human B peptide was measured by enzyme-linked immunosorbent assay (ELISA). Details of protein expression, purification, and ELISA are described in the Supporting Information. Molecular Docking. The structure file of PB2cap bound with natural ligand m7GTP was retrieved from the RCSB protein data bank (PDB 4CB4).19 Co-crystallized ligands were removed by the PDB editor in order to perform docking analysis. The 3-D structure of Human B peptide was modeled through Pep-Fold server,20 followed by hydrogen molecule addition and energy minimization with Avogadro software.21 Molecular docking was carried out by utilizing the patchdock and firedock algorithms22,23 with the protein structure considered to be a rigid body while the ligand was fully flexible. One hundred docking solutions were computed for the ligand−protein interaction, and all other parameters were set to default. The solutions were visualized and validated by Schrodinger maestro (Schrödinger).24 Statistical Analysis. The data were evaluated for statistical significance using a one-way ANOVA or log-rank (Mantel-Cox) test or the student’s t test as indicated in the figure legends (Prism 6.0, GraphPad, Inc.). EC50 was calculated using Sigma

Figure 5. Mapping of the binding site on PB2. (a) Functional domains of PB2 were expressed with fusion tags (i.e., His-tagged PB2cap and GST-tagged 627-NLS) or synthesized with biotin labeled (PB1-binding) as specified in the methods. The binding capacity of individual domain to peptide biotin-Human B was measured with ELISA, in which a 1 μM peptide probe was coated per well, followed by the addition of an individual domain at various concentrations. The pET32a-blank and GST-blank proteins were included as negative controls. (b) Competitive binding assay were carried out using unlabeled Human B peptide as a positive control inhibitor. Concentrations of biotin-Human B probe and PB2cap protein were fixed at 1 μM, and unlabeled Human B peptide was serially diluted (200 to 3.125 μM) and incubated together with PB2cap protein. The experiments were conducted in triplicate and repeated twice for confirmation. Data are shown as the mean value ± SD.

by 13 conserved residues.16 Intriguingly, the Human B bound residues overlap with these amino acids that are essential for cap binding (e.g., Phe 323, Ser 324, Lys 339, Asn 429, and His 432). The result explained why Human B peptide was able to provide cross-protection against influenza virus infections (Figure 2b). In addition, investigation of whether Human B inhibited the virus cap-binding activity might be performed in a future study. The influenza virus is obliged to compete for host-cell factors and manipulate the cellular translation apparatus to ensure a high proficiency of viral protein synthesis.17 In this study, we identified the PB2-eIF4G1 binding site as a druggable target and revealed the capacity of the eIF4G1-derived peptides in blocking the PB2-eIF4G1 interaction so that they could efficiently inhibit the replication of influenza A viruses in vitro and in vivo. Also, the screening of small-molecule compounds that displace the binding of Human B peptide from PB2cap is feasible as an option to develop novel anti-influenza drugs. Furthermore, we infer that the investigation of other eIF4G1-mediated activity of other viruses (e.g., rotaviruses, adenoviruses, and enteroviruses) as therapeutic targets is appropriate.17 E

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

Letter

Figure 6. Docking simulation of Human B peptide with the influenza PB2cap domain. A virtual docking representation of the Human B peptide with PB2cap is shown with a ribbon diagram (a) and a ligplot diagram. (a) 3D interaction diagram depicts the peptide in red contacting amino acid residues Ser 322, Phe 323, Ser 324, Lys 339, Arg 427, Asn 429, and His 432 that are located on PB2cap. (b) In the two-dimensional analysis, interacting residues on PB2cap are shown in brick red, and key amino acids of the peptide ligand (i.e., Arg 2, Ala 4, Ser 6, Thr 7, Arg 10, Phe 11, and Gln 15) are represented in pink. Hydrogen bonds are shown as green dashes. Blue, red, and black balls represent the carbon, oxygen, and nitrogen atoms within a specific amino acid, respectively.

Plot 12.0 (SPSS). Values of p < 0.05 were considered to represent a statistically significant difference.

■

ACKNOWLEDGMENTS

■

REFERENCES

This study was supported in part by the Innovation and Technology Commission, Government of Hong Kong SAR (UIM/278).

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsinfecdis.6b00064.

■

■

(1) Hayden, F. G., and de Jong, M. D. (2011) Emerging influenza antiviral resistance threats. J. Infect. Dis. 203, 6−10. (2) Hai, R., Schmolke, M., Leyva-Grado, V. H., Thangavel, R. R., Margine, I., Jaffe, E. L., Krammer, F., Solorzano, A., Garcia-Sastre, A., Palese, P., and Bouvier, N. M. (2013) Influenza A(H7N9) virus gains neuraminidase inhibitor resistance without loss of in vivo virulence or transmissibility. Nat. Commun. 4, 2854. (3) Lou, Z., Sun, Y., and Rao, Z. (2014) Current progress in antiviral strategies. Trends Pharmacol. Sci. 35, 86−102. (4) Ma-Lauer, Y., Lei, J., Hilgenfeld, R., and von Brunn, A. (2012) Virus-host interactomes–antiviral drug discovery. Curr. Opin. Virol. 2, 614−621.

Additional biological assay protocols, data, and expanded description of the results (PDF)

AUTHOR INFORMATION

Corresponding Author

*Tel: +852 22554383. Fax: +852 28551241. E-mail: bzheng@ hkucc.hku.hk. Notes

The authors declare no competing financial interest. F

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX

ACS Infectious Diseases

Letter

(5) Das, K., Aramini, J. M., Ma, L. C., Krug, R. M., and Arnold, E. (2010) Structures of influenza A proteins and insights into antiviral drug targets. Nat. Struct. Mol. Biol. 17, 530−538. (6) Morley, S. J., Curtis, P. S., and Pain, V. M. (1997) eIF4G: translation’s mystery factor begins to yield its secrets. RNA 3, 1085− 1104. (7) Aragon, T., de la Luna, S., Novoa, I., Carrasco, L., Ortin, J., and Nieto, A. (2000) Eukaryotic translation initiation factor 4GI is a cellular target for NS1 protein, a translational activator of influenza virus. Mol. Cell. Biol. 20, 6259−6268. (8) Zhou, H., Zhu, J., Tu, J., Zou, W., Hu, Y., Yu, Z., Yin, W., Li, Y., Zhang, A., Wu, Y., Yu, Z., Chen, H., and Jin, M. (2010) Effect on virulence and pathogenicity of H5N1 influenza A virus through truncations of NS1 eIF4GI binding domain. J. Infect. Dis. 202, 1338− 1346. (9) Yanguez, E., Rodriguez, P., Goodfellow, I., and Nieto, A. (2012) Influenza virus polymerase confers independence of the cellular capbinding factor eIF4E for viral mRNA translation. Virology 422, 297−307. (10) Heaton, N. S., Moshkina, N., Fenouil, R., Gardner, T. J., Aguirre, S., Shah, P. S., Zhao, N., Manganaro, L., Hultquist, J. F., Noel, J., Sachs, D. H., Hamilton, J., Leon, P. E., Chawdury, A., Tripathi, S., Melegari, C., Campisi, L., Hai, R., Metreveli, G., Gamarnik, A. V., Garcia-Sastre, A., Greenbaum, B., Simon, V., Fernandez-Sesma, A., Krogan, N. J., Mulder, L. C., van Bakel, H., Tortorella, D., Taunton, J., Palese, P., and Marazzi, I. (2016) Targeting Viral Proteostasis Limits Influenza Virus, HIV, and Dengue Virus Infection. Immunity 44, 46−58. (11) Fawell, S., Seery, J., Daikh, Y., Moore, C., Chen, L. L., Pepinsky, B., and Barsoum, J. (1994) Tat-mediated delivery of heterologous proteins into cells. Proc. Natl. Acad. Sci. U. S. A. 91, 664−668. (12) Sugiyama, K., Obayashi, E., Kawaguchi, A., Suzuki, Y., Tame, J. R., Nagata, K., and Park, S. Y. (2009) Structural insight into the essential PB1-PB2 subunit contact of the influenza virus RNA polymerase. EMBO J. 28, 1803−1811. (13) Guilligay, D., Tarendeau, F., Resa-Infante, P., Coloma, R., Crepin, T., Sehr, P., Lewis, J., Ruigrok, R. W., Ortin, J., Hart, D. J., and Cusack, S. (2008) The structural basis for cap binding by influenza virus polymerase subunit PB2. Nat. Struct. Mol. Biol. 15, 500−506. (14) Tarendeau, F., Crepin, T., Guilligay, D., Ruigrok, R. W., Cusack, S., and Hart, D. J. (2008) Host determinant residue lysine 627 lies on the surface of a discrete, folded domain of influenza virus polymerase PB2 subunit. PLoS Pathog. 4, e1000136. (15) Tarendeau, F., Boudet, J., Guilligay, D., Mas, P. J., Bougault, C. M., Boulo, S., Baudin, F., Ruigrok, R. W., Daigle, N., Ellenberg, J., Cusack, S., Simorre, J. P., and Hart, D. J. (2007) Structure and nuclear import function of the C-terminal domain of influenza virus polymerase PB2 subunit. Nat. Struct. Mol. Biol. 14, 229−233. (16) Liu, Y., Qin, K., Meng, G., Zhang, J., Zhou, J., Zhao, G., Luo, M., and Zheng, X. (2013) Structural and functional characterization of K339T substitution identified in the PB2 subunit cap-binding pocket of influenza A virus. J. Biol. Chem. 288, 11013−11023. (17) Walsh, D., and Mohr, I. (2011) Viral subversion of the host protein synthesis machinery. Nat. Rev. Microbiol. 9, 860−875. (18) Zheng, B. J., Chan, K. W., Lin, Y. P., Zhao, G. Y., Chan, C., Zhang, H. J., Chen, H. L., Wong, S. S., Lau, S. K., Woo, P. C., Chan, K. H., Jin, D. Y., and Yuen, K. Y. (2008) Delayed antiviral plus immunomodulator treatment still reduces mortality in mice infected by high inoculum of influenza A/H5N1 virus. Proc. Natl. Acad. Sci. U. S. A. 105, 8091−8096. (19) Pautus, S., Sehr, P., Lewis, J., Fortune, A., Wolkerstorfer, A., Szolar, O., Guilligay, D., Lunardi, T., Decout, J. L., and Cusack, S. (2013) New 7-methylguanine derivatives targeting the influenza polymerase PB2 cap-binding domain. J. Med. Chem. 56, 8915−8930. (20) Shen, Y., Maupetit, J., Derreumaux, P., and Tuffery, P. (2014) Improved PEP-FOLD Approach for Peptide and Miniprotein Structure Prediction. J. Chem. Theory Comput. 10, 4745−4758. (21) Hanwell, M. D., Curtis, D. E., Lonie, D. C., Vandermeersch, T., Zurek, E., and Hutchison, G. R. (2012) Avogadro: an advanced semantic chemical editor, visualization, and analysis platform. J. Cheminf. 4, 17.

(22) Schneidman-Duhovny, D., Inbar, Y., Nussinov, R., and Wolfson, H. J. (2005) PatchDock and SymmDock: servers for rigid and symmetric docking. Nucleic Acids Res. 33, W363−367. (23) Mashiach, E., Schneidman-Duhovny, D., Andrusier, N., Nussinov, R., and Wolfson, H. J. (2008) FireDock: a web server for fast interaction refinement in molecular docking. Nucleic Acids Res. 36, W229−232. (24) Zhu, K., Day, T., Warshaviak, D., Murrett, C., Friesner, R., and Pearlman, D. (2014) Antibody structure determination using a combination of homology modeling, energy-based refinement, and loop prediction. Proteins: Struct., Funct., Genet. 82, 1646−1655.

G

DOI: 10.1021/acsinfecdis.6b00064 ACS Infect. Dis. XXXX, XXX, XXX−XXX