Inhibitors of eIF4A and Derivatives of Pateamine A

with Antiviral Activity and Use

Cross Reference to Related Applications

pKMMj This application claims the benefit of priority to U.S. Provisional Patent

Application Serial No. 1 621 ,265 filed on April 6, 2012, which is incorporated herein by reference in its entirety.

Federal Funding Legend

[0002] This invention was produced in part using funds from the Federal

Government under I O! Ai 103311-0.1 _s "The Role of Host and Viral Translation Factors During HCMV Infection."

Field of the Invention

pKM)3J The invention generally relates to the field of virology and antivirais for the treatment of viral disease. More specifically, this invention is directed to methods and compositions for inhibiting eukaryotic initiation factOi 4A (elP4A) in a patient in need of antiviral treatment.

Background of the Invention

[00O4J in order to replicate, viruses depend on host cell mRNA translation machinery for the synthesis of viral proteins. As such, drugs that target the specific components of the host translation machinery utilized by said viruses have the potential to inhibit viral replication, and prove beneficial as antiviral therapies. While viruses encode many proteins that prevent the host cell, from regulating protein synthesis during infection, virtually all viruses utilize cellular proteins to translate viral m NA into protein. This is doe to viral inability to generate new infectious progeny in the absence of viral protein synthesis. Therefore, the interaction of viruses with the host mR A translation machinery is a promising target for the development, of novel anti v iral drugs. iOSj The elF4Alli R A heSiease is a novel host factor required for HCMV replication. Depletion ofeIF4AHi significantly reduces virus replication, and the elF4AlIi inhibitor pateamine A halts progression of the HCMV ly tic cycle at all stages of infection . Inhibition ofeIF4AUI activity presents a promising new avenue for the development novel antiviral therapeutics capable of limiting HC V disease.

[0006) Pateamine A and hippuristanoi works by targeting cellular proteins, rather than viral proteins. As described by the studies disclosed herein, pateamine A and hippuristanoi inhibits cellular m NA translation initiation, factor cIF4A, which is critical for the replication of a broad range of viruses.

[0007! Pateamine A functions as an allosteric inhibitor of elF4A by preventing the release of eI.F4AHJ from cognate mRNAs. Pateamine A inhibits the replication of HCMV and the related alphaherpesvirus HSV1 , importantly pateamine A inhibits viral protein synthesis and the production of progeny virus at any stage of infection, indicating an ongoing requirement for eIF4AM. throughout the virus lytic cycle. HCMV utilizes specific components of the host exon junction complex (EJC) to promote efficient virus replication. Likewise, DMDA-pateamrne A, is the structurally simplified derivative of pateamine A and has equivalent or greater activity to pateamine A in viiro, (Kuznetsov, G. _; et al. (2009) AfoL Cancer Then 8: 1250-60). elF4A10. is a novel target for the development of new antiviral therapeutics.

[0008] Hippuristanoi, a recently developed inhibitor of the c!.P4A RNA heiiease, is a small molecule inhibitor of elF4F activity thai does not affect the acti vity of signaling pathways that control e!F4F integrity. Hippuristanoi binds directly to eIF4AI and ciF4AlI and inhibits their RNA helicase activity in an ATP-competitive .manner. elF4A is a subunit of the elF4F complex. Consistent with the requirement ofeIF4A activity for the translation of most cellular mRNAs, hippuristanoi potently inhibits global protein synthesis in mammalian cells, importantly, hippuristanoi is a specific inhibitor of elF4.41 and eIF4AIl, and it does not affect the activity of other RNA helicases, including the closely related eIF4AHI.

[00091 Post-trai>.seriptionai control of gene expression is a highly regulated process in mammalian ceils. Following transcription primary transcripts undergo substantial remodeling in the nucleus. Two critical steps in mRNA maturation are the removal of in.trons and Ore subsequent export of the mature mRNA to Ore cytoplasm. A multi- component complex termed the spliceosorae recognizes and removes mtrons from primary mRNAs. (Jurica, MS. & Moore, M (2003) Mid. Ceil, 12:544). Concomitant with splicing, RNA binding proteins are deposited On the mRNA to signal that the mRNA is properly spliced and ready for export: the cytoplasm. The .mature mRNA then transits through the nuclear pore to the cytoplasm where it translated into protein. (Drcyfuss, G., Kim, V.N, & Kataoka, N. (2002) Nat. Rev. Mo I. Cell Biol., 3:195-205). Each step in this process is requires a distinct complement of protein complexes, and each, step is regulated in response to e ironmenial cues. (Chang, T.H., et al., (20.13) BUK'him Biophys Acid) The coordinated actions of the splicing machinery and subsequent RNA binding proteins ensures that only mR s that have undergone the proper processing steps are released to the cytoplasm tor translation into protein. (Wang, Z. & Burge, C.B., (2008) UNA, 14:802- 13). pstMiJu] The eIF4A0! RNA hehcase plays several important roles in the export of spiked mRNAs. Interactions with the splicing machinery positions elF4AIM

approximately 24 base pairs 5' of newly ligated exon junctions. (Shiimori, M., inoue, MX, & Sakamoto, H. ₅ (2013) Mol Cell Biol, 2013. 33: 444-56; Ferraiuo!o, M.A., et al, (2004) rac Noil Acad Sal U S A, 1.01 :411.8-23; Shibuya, T. ₅ et al, (2004) Nat Struct Mai BioL 1 1:346-51). elF4AUI itself has no RNA binding specificity, howe ver two associated proteins, Y .14 and magoh coordinate elF4Aill deposition on mRN A with the action of the spiiceosorne. (Bono, F. & Gehring, N.H., (201 1) UNA Biol, 8:24-9). In addition, binding to Y14 and niagoh. stabilizes the association of clF4Alll with the spliced mRNA. Toge her with the barentsz protein, eIF4AlH, Y14, and magoh co mprise the core proteins of the exon junction complex, or EJC. (Palacios, 1M, et al., (2004) Nature, 427:753-7). After nucleation of the EJC core on the spliced mRNA, additional proteins are recruited to the complex to facilitate nuclear export of the ribonueleoproteia complex . (Bono, F. & Gehring, N.B., (201 1) RNA BioL 8:24.-9) pHMM 1} In the cytoplasm the ESC also facilitates the translation of associated mRNAs in the pioneer round of translation, (Le Fiir, H. & Seraphm, B, (2008) Cell, 133:213-6; Lejeune, F. & Maquat, L.E., (2005) Curr Opin Ceil Biol 17:309-15). in the pioneer round, FJC-bound mRN As undergo a. single round of translation thai serves to assess the quality of the mRN A. (Ishigaki, Y.„ et al., (2001 ) Cell, 106:607-17). As the pioneer round of translation proceeds, the translocating ribosome displaces EJCs from the mRNA, freeing the mRNA for subsequent rounds of translation. (Chang, Y.F., torn, IS. & Wilkinson, M.F, (2007) Atmu Rev Biockem, 76: 51-74; Lejeunc, ¥., et al., (2002) EMBOJ, 21 :3536-45). Should translation terminate prior to the removal of all EJCs, the remaining EJCs serve as scaffolds for the rc r itoient of the nonsense-mediated decay (Aiexaodj-ov, A., et a!., (2012) Proc Nail Acad Sci U SA, 109:21313-8) mRNA degradation complex. (Chang, Y.F., Imam, J.S. & Wilkinson, M.F. (20ΪΠ) Anna Rev Biochem, 76: 51 -74). The resulting destruction of the mRNA through the NMD pathway prevents the expression of potentially deleterious truncated proteins. Thus elF4AHI links post-transcriptional R A processing to mRNA quality control pathways to maintain the integrity of the proteome. MH Zj Like host mRNAs, human cytomegalovirus (HCMV) mRN As are actively transported from the nucleus to the cytoplasm., (Sandri-Goldin, .R..M., (2001) Curr Top Microbiol Immunol, 259:2-23). Much work has been done to determine the role of viral factors m this process. These studies have primarily focused on the conundrum of how itnspiieecl viral mRN As are efficiently exported to the cytoplasm. Id, Traditionally herpesvirus mRNAs have been considered to linearly encoded, i.e. splicing is thought to have a limited role in expanding the genetic complexity of herpesvirus proteomes. As uiispliced mammalian mRNAs are typically retained in mis nucleus (Valencia, P.. Dias, A.P & Reed, R., (2008) Proc Nail Acad Sci U SA, 105:3386-91 ), this raised the question as to how herpesvirus mRNAs are exported to the cytoplasm. Several elegant studies have revealed that herpesviruses encode a family of mRN A-assoeiate proteins, the ICP27 family, which facilitates trafficking of unsphced viral mRNAs to the cytoplasm. (Sandri- Goldin, R.M., (200.1 ) Curr Top Microbiol Immunol, 259:2-23). As might he expected ICP27 family members are critical for efficient viral replication. While much is the role for -viral proteins in the export ofherpesvtrus mRNAs is well-described, relatively little is known regarding the role of host proteins in this process. pMM)13j Recent next generation sequencing studies have changed our iew of the potential for alternative splicing t expand the coding capacity of the HCMV genome. (Stern-Ginossar, NL, et aL (2012) Science, 338: 1088-93; Gatherer, U, et aL, (201 1) Proc Natl Acad Sci U A, 108:19755-60). These studies have revealed that up to a third of ail HCMV mRNAs are splicc&CGatherer, D„ et aL (201 \) Proc Natl Acad Sci US A, .108: 19755-60). Presumably spliced viral .mRNAs also require .an active transport process to leave the nucleus. However the role of host factors such as the BJC in the export of spliced viral mRN As and virus replication has not been investigated.

\ mM\ To date no virus has been shown to encode a. functional ribosome. Rather viruses recruit host ribosomes to viral mRNAs for the synthesis of viral proteins. Due to thei reliance on host factors, viral mRNAs must compete with host messages for access to the protein synthetic machinery. In orde to limit competition many viruses inhibit host protein synthesis, thereby freeing host ribosomes for the translation of viral m NAs.

However human cytomegalovirus (HCMV) does not inhibit the iraaslation of host mRNAs during infection. (Herdy, B„ M. et al. (2012) Nat Immunol 13:543-50; Mutltukrishnan, S. et at., (.1 75), Nature 255:33-7). This suggests that HCMV mRNAs either avoids

competition with host mRN As, or alters the abundance of host translation factors to support the continued translation of both host and viral mRNAs. I either case, the mechanisms by which HCMV mR As reeruit host ribosomes and translation factors are critical determinants of virus replication, jtidOlS! The recruitment of a ribosome to an roRNA occurs through an ordered assembly of translation factors on the 5' terminus of the message. Most cellular mRN As contain a 5 ^" 7~methyiguanosine cap (ro G cap) that facilitates their translation.

(Muthukrishnan, S. et ah, (1975), Nature 255:33-7). The m ⁷ G cap promotes mRNA translation by nucleating the formation of translation initiation complexes on the .5' end of the mRNA. The primary translation initiation comple used by capped hos mRNAs is the eIF4F complex. (Brown-Luedi, M, L, et al ( 1982), Biochemistry 21 :42Q2<6; Gmgras, A. C, Raught B, & Soncnberg, N. (1 99) Amu Rev Biockem 68:913-63; Grifo, J, A„ et al, {\ 9 } JBiol Chem 258:5804-] ). The el.F4F complex provides three functions that facilitate the translation of capped mRNAs: 1) the nueieatio of translation factors on the 5' end of capped mRNAs 2) the recruitment of 40S ribosoraal subunits to the roRNA and 3) the removal of mRNA structures ^' that would otherwise impede ribosome scanning, (Gmgras, A. C, Raught, B. & Sonenberg, N. ( Ι 999) ,½Η?/ Rev Biockem 68:913-63).

1<HJ016] The three subunits of the elF4F complex each play a role in at least one of these functions. The eIF4E summit binds directly to the m ^' G mRNA cap and nucleates the assembly of the eIF4F complex on the 5' end of the mRNA. The ei ^' F4G subtimt binds to eIF4 ^' E and serves as a scaffold that coordinates the interaction of multiple translation factors with the elF4f complex and the mRNA (Jackson, R. I, Helen, C.U. & Pestova, T.V., (2010) Nat. Rev Mol Ceil Biol 1 1 : 1 J 3-2?). For example elF4G binds the eIF3 complex, which in turn brings 40$ ri ^' bosoma! subunits to the 5' end of the mRNA. id. The interactio of elF4G with the poly A binding protein (PABP) promotes mRNA circularization through its interaction. (Deny, M. C, ct at (2006) Cold Spring Barb S mp Quant Biol 71 :537-43). elF4G also recruits the third subuntt of the elF4F complex, the elF4A RNA heliease, elF4A stimulates translation initiation by resolving secondary structures in the 5' untranslated region (5 ^! UTR) of the mRNA m an ATP-dependent manner. (Griib, J. A, et al. (1982) J Biol Chem 257:5246-52). e!F4A helicase activity allows the 40S ribosomal subunit to scan the 5 ^" UTR ntil it reaches the initiating codon, at which time protein synthesis commences. The importance of the eiF4f complex for the translation of capped mRNAs is highlighted by the finding that inhibiting or disrupting the elF4F complex results in a global inhibition, of protein synthesis. (Thoreen, C, C, et al. (2009) J Bio} ( ¾:·;;·?. 284:8023-32).

{«8017] The elF4G arid elF4A subunits can also participate in alternative modes of translation initiation outside of the context of the eIF4F complex. Both elF4G and eJF4A are often required for translation initiation at internal ribosome entry sites (IRESs).

(Thompson, S. R., (2012) Trends Microbiol 20:558-66). This mode of translation initiation differs from the canonical eIF4F- ependent pathway in thai 40S ribosomal subanits recruited to IRESs do not scan the 5' UTR, but rather are thought to bind the rn ^' RNA at or near the site of translation initiation. Specific IRES-associatcd trans-acting factors,, or ITAFs, recruit elF4G and elF4A to the IRES in a manner that does not require their interaction with ciF4E. (Pestova, T, V,, et al. (20 1) Pr c Nail Acad Set U SA, 98:7029-36). While IRES-dependent translation initiation is common for RNA viruses RNAs, IRES elements are relatively rare in messages encoded by DMA viruses, although specific examples do exist. (Grainger, L„ et al (2010) J Virol 84:9472-86).

[00018] Herpesvirases do not encode obvious homologs of eSF4F subunits, and

^' herpesvirus mRNAs are thought to be translated in a eap-depemtent manner. (Walsh, D., & Mohr. I. (20.1 1 ) Nat Rev Microbiol 9: 860-75). This suggests that the elF4F complex is an. important determinant of viral protein synthesis. Consistent with this idea, several previous studies have shown that herpesvirus infection promotes assembly of the elF4F complex. For example HSV 1 and HCMV infections increase the abundance of eiF4F subunits. (Walsh, D., & Mohr, I. (2006) Genes ev 20:461-72; Walsh, D., et al. (2005) J Virol 79:8057-64), in addition members of the alpha, beta, and gairanaherpes virus f ami Iks have been found to stimulate formation of the eIF4F complex by manipulating host signaling pathways. Members of each herpesvirus family have been shown to activate the mTOR kinase, (Buchkovich, N. I, et al (2008) Nat Rev Microbiol 6:266-75). mTOR activation facilitates eIF4F compiex formation, by phosphoiylating and antagonizing the translation repressor 4EBP-.I , which binds el ^' F4E.and blocks its interaction, with elF4G. (Pingar, D. C, et al (2002) Genes Dev 1 : 1472-87}, Infection with HSVl , HCMV,

KSHV, or EB lso stimulates P.I3K activity, which can increase mTOR activity, and subsequently promote eIF4F complex formation. (Buchkovich, N. J., et al. (2008) Nat Rev Microbial 6:266-75). HSVl and HCMV also encode viral proteins that antagonize the tuberous sclerosis compiex, a negative regulator of mTOR activity, (Moorman, . I, et ai. (2008) CM Nasi Microbe 3:253-62). In addition, infection with either HSVl or HCMV activates the ERK and ΜΈ kinase cascades, resulting _. hi activating phosphorylation of eIF4E by the Mnkl/2 kinases. (Walsh, D, _? & Mohr, L (2004) Genes Dev 18:660-72), ΜΗ ϊ9| The above studies suggest that herpesvirus infection increases elF4F abundance and activity to limit competition between host and viral mRNAs for host translation initiation factors- However several recent studies have suggested a more complicated role for elF4F during herpesvirus infection. For example, while inhibiting the mTOR. kinase from the start of HCMV infection efficiently disrupts the eIF4F comple and limits virus replication, some HCMV mRNAs continue to be efficiently translated. (Clippings; A, j„ Maguire, .G., & Alwinc, lC, (201 1) J Virol 85:3930-9; Moorman, N, i, & Shcnk, T. (2010) J Virol 84:5260-9). Both total protein synthesis and virus

replication become increasingly resistant to the effec ts of mTOR inhibitors as HCMV infection progresses. (CUppinger, A. I _s Maguire, T.G., & Alw ne, J.C, (201 1) J Virol 85: 3930-9). Similarly, mTOR inhibitors do not affect HSVl replication following infection at high multiplicities ^' of infection (MOi) despite a significant redaction in the abundance of the ei.F4F complex. (McMahon, R., Zaborowska, I, & Walsh, D. (2011) J Virol 85:853-64). Together these studies suggest that, at least under some circumstances, ^' herpesvirus mRNAs can be translated under conditions that limit. eIF4F complex, abundance.

100020] While pre vious studies have used of inhibitors of host signaling pathways that control elF4F activity to examine its role in viral mRNA translation, the requirement for elF4F activity for the translation of HCMV mRNAs has not been directly before the studies presented herein.

(00021] One benefit of modulating cellular proteins critical for viral replication, is that this mechanism should limit the ability of the virus to evolve imitations (<¾ ^y ., escape mutations), which inhibit drug effectiveness. While the coding content of viral genomes is relatively amenable to mutations thai reduce the effectiveness of antiviral drugs, the human genome is much less prone to mutation. One particular benefit of

compositions/treatment that targets host cell machinery rather than virus encoded processes or proteins is a reduction in the development drug resistant viral strains,

[i¾!<)22{ An. additional benefit is that patearnine A can be tolerated for a sufficient periods of time to limit viral disease. Previous studies in mice demonstrated that

patearnine A was well tolerated for extended periods of time (e.g., greater than 30 days), suggestin thai patearnine A is not acutely toxic to mammals. Kuznetsov el ai. Mol Cancer Ther. 2009 May;8(5): 1250-60. Epub 2009 May 5.

Summary of the Invention

[WQ2$\ The invention provides methods. and compositions for inhibiting the elf4A RNA helicases, as an antiviral treatment in a patient in need thereof Compositions utilized by the methods of the invention comprise inhibi tors of eIF4A, particularly patearnine A, DMDA -patearnine A, derivati ves thereof, and hippuristanol and deri vatives thereof. The invention, further provides methods for administering a therapeutically effective amount of an antiviral pharmaceutical composition comprising an inhibitor of eIF4A to a patient suffering from a viral infection. In a particular embodiment, the virus to be treated includes but is nor limited to double stranded DNA virus, herpesvirus, herpes simplex virus type 1 (HSV), human cytomegalovirus (HCMV), enveloped positive sense RNA virus, flavi irus, dengue virus, enveloped negative sense RNA. virus, Ross River virus (RRV) ₅ paramyxyomviros, respiratory syncytial virus (RSV), parainfluenza virus. (PI V), alphavirus, Gnkungunya virus (CHIK), parainfluenza virus, enterovirus, human rhinovirus, HIV. and influenza.

| t S24i In. one aspect of the invention, the method comprises administering to a patient in need of such treatment an effective amount of an inhibitor of niRNA translation initiation factor eIF4 A or combinations of inhibitors thereof The invention thus provides advantageous alternatives to current methods and pharmaceutical compositions for treating viral infection by utilizing methods and. compositions for inhibiting cellular transcriptional machinery important for viral replication. In particular the pharmaceutical formulations and compositions of the invention comprise elF4A inhibitors, including but. not limited to: ^• particularly paicaraine A, DMDA-pateamine A, derivatives thereof, hippurisianol and derivatives thereof, and stlvestrol and derivatives thereof

[0002-5] Specific particular embodiments of the present invention will become evident from the following more detailed description of certain particular embodiments and the claims.

Brief Description of the Drawings

[00026] This invention can be further appreciated and understood from the following detailed description taken in conjunction with the drawings wherein:

[0 27] Figar 1 demonstrates that prior depletion of eIFA4IH limits progression through the HCMV lytic cycle. Figure 1A is a Western blot demonstrating viral protein expression is decreased, when e!F4A! ^' U expression is diminished. Figure IB is a graph demonstrating that depletion of eIF4AIH by shRNA resulted in a greater than. 150 fold defect in. virus production, at all times measured as quantified using the TCID50 method, (scr ^::: control cell transduced with a. !entivims expressing a scrambled sfaRNA)

[00028] Figare 2 demonstrates that e!.F4AIiI i necessary for the efficient nuclear export of iCMV mR s. Figure 2 A is a. Western blot demonstrating that the fractionation protocol rcsuits in separation of nuclear and cytosoiic fractions. Figure 2B is a graph demonstrating that less viral niR A was found in the cytosol of cells depleted of elF4Alli (4ΑΪΪ.Ϊ) as compared to control ceils (Scram) as measured by quantitative realtime PGR, The data are normalized to the percent of the mRJ A in the cytoplasm in control cells.

[00029] Figure 3 demonstrates that loss of eIF4AIII does not affect the association of viral mRNAs with polysomes. Figure 3A i a 10-50% linear sucrose gradient visualized on an agarose gel demonstrating that polysomes were present in elF4.A!l]- depieted cells. Figure 3B is a graph demonstrating iEI R A analyzed from each fraction of the sucrose gradient by quantitative real -time PCX (qPCR . Figare 3C is a graph demonstrating pp28 RNA analyzed from each fraction by qPCR. Figure D i a graph demonstrating that upon treatment with EDTA, viral. mRNAs were shifted to the lighter fractions of the sucrose gradient. The abundance of the E I niR A in each fraction was determined by qPCR. f0tR*3uj Figure 4 ^' demonstrates that depletion of the EJC core component magoh does not decrease HCMV replication. Primary fibroblasts were transduced with .magoh- specific or scrambled sh ^' R As as in figure J . Figu re 4a is a graph demonstrating that the depletion of magoh had a minimal impact on the production of cell free virus when compared to the depletion of eIF4AIII as quantified by qPCR, Open bars show the percent of magoh mRNA remaining at each time post infection in cells transduced with a lent! virus expressing a magoh-specific shRNA. The amount of magoh mRNA at each time point in ceils transduced with a co trol shRNA. is set to .100% Figure 4h is a graph demonstrating that magoh depletion does not reduce HCMV replication. The amount of virus present in culture supernatants at the indicated times post infection was determined by the TC1D50 method.

[iKMHl Figure 5 is a graph demonstrating that doses up to 250 nM of DMDA- pateamine A are not toxic to KPFs over a time period of 5 days as determined by the lactate dehydrogenase (LDH) assay.

[00Θ32] Figure 6 demonstrates thai, chemical inhibition of elF4Alil inhibits HCMV immediate early protein expression. Figure 6A is a Western blot demonstrating that DMDA ,-pateamine A treatment reduced the expression of the immediate early IE1 protein, while hippuristano! did not limit IE.1 protein expression, (M™ mock,, NT - HCMV ^* infected, no drug, CHX - cyeloheximide, HP - hippuristanol, PaiA ~ DMDA-pateamine A) Figure 6B is a graph demonstrating that neither hippuristano] or DMDA-patcammc A affect ΙΕΊ. mRNA expression as quantified b qPCR. Figure 6C is a graph demonstrating that both hippuristano! and DMDA-pateamine A treatment significantly reduced global protein synthesis in HCMV-infected cells. Cells were infected as in figure 6A, and the rate of total protein synthesis was determined by measuring the amount of radiolabeled amino acids incorporated into acid-insoluble proteins during the final 30 minutes of the six hour drug treatment.

]WMB3] Figure 7 demonstrates that pateamine A inhibits HCMV replication.

Figure 7 is a Western blot demonstrating that I E! levels remained reduced throughout the time course of infection in DMDA-pateamine A treated cells ( 100 nM). Figure 7B is a graph demonstrating DMDA-pateamine A (100 nra) limited viral DNA replication as efficiently as phosphonoacetic acid (200 pg/rnl). which is a known inhibitor of the HCMV D A polymerase. At 96 hours post infection ceils were harvested and viral genomes were quatttificd by qPCR. Figure 7C is a graph demonstrating that no virus was found in the supernatant of DMDA-patea.roi.ne A treated cells at anytime post infection. Solid bars represent no treatment witn DMDA-pateamtne A, and clear bars represent treatment with 100 «M DMDA-pateamine A. Figure 7D is a graph, demonstrating that DMDA-patea.rai.ne A inhibited HCMV in a dose dependent manner. The amount of virus in. the culture supernatants was determined at 120 hoars post infection by the TC1D50 method.

[00034] Figure 8 demonstrates that pateamine A inhibits HCMV protein synthesis and replication at all stages of the viral life cycle. Figure 8A is a Western blot gel ^• demonstrating that 100 nM DMDA-pateamine A was capable of stopping the further accumulation of viral proteins when added at any time between 0-120 hours post infection of HFFs with HCMV. Figure SB is a SDS-PAGE gel visualized by autoradiography demonstrating that DMDA-pateamme A prevented protein synthesis at any time postinfection. Radiolabeled amino acids were included in the media during the final 30 minutes to label nascent proteins. Figure SC is a graph demonsirating that treatment with pateamine A is capable of inhibiting HMCV lytic cycle at any stage of infection. Infected ceils were treated with DMDA-pateamine A (100 nM; open bars) for 24 hours at the indicated times post infection. At the end of the 24 hour period the amount, of vires in the culture supematants was determined by tire TCID50 method. Control samples were taken, from untreated samples (filled bars) at the start and finish of each treatment period for comparison.

100035] Figure 9. Demonstrates that pateamine A inhibits HSV1 replication at ail stage of the virus lytic cycle. Figure 9A is a graph demonstrating that treatment with 100 nM of DMDA-pateamine A immediately following HS VI infection of Vero cells prevented virus replication as quantified by plaque assay. Vero cells were infected with HSVl at a multiplicity of ten. (Closed boxes ~ untreated; open boxes ~ pateamine treated). Figure 9B i a graph demonstrating that the treatment of DMD A- ateamine A to Vero cells at 4, 8, 12 or 1 hours post infection prevented any further accumulation of infectious HSVl particle as determined by plaque assay , (Closed boxes = untreated; open boxes— pateamine A treated).

1000361 Figure JO is a graph demonstrating that treatment with 1 0 nM of DMDA- pateamine A immediately followin either Chikungunya virus (CHKV) or Ross River virus (RRV) infection in Hela cells decreased vims replication by greater than 100 fold as quant fied by plaque assay. p ⁾ 037| Figure 11 is a graph demonstrating thai treatment with I 00 nM of DMDA- pateara ne A immedi te following dengue virus infection of Hela cells decreased vims replication at 72 hows post infection as dcterrained by the TCID50 assay.

[00038] Figure 12 is a graph demonstrating that treatment with DMDA-pateamine

A limited human rhinovirus (RRV) replication in Hela cells by greater than 1000 fold as determined, by the TCID50 assay.

[00039] Figure J 3 is a graph demonstrating, that treatment with DMDA-pateamine A accelerated the clearance of respiratory syncytial virus (RSV) titers from the apical wash of primary human, airway epithelial (HAE) cells as determined by the TCID50 assay.

[00040] Figur 14 is a graph demonstrating that treatment with DMDA-pateamine

A accelerated the clearance of RSV infected cells from HAE raft citltiires as measured by fluorescence microscop .

100(1411 Figure 15 is a graph demonstrating that treatment with DMDA-pateamine

A prevented parainfluenza virus (PIV) titers from increasing over the test period in the apical wash of HAE cultures as determined by the TCID50 assay.

[80042] Figure 16 is a graph demonstrating that treatment, with DMDA-pateamine A prevented P1Y infection based on the absence of GFP positive cells.

[00043] Figure 17 is a graph demonstrating that treatment with DMDA-pateamine

A reduced RSV replication in HEP2G cells by greater than 100 fold at ί -4 days post inoculation of RSV,

[00044] Figure 18 demonstrates human immunodeficiency virus (HIV) replication is inhibited b pateatnine A. Figure 18A is a graph demonstrating that DMDA-pateamine A ( 100 nM) inhibits the replication of HIV strain L4 as quantified by p24 ELISA assay. Figure 188 is a graph demonstrating that DMDA-pateamine A. (i O nM ) inhibits the replication of HI strain R3A as quantified by p24 ELISA assay. Figure 18C are light microscopy images demonstratmg that HIV strain R.3A induces extensive syncytia formation, and DMDA-pateamine A ( 100 nM) completely inhibits such syncytia formation. 1000 5] Figure 19 is a graph demonstrating that 100 nM of ^' hippuristanoi was not toxic to HFFs over a treatment of 5 days as detenmned by the LDH assay,

100046] Figure 20 demonstrates -that hippuristanoi, an A TP-competitive inhibitor of cIF4A RNA hciicasc acti vity, docs not prevent HC'MV H i protein expression. Figure 20A is a Western blot demonstrating that hippuristanoi does not inhibit mTOR activity. Uninfected (M) or HC ^' MV infected cells (multiplicity of 3) were !eff untreated (NT), or treated with hippuristanoi (HP; 100 nM) or Torini (T; 250 nM) at the time of infection. mTOR, activity was determined by measuring the degree of phosphorylation of the S6 ribosomal protein f rpS6) Figure 20B is a Western blot demonstrating that inhibition of eIF4A activity docs not prevent IE! protein accumulation. Cells were treated with HP or Torin as in A, or with cyclohexittude (CHX; 100 pg rai) or actmomycin D (Act D;

Spg ml) at the time of infection. Figure 2 C is a graph demonstrating inhibition of ei.F4A activity with hippuristanoi does not prevent IE1 mRNA accumulation as measured by qPCR. Figure 20D is a graph demonstrating that hippuristanoi potently limits the rate of total protein synthesis in HCMV-infected cells. Cells were infected and treated as ia figure 26A, and the amount of radiolabeled, amino acids incorporated into protein during the final thirty minutes of the drug treatment was determined by scintillation counting.

10ΘΘ47] Figure 21 demonstrates the effect of eIF4A. inhibition with hippuristanoi 33 HC ^' MV protein synthesis. Figure 21 A is a Western blot demonstrating that the IE I protein was expressed throughout the HCMV" ly ic cycle in infected cells (multiplicity of 3} treated with hippuristanoi (HP; 100 nM). hi contrast the expression of the early protein plJL44 and the late protein pp28 was reuced and delayed in the presence of hippuristanoi. Figure 21B is a graph demonstrating that hippuristanoi significantly decreased the accumulation of the late mRNA encoding pp28 as measured by qPCR. 0 48] Figure 22 demonstrates that inhibiting eIF4A activity with hippuristanoi prevents viral DNA accumulation and replication, .Figure 22A is a graph demonstrating that inhibiting eIF4A with hippuristanoi (HP; 100 nm) or phosphonoacetic acid (PAA; 200 ug/ml) at the time of infection reduced viral DNA accumulation by two orders of magnitude. D A was extracted form infected cells at the indicated times post infection, and. viral DNA was quantified by quantitative real-time PGR using primers specific for the major immediate early promoter, The total DNA content of each sample was normalized to the abundance the host GAPDH gene. Figure 22B is a graph demonstrating that hippuristauol (100 nM) inhibits the production of ccU-frec progeny virus (clear bars). Solid bars represent no treatment The amount of virus present in culture supernatants was determined at the indicated times post-infection by the TCID50 assay . The limit of detection in this assay is 6.7 X U TCID50/ml. KH»49| Figure 23 demonstrates that prior depletion of elF4AI limits progression through the HCMV lytic cycle. Figure 23A is a Western blot demonstrating that primary human fibroblasts transduced with lentivirus expressing eIF4AI-speci.fic sh As or a •scrambled control exhibited diminished and delayed expression of a viral early (pU ^" L44) and late (pp28) protein. Figure 23B is a graph demonstrating that limiting eJF4Ai expression prior to infection also decreased the yield of free cell virus by approximately 150 fold. Cells were transduced and infected as in A, and the amount of virus in the culture media was quantified using the TCID50 method, p ⁾ 650| Figure 24 demonstrates the temporal requirement for eiF4A activity for the synthesis of HCMV proteins. Figure 24A is a Western blot, demonstrating that representative immediate earl (IE 1 ), early iplJL44) and late (pp28) viral proteins increased in abundance during a twenty four hour treatment with hippuristanol (HP; 100 nM). Figure 24B is a Western blot demonstrating that HCM proteins can be transl ted in the absenee of elF4A activity during later stages of infection. Cells were left untreated for the first 48 hours of infection, and the treated with either hippuristanol (HP; 100 nM) or cycldheximide (CHX; 00p.g ml) from 48 to 72 hours post-infection. Figure 24C is a SDS-PAGE gel visualized by autoradiography demonstrating that similar amounts of newly synthesized HCMV viral proteins were isolated from cells treated with

'hippuristanol for a twenty four period beginning at 72 hours post infection. Radiolabeled amino acids were included during the final thirty minutes of the assay t lable newl synthesized proteins. The viral proteins wer then captured by immunopreeipitation and visualized by autoradiography. Treatment with CHX served as the negative control,

J OSl I Figure 25 demonstrates HCMV replication becomes resistant to hippuristanol as infection progresses. Figure 25A is a graph demonstrating that there wa no impact on HMCV replication when hippuristanol (HP; 100 nM) was added at later times in infection. The amount of virus in the culture supematants was quantified by the TCID50 method. Figure 25B is a SDS-PAGE gel visualized by autoradiography demonstrating that hippuristanol treatment inhibited total protein synthesis in uninfected cells by greater than 70%. Acid-insoluble proteins were isolated from infected eel! fysaies and resolved on featuring acryiaroide gels. Newly synthesized protein was visualized by autoradiography,

190052) Figure 26 demonstrates that hippuristanol differentially affects the accumulation of herpes simplex virus (H ^' SV i) progeny virus in Vero and primary human foreskin fibroblast (HFF) cells. Figure 26A is a graph demonstrating that 100 nM of hippuristanol had no effect on the accumulation of HSV1 in the culture supernatant of infected Vero cells as measured by plaque assay. Figure 26B is a graph demonstrating that 100 «M of hippuristanol suppressed HSV 1 replication in HFF cells as measured by plaque assay.

Detailed Description of the Preferred Embodiments

[00053] The invention is more specifically described below and particularly in the

Examples set forth herein, which are intended as illustrative only, as numerous

modifications and variations therein will be apparent to those skilled in the art.

10805 ] As used in. the description herein and throughout the claims that follow, the meaning of "a", "an", and "the" includes plural reference unless the context clearly dictates otherwise. The terms used in the specification generally have their ordinary meanings in the art, within the context of the invention, and in the specific context where each term is used. Some terms have been more specifically defined below to provide additional guidance to the practitioner regarding the description of the invention,

[00055] The invention involves methods of treating viral diseases by inhibiting mRNA translation initiation factor elF4A, and in particular, using pateamine A, D DA- pateamine A, derivatives thereof or hippuristanol.

[00056] The term "therapeutically effective amount" or "pharmaceutically effective amount" generally refers to an amount sufficient to affect a desired biobgical effect, such as a beneficial result, including, without limitation, prevention, diminution, amelioration or elimination of signs or symptoms of a disease or disorder. Thus, the total amount of each active component of the pharmaceutical composition or method is sufficient to show a meaningful patient benefit, for example, hut sot limited to, treatment of viral infection. Thus, a "pharmaceutically effective amount" will depend upon the context in which it is being administered. A pharmaceuticall effecti ve amount may be administered in one or more prophylactic or therapeutic administrations. When applied to an individual active ingredient, administered alone, the term refers to that ingredient alone. When applied to a combination, the term refers to combined amounts of the active ingredients that result in the therapeutic effect,, whether administered in combination, serially or simultaneously. KHJ57j Administration routes for the antiviral compositions of the invention include orally, through injection by intravenous, intraperitoneal, intracerebral (intra- pareuehynial), intracerebroventricular, intramuscular, intra-oenlar, intraarterial, iniTaportal, or intralesional routes; by sustained release systems or by implantation devices. The antiviral ^' compositions may be administered by bolus injection or continuously by infusion, or by implantation device. The antiviral composition also can be administered locally via implantation of a membrane, sponge or another appropriate material on to whsch the desired molecule has been absorbed or encapsulated. Where an implantation device is used, the device may be implanted into any suitable tissue or organ, and delivery of the desired molecule may be via diffusion, timed-release bolus, or continuous administration,

1 0058] Optimal antiviral compositions can be determined by one skilled in the art depending upon, for example, the intended route of administration, delivery format and desired dosage. See, for example, REMINGTON'S PHARMACEUTICAL SCIENCES. See

REMINGTO 'S PBA R M A CEU1ICA L SCIBNC -S (18th Ed., A. &. Gennaro, ed.. Mack Publishing Company 1 90). Such, compositions may influence the physical state, stability, rate of w vivo release and rate of in vivo clearance of the antibodies of the invention. It is within the .knowledge of one of skill in the art to choose the type and adjust the amount of a pharmaceutical excipient, diluent or carrier for use in the pharmaceutical compositions provided by the invention to be used with the inventive methods described herein.

|0tRl5i>] The antiviral compositions of the invention can further comprise a pharmaceutically acceptable excipient, diluent or carrier. The primary vehicle or carrier in a pharmaceutical composition .may be either aqueous or non-aqueous in nature. For example, a suitable vehicle or carrier for injection may be physiological saline solution.

10006 ] A "patient" or "subject" to be treated by the subject method can mean either a human or non-human animal, and in certain embodiments is a mammal, hi particular embodiments, a "patient in need of treatment" is a human suffering- ^' from a viral disease. Viral disease can be identified when sufficient e vidence of viral infection is present to establish a positive diagnosis, or when viral titers or antigen levels reach levels consistent with clinically accepted standards for a positive diagnosis.

[00 61 J In the method of the present invention, antiviral compositions can be administered to any member of the Vertebrate class, Mammalia _* including, without limitation, primates, rodents, livestock and domestic pets. Livestock include mammals to be consumed or that produce useful products. 0 62] The antiviral compositions administered in accordance with this invention can include compounds that inhibit the normal function of the eukaryotic initiation faetor- 4A family of RNA helieases {eIF4AI, elF4AH. and elF4AIii).

[80063] e.lF4A is a founding member of a "DEAD-box" family of ATP-depcndent helieases whose activity as an ATPase and RNA heiicase have been extensively characterized both biochemically and structurally. eIF4AI/l.I comprises two distinct domains, with, an N-terminal domain serving mainly as a cetiter of ATPase activity and C-tenrikus acting primarily for RNA binding. The two domains are connected through a short linker, and residues on both domains are required tor function. Enzymatic activities of elF4Al U are stimulated by elF4G, that primarily binds to N-terminal domain of eIF4A. in addition to eIF4G, heiicase activity of elF4Al/ either alone or as part of the ciF4F complex is also stimulated by ei.F4S, another protein that participates in eukaryotic translation initiation. Unlike eIF4G, that .forms a stable complex with eIF4Al/IL association between elF4B and eIF4Al ll appears to be transient, as a stable complex between the two has not been, observed.

}(HK»i>4 j Suitable compounds that inhibit the normal function of the elF A family of

RNA helieases include pateamine A, DMDA-pateamine A, derivatives thereof and hipppuristanol. Pateamine A is a. marine metabolite isolated from Myc !e sp.,. and. is of chemical formula:

The simplified Pateamine A derivative is des-methyl, des-ammo paieamine A (also DMDA-patearoine A). This derivative is of formula:

DMDA-pateamine A, is the structurally simplified derivative of pateamine A and has equivalent or greater activity to pateamine A in vitro.

in particular embodiments, hippuristaaot is used in the treatment of viral infections.

Hippumtanol has the fol

(3R,5SJ 1 S,20R,22R _J 24S)~24-Me{hyl~22,25-epox ^, furostan-3,.i 1 ,20-triol.

[fltMMS] The "viral disease" which can be treated in accordance with the methods of the invention refer to diseases caused fay any of various simple submicroscopic parasites of plants, animals, and bacteria, that cause disease. Exemplary viral diseases that arc suitable as fargets for the inventive methods include but are not limited to diseases caused by double stranded DMA viruses, enveloped positive sense RNA viruses, enveloped negative sense RNA viruses and non-enveloped posiiive sense RNA viruses, and non-en veloped ^• negative sense RNA viruses. More particular types of viral diseases include diseases caused by herpesviruses, paramyxyomviruses, flavivtmses, alphaviruses, enteroviruses, HIV, and influenza. More advantageous viruses include human cytomegaloviruses, .herpes simplex viruses type 1, respiratory syncytial viruses parainfluenza virus, dengue viruses,, Chikunguriya viruses, parainfluenza viruses, human rhmovirus and respiratory syndrome viruses, and influenza A viruses,

[00066] The term "mediated" refers to the cause of viral disease including disease pathology .and conditions of the disease. In particular embodiments the viral disease is mediated by a double stranded DNA virus. In certain embodiments the viral disease is mediated by herpes virus that iaelud.es, but not limited to, human cytomegalovirus or ^' herpes simple virus type 1.

Examples

[00067] T e Examples which follow are illustrative of specific embodiments of the invention, and various uses thereof. They arc set forth for explanatory purposes only, and are not to be taken as limiting the invention. xample il: _: E.ffecjof eI.F Ai1I on efficient HCMV replication

[00068] To examine a potential role for the EJC in HCMV replication, cells were depleted of etF4AIII, the EJC component that: nucleates its deposition on spliced ni As, and the production of HCMV progeny virus was assessed .

[000691 Primary human foreskin fibroblasts (HFFs) were grown in DMEM containing .10% newborn calf serum. Cells were used between passage 5 and 15 in all experiments. Replication defective lerrtiviruses expressing elF4Alli-or magoh-specifte sh.lR.NAs were generated by transacting the shJRNA expression plasmtd (obtained from the UNC Lentivirus Core Facility) together wit packaging plasrnids into 293T ceils. Forty eight hours post transfection lentivirus was isolated from cell-free supernatants and passaged through a 0.22 μηι filter to remove cell debris. In the some cases the lenti virus stocks were generated by the University of North Carolina Lentivirus Core Facility using the same protocol.

100070] For transductio of HFFs, lentivirus stocks were incubated o vernight: with confluent HF ^' F cultures in the presence of polybrene (5 nig/ml). The following day the media was replaced with serum tree DMEM, and the cells were used for experiments at: 72 hours post transduction. Efficient depletion was routinely assessed by quantitative realtime PCR or Western blot. The same protocol was used to generate ieiiiivrrus expressing magoh-speciftc shRNAs .

100071] A variant of the HCM V ^" AD 169 strain containing a green

fluorescent protein expression cassette inserted in the UL21.5 locus was used as the wild type virus strain in all studies, (Wang, D. ₅ Bresnahan, W.„ & 8henk _? T.. (2004) Proc. Nail Acad. Sci. .1 8: 19755-60). Unless otherwise noted, all infections were performed at a multiplicity of infection (MOl) of three in serum free media in a minimal volume of media, infected cells were harvested by scraping and stored at -8( C until analysis.

Quantification of cell-free virus by the TCID50 method was performed as described previously. (Moonnaji, N.i & Shenk Ϊ, (2010) J. Virol., 84:5260-9).

100072] Confluent fibroblast cultures were depleted of clF4AIlI by transduction with a leffiivirus expressing eiF4A.lii-speeific shRNAs. Cells transduced with a scrambled shR. A served as a control. The ceils were then infected with HCMV, and the production of cell free virus in the supernatant was assessed by the TCID5 method over a single round of viral replication. As shown in figure 1 A, the shRNA efficiently depleted eiF4Alil form the ceils throughout the time course of infection. When the amount of virus in the supernatant of elF4AI.I! -depleted cells was compared to that in control supernatan.Cs we found that depletion of eIF4AIlI resulted in a greater than 150 fold defect in virus production at all times measured (Fig. IB). Thus cIF4AHI expression is required for efficient HC V replication

{00073) To determine what stage of the vims life cycle requires e.IF4AI.Ii

expression., the expression of representative immediate early, early, and late proteins throughout a time course of infection was measured in control or eIF4AHI depleted cells. In both control and elF4Aii! depleted cells the immediate earl protein IE 1 was efficiently expressed at all times post infection. In contrast the expression of an early protein, pUL44, was diminished, as was the expression of the late protein pp28 (FIG. 1 A). This result demonstrates that ei.F4AlH is required for the transition from the immediate earl to early stage of infection.

Example 2: The Effect of eIF4AIIl o« Efficient Nuclear Export of Viral Transcripts 100074] The ETC plays an ^' important role in the post-trauscriptional regulation of host ceil gene expression. The ^' EJC binds to nascent spliced transcripts and facilitates their export from the nucleus to the cytoplasm. To test whether clF4AlfI might facilitate virus replication by playing a similar role is the export of HCMV mRNAs from the nucleus, the ratio of viral .mRNA in the nucleus and cytoplasm in tTF4Alll.~depieted cells was

measured.. Primary fibroblasts were transduced and infected as described in Example 1. Control or el.F4Al.ii depleted cells were infected with HCMV, and then fractionated into nuclear and cytoplasmic fractions at 72 hours post infection. The purity of the fractions was assessed by Western blot using antibodies for the cytoplasmic protein tubulin and the nuclear laroio A/C.

J0Q075] Ceils were harvested by scraping, pelleted, and re-suspended in

fractionation buffer {20 m. Trts-MCL, pH 7.4, 140 roM KCi, SmM MgC½, l% Trttoa-X 100, 10 mM DTT). The !ysate was passed five times through, a 27 gauge needle, and then spun at 300 X g for five minutes. The resulting supernatant contained the cytosolic fraction, while the -pellet contained the nuclear fraction, in each experiment a portion of the nuclear and cytosolic fractions was analyzed by Western blot to ensure efficient fractionation. Antibodies specific for tubulin or laminin A/C were used to monitor efficient cytoplasmic and nuclear fractionation, respectively.

[00076] Figure 2A demonstrates that the fractionation protocol results in separation of nuclear and cytosolic fractions, as tubulin was found exclusively in the cytosolic fractions while iamin A/C was only detected in nuclear fractions.

198077} The abundance of representative viral mRNAs in both locations was then, determined by quantitative real-time PCR (qPCR). .RNA. was extracted from die nuclear or cytoplasmic fractions of control or eiF4AIlI~depiefed. cells. The amount of each viral mRNA in the nuclear and cytoplasmic fractions was then measured by qPCR. The ratio of nuclear ^' to cytoplasmic mRNA in control or el 4Alll depleted cells was calculated. As shown in figure 2B, less viral mRNA was found in the cytosol of cells depleted of eIF4AIIl as compared to control cells. A similar reduction in. cytosolic mRNA abundance was observed tor all three viral mRNAs tested. This was surprising as eI.F4A ^' HI is thought to facilitate the export of spliced mRNAs, and neither the IJL44 nor U L mRNAs are known to be spliced. Result indicate that elF4AIH expression is required fo the efficient nuclear export of both spliced and unspliced HCMV mRNAs. ">7

Exam le 3: eIF4AflI are Dispensable for Viral iti NA Translation. l ⁽ HKi78 S The exon junction complex (EJC) also contributes to the translation of its associated mRNAs in the cytoplasm. (Le Hk. H. & Seraphin, B. (2008) Cell 133; 213-6) eIF4AIH as part of the EJC interacts with the cap binding complex, which is required for the pioneer round of translation, (Hwang, X, et al. (2010) Mol Cell 39: 396-409).

Translation in the pioneer round is thought to precede and facilitate subsequent translation driven by the canonical elF4F translation initiation complex. (Chin, S.Y., et al. (2004) Genes Dev. 2004. 18: 745-54). The reduction in nucleo-cytopiasmic transport of viral niRNAs when. e!F4AHJ was depleted suggested that eiP4A!li might also facilitate the translation of virai niRNAs in die cytoplasm.

P M>79] First, to determine whether depicting ceils ofelF4AIU affected polysome abundance in HCMV infected ceils, polysomes were separate in a sucrose gradient. Cells were treated with eydoheximide (CHX; 100 mg nil) in media for ten minutes, washed twice in PBS with CHX, and then collected by scraping. The cells were pelleted by centrifugarioti at 5,000 X g for ten minutes, and then resuspended in 1 nil of polysome lysis buffer (20 mM Tris-HCL, pH ^" 7.4, 140 inM KCI, 5mM MgCtj, t% Triton-X 100, 10 IBM DTT) containing CHX. Cells were incubated on ice for ten minutes and then disrupted by five passages through a 27 gauge needle. Nuclei were removed by spinning the iysaie for five minutes at 300 X g. The supernatant was cleared of mitochondria by a subsequent ten minute spin at 15,000 rpni in a microcentrifuge. The resulting supernatant was layered onto a 10-50% linear sucrose gradient (made in polysome buffer) containing CHX, The gradients were span for 2 hours at 4''C in a SW41 swinging bucket rotor at 35,000 rp with no brake. After centrifugation the gradients were manually fractionated from the top of the gradient into 750 μΙ fractions.

[fKiMOj To visualize the location of polysomes in the sucrose gradients, RNA was extracted front 250 μί from each gradient fraction The RNA pellet was resuspended in 30 μΐ of gel loading buffer containing 1% et idiurn bromide and resol ved on 2% agarose gels, images were obtained using a Biorad. GelDoc. K iSl ] For quantification of viral niRNAs i each fraction of the sucrose gradient,

RNA was extracted from 250 μΐ of each gradient traction using Trizo! reagent and reverse transcribed to cDNA as described below. The abundance of specific virai mRNAs in each fraction was quantified b quantitative real-time PGR as described below and previously. (Terbium S.S. et a!. (2010) PLos P t og., 6:eIO0O965).

[00032$ Cytoplasmic extracts of control or eIF4AIil-depleted ceils were resolved through linear sucrose density gradients. The gradients were manually fractionated and the location of monosome and polysome containing fractions was determined by monitoring the presence of ribosomal RNAs in each fraction. Figure 3 A demonstrates that polysomes were present in eiF4A10-depleted ceils although at a somewhat reduced level. This results consistent with revious results demonstrating that elF4AI0 is not essen.ti.al for niR. A translation, bat rather serves to enhance the translation of its associated .mRNAs.

[00083] Next to determine whether depletion of elF4Aill affected the association of specific HCM V transcripts with polysomes, mRNA was distributed across a sucrose gradient. raRNAs that are actively undergoing translation are bound by -multiple

ribosomes, or polysomes, and therefore migrate further into a sucrose gradient, inhibition of roRNA translation or disruption ofribosomes decreases the number of ribosomes associated with an mRNA, resulting i slower migration through, the gradient. Therefore the degree to which an mRNA is being translated can be inferred from the distribution of tire mRN A across the sucrose gradient (Figs. 3B & 3C).

10008 1 To determine if eIF4AHI was required for efficient HCMV mRNA translation, the abundance of specific viral niRNAs .throughout a sucrose density gradient in cells depleted of eIF4AlIl was measured. Importantly, only cytoplasmic niRNAs are analyzed in this assay, thereby allowing us to discriminate the effects of elF4AIH depletion on viral mRNA export from its potential role in the translation of -viral mRNA in the cytoplasm. Total RNA was extracted from, each fraction, of the sucrose gradient and the abundance of viral mRNAs in each fraction was determined by qPCR.

100085] R A was isolated from cell, pellets by Trizol extraction per manufacturer's instructions. Briefly, the sample (either ceil pellet or 250 μΐ of liquid sample) was mixed with I ml of Trizol reagent (Invitrogen, Carlsbad, Uni ted. States) . The mixture was thoroughly vortexed and then, extracted once with chloroform. The aqueous phase was removed to a new tube and an equal volume of isopropanol was added and mixed. The mixture was incubated at -2 T ^" for at least one hour and the RNA was pelleted in a microcentrifuge. The RNA was washed once in 70% ethanol, allowed to air dry and resuspendcd in distilled wafer, RNA concentration was determined using a NanoDrop spectrophotometer. The RNA was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Life Technologies, Carlsbad, United States). Quantitative reverse transcriptase real-time PCR (qRT-PCR) was performed using a Roche LC480 using SYBER Green master mix (2x.) and primers specific for the indicated m NA. Each sample was analyzed in duplicate, and ail experiments were performed at least three times. MtSiij To quantify viral DNA .accumulation, DNA was extracted from infected cells by phenol: chloroform extraction. Five hundred nanograms of DNA was analyzed by quantitative real-time PCR (qPCR) using primers specific for die HCMV major immediate earl promoter, as previously described. (Terhune, S.S. et al. (2007) J. Virol., 81 :3109-23). The abundance of viral DNA was determined by comparison to a standard curve consisting of .10* to 10* copies of genomic HCMV DNA. In all experiments the r value of the standard curve exceeded 0,98. The results were normalized to the abundance of GAPDH in the sample to correct for variations in loading. Viral standards were also used to determine the percentage of a given mRN A present in the nucleus and cytoplasm. The resulting ration was normalized to the amount of the mRNA in a total RNA isolation prepared from an aliquot of the sample prior to fractionation. (Crtstea, I.M,, et al (2010) . Virol., 84:7803-14),

(0008?] Fold change in viral mRNA abundance was calculated by the AACT method as described previously.

[00088! Depletion of elF4Aill did not alter the migration of the viral mR As in the gradient, indicating that eJF4AIJI is not required for translation of viral messages. This was true for both the spliced HCM IE I mRNA, and the unsp!iced UL99 mRNA. To ensure that the migration of the viral mRNA in the gradient reflected its association with ^• polysomes rather than a potential interaction with HCMV virions, we detemiined if EDT A treatment altered the migration of the mRNA in the gradient. EDTA chelates magnesium cations that are required to maintain ribosome integrity. Upon treatment with EDT ribosomal RNAs were lost from the denser fractions of die gradient (data not shown) and viral mRNAs were shifted to the lighter fractions ( Fig. 3D). This demonstrates that the observed migration pattern of the viral mRNAs in the sucrose gradient reflected their association with, polysomes, rather than thei incorporation into HCMV virions. From this data we conclude that while eIF4AIM is required for the efficient export of viral raRKAs from the nucleus, eIF4AIH plays a minimal role in the translation of viral messages in the cytoplasm.

[00 8 ) The above data demonstrates that eIF4A01 was an important host determinant ofHCMV replication. To further determine whether elF4AHI functioned outside of its role in the EJC, HCMV replication and protei expression in ceils depleted of the additional EJC core component magoh was measured. Magoh-depleted primary fibroblasts were infected with HCMV and the production of cell free virus throughout a single round of virus replication was determined by the TCID5 method. The depletio of magoh had a minimal impact of the product ion of cell free virus when compared to the depletion of eIF4AHI (compare figure IB and figure 4A). Consistent: this result, the expression of a representative IE, early or late protein was not affected by magoh

depletion (fig 4B). In addition, the efficient depletion of magoh did not affect the export of viral niR As from the nucleus (not shown). These data suggest that the EJC itself is dispensable for virus replication. Rather elF4AHl facilitates HCMV replication and the nuclear export of viral mRNAs through as unknown EjC-iiidependeiit function.

Example 4: Inhibition of Human Cytomegalovirus (HCMV ) Replication

by Pateamine A in Primary _^ Hnman _^ CeUs

{00090J Recently a several inhibitors of elF4A heiicases have been developed that have potent in vitro and in vivo activity. One such inhibitor is the natural product pateamine A. Pateamine A was originally identified as a translation inhibitor made b sea sponges found off the coast of Okinawa. Subsequent studies demonstrated that pateamine A binds specifically to the eIF4A family of RNA heiicases. This includes elF AI/l ^' i, R A heiicases involved in translation initiation, as well as elF4A01. To determine if Pateamine A has anti iral properties in additional to its proven efficac as a chemotherapeutic agent, viral protein and nucleic acid were quantified in cells treated with DMDA-pateamme A, the structurally simplified derivati ve of pateamine A. All experiments contained in this Example and the corresponding figures utilized DMDA-patearoine A.

[00091 J To determine the effect of pateamine on HCMV replication, primary human foreskin fibroblasts (HFFs) were infected with the AD 169 sixain of HCMV at a multiplicit of infection {MOT) of three. DMDA-pateamine A (100 nM) was added to the infected cultures at one hour post infection. Ceils were harvested at the 24, 48, 72, 96, or 120 hours post infection, and frozen at -80 degrees Celsius until analysis. The effect of pateamine A on HCM ^' V replication was determined by measuring the amount of infectious virus present in the supernatant using the tissue culture infectious dose 50 (TCI.D5 ) assay. The effects of paieamine A on cell viability were determined b .measuring the amount of lactate dehydrogenase (LDFI) present in the supernatant of treated cells using the standard biochemical assay. The effect of pateamine A on viral m NA expression was determined by quantitative reverse transcriptase real time PCR using primers specific for the HCMV IE1 mRNA. The .impact of _. pateamine A on global protein synthesis was determined by measuring the amount .of radiolabeled amino acids incorporated into acid-insoluble protein 30 minutes post injection,

[000921 DMDA-pateamine A was not toxic to primary human ceils at doses up to

100 uM over a time cours of 5 days as determined by lactate dehydrogenase (LDH) assay (Fig. 5)

[00093 _. 1 T determine if pateamine A treatment inhibited t he expression of HCMY

IE proteins, ceils were infected with HCMY, and DMDA-pateamine A was added to the cultures upon removal of the virus inocuhim. Cells were harvested six hours after the addition of the drug and the expression of the IE! protein was measured by Western hot. In parallel samples, the abundance of the ΪΕ1 mRNA was also measured by quantitative ^• real-time PCR, DMDA-pateamine A treatment at the time of infection reduced the expression of the immediate early IE! protein (fig. 6A) despite the efficient synthesis of I.EI mRNA (fig. 7C). Therefore, treatment with DMDA-pateamine A immediately following infection prevented virus protein production, but did not prevent HCMV mRNA expression. As discussed, above Paieamine A inhibits both, the eiF4Al/I.i belieases utilized by translation initiation complexes and the eIF4AIII helicase.

1 009 ] To determ ine if the effects, of pateamine A were due to inhib ition of eIF4AI/lI o inhibition of eI.F4AIIl, the effects of hippuristanol were compared to that of DMDA-pateamine A. Hippuristanol specifically inhibit el.F4Al/.li but doe not affect elF4AlU activity. Hippuristanol treatment did not limit IE 1 protein or mRNA expression (figs. 6A & 6.B, respectively) despite significantly reducing global protein synthesis as measured by the incorporation of radiolabeled amino acids into acid-insoluble protein during the final thirty minutes of the infection (fig. 6C). |04M)9Sj HFF cultures were incubated in 1 ml of methionine and cysteine free media for fifteen minutes. 125 uCi of "S-labeled methiomrie and cysteine were then added directly to the media. Where indicated inhibitors were present during both amino acid starvation and labeling periods. Thirty minutes later the labeling media was removed and the cells were rapidly washed two times in ice cold PBS containing .1 <K pg/ml CHX. Cells were then scraped, pelleted, and resuspended in ice cold RIPA buffer (20 mM Tris-HCI (pH 7.5), 150 mM Nad, 1 mM EDTA _; 1% NP-40, 1% sodium deoxycholate) containing protease inhibitors (complete EDTA-free protease inhibitors, Roche). To determine the rate of nascent protein synthesis, fifteen microli ters of the lysate was mixed with 150 μΐ of a 1% BSA solutio in water. TCA was -added to -a final concentration of 20%. and the mixture was incubated on ice for thirty minutes. The mixture was then applied to the center of a glass microfiber disk under vacuum. The filter was washed three times with 20% TCA solution, and once with 100% ethanot. The filters were allowed to air dry and then the amount of bound radioactivity was measured using a scintillation counter. The amount of radioactivity was normalized to the amount of protein in each sample as determined by the Bradford assay. To visualize nascent proteins, ten inicrogmms of each sample was resolved on a 10% SDS-Page gel. The gel was fixed in a I?; 1 mixture of acetic acid: methanol for ten minutes, dried, and exposed to film.

[I¾i0<>6] DMDA-pateamiiie A inhibited total protein synthesis to a similar degree as hippuri.stanol (fig. 6C), however only DMDA-pateamine A prevented IE! protein expression. As pateaoiine A inhibits all three- isoforms of eIF4A while hippuristaiioS inhibits inly el Al/il, these data suggest that the effects of Pateamrae A on IE I protei expression are due to inhibition of elF4A0l rather than eIF4AI/il. They also show that the cIF4AIH inhibitor Paieamine A prevents the translation of an. HC V IE protein.

1W ?{ To determine effects of Pateamine A on viral protein expression throughout a time course of infection, DMDA-pateamine A was added to infected cells at the time the inoculum was removed, and viral protein expression was measured by Western blot throughout a time course of infection. At 24 hours post infection the IE] protein expression was present in .DMDA-pateamine A -treated, cells, although at reduced leveis as compared to untreated cells. IE! levels remained redueed throughout the time course of infection in DMDA-ateamme A treated cells ( fig. 7A). Very little to no expression of an HCMV early (pUL44) or late (pp28) protein was observed at. any time post infection in cells treated with Pateamine A (Fig. ? A). Consistent with the reduction in IE and early protein synthesis, DMDA-paieamine A treatment inhibited the accumulation of viral DNA in infected cells as determined by quantitative real-time PGR. Figure 7S shows that DMDA-pateamine A limited viral DNA replication as efficiently as phosphonoacetic acid, a known inhibitor of the HCMV DNA polymerase, it should be noted that in both the analysis of viral protein expression and viral DNA accumulation a single dose of DMDA- pateaniine A was given a the time of infection, and the media was not replaced throughout the time course .

[00098] To determine the effect of Pateamine A treatment on HCMV replication,

DMDA -pateamine A was added following the removal of the virus inoculum (multiplicity of three) and the presence of cell-free virus in the supern atant was quantified throughout a single round of virus replication, A single dose of DMDA-pateamtne A was gi ven at the time of infection, and the drug was not refreshed throughout the time course. No virus was found in the supernatant of DMDA-patearaine A treated cells at any time post infection, indicating that Pateamine A inhibits HCMV replication (fig. ?C). Figure 7D shows that the DMDA-ateamine A inhibited HCMV replication in a dose dependent manner,

demonstrating the specificity of the drug. DM -ateamine A was added to infected cells at various concentrations immediately following virus adsorption, and the amount of virus in the supernatant was measured at 120 hours post-infection. Concentrations of DMDA- pateamine A as low as 10 nM inhibited HCM V replication (Pig. 7D), suggesting that the IC50 for Pateamine A is likely in the low nanomolar range.

}00β99{ At concentrations of 100 nM or greater no cell free virus was observed

(FSG. 7D). While no toxicity was observed with 100 nM DMDA-pateamine A (Pig. 5), significant toxicity was seen with the 500 nM dose. Therefore, effect of the 500 nM dose on virus replication is likely due to cell death. Together these data demonstrate that Pateamine A is a potent, non-cytotoxie inhibitor of HCMV replication that is effective at low nanomolar concentrations.

Example 5: The Impact of Pateamine A on the HMCV Lvtic Cycle at Different

Stages of Infection 00010 } inhibiting elF4Al/iI activity from the start of infection decreases HCMV replication. However as infection progresses virus replication and protein synthesis become increasingly resistant to ctF4AI/H inhibition. During the later stages of infection (72 hours and beyond) the eIF4AI/H inhibitor hippuristanol had no effect on the production of progeny virus. To determine ifHCMV infected ceils similarly developed resistance to the effects of Pateaniine A as infection progressed, infected cells were treated with DMDA-pateamme A for 24 intervals beginning at 24 hours post infection (e.g .from 24-48 hps, from 48-72 hpi, etc.). Ah experiment contained in this Example and the corresponding figures utilized DMDA-patearaine A. Untreated samples were harvested at the beginning and end of each treatment interval. As shown in figure 8A, at all times post infection the addition of DMDA-pateamine A prevented further increase in the abundance of viral proteins during the 24 treatment For example, the HCM.V Sate protein pp28 was first detected at low levels at 72 hpi in untreated cells. Between 72 and 96 hpi pp28 protein abundance increases in untreated cells. However the addition of DMDA-pateamine A. at 72 hpi prevents any further increase in pp28 protein abundance. Similar results were observed for the HC V early protein pUL44. While DMDA-patearaine A treatment did not limit IE J expression in these experiments, this likely reflects the long half-life of the IE I protein Teng, M.W. et at., (2012) ,/ Virol 86:7448-33) rather than the continued synthesis of iEl in the presence of Patearanie A.

[<XH)10i{ We also tested the effect of Paieamine A on total protein synthesis in infected cells at different temporal stages of infection. Cells were infected with HCMV, and then DMDA-pateamine A. was added for a twenty four hour interval as above. During the final thirty minutes of DMDA-patearaine A treatment radiolabeled amino acids (labeled according to the method in Example 4) were added to the cultures to label any newly synthesized proteins. Acid-insoluble proteins were then extracted from the cell lysaf.es and visualized on polyacrylamide gels. As shown i figure 8B, the addition of DMDA-pateamine A at any time post-infection completely prevented farther protein synthesis. Surprisingly, this effect was more pronounced in infected cells than in uninfected cells as shown by the continued synthesis, albeit at reduced levels, of host proteins in mock infected cultures. p)(i!i)2! To determine the impact of Pateamine A on the production of cell free virus during different temporal stages of infection, as abo ve, DMDA-Pateanirae A was added to infected cells at various times post-infection for a twenty four hour period, and the amount of virus in culture supernatants at the end of the treatment interval was determined by the TCID50 assay. Untreated samples were harvested at the beginning and end of each interval as controls. As with viral protein synthesis, the addition of D DA-Patearaine A at any time post-infection completely prevented the production of additional progeny vims, (Fig, 8C). This was true at all times post-infection, demonstrating that Pateamine A is capabie of inhibiting the HCMV lytic cycle at any stage of infection.

[OOOtOBj Together these data show that Pateamine A effectively stops the progression of the HCMV lytic cycle at any stage of the virus life cycle. These data are markedly differ from previous results using the είΡ4ΑΪ/1Ι inhibitor MppuristartoL While viral protein synthesis and replication become resistant to inhibition by hipparistanol during the later stages of infection, Pateamine A stops further progression through the lytic cycle at all times post infection. These data support the conclusion that the inhibitory effects of Pateamine A on the HCMV lytic cycle primarily reflect its impact on elF4Aiii function. These data also show that Pateamine A or similar compounds targeting eIF4AHI have the potential to be potent and effective anti iral drugs capabie of inhibiting infection at all stages of the viral life cycle

E sample 6: Inhibition of Herpes Simplex V it us type 1 (HSVl) Replication in Hnrnaa Fibroblast Ceils by Pateainirte

[000104! To determine the effect of pateamine A on HSVl replication, Vero cells were infected with HSVl at a multiplicity often, and the amount of virus in the culture supernatant was measured by plaque assay over a single round of virus replication (fig 9A). All. experiments contained in this Example and the corresponding figures utilized DMDA-pateamine A. DMDA-Pateamine A ( 100 ti ) was added to the superaatants when the inoculum was removed. When DMDA-pateamine A was added at the time of infection the production of infectious HSVl particles was completely inhibited. In fact the amount of HSVl in the supernatant fed below the levels found in the inoculum by the end of the time course. f 000} 5) Additionally, HSVl replication remained sensitive to DMDA-pateamine A throughout the virus life cycle. Vero cells were infected with HSVl , and exposed to DMDA -pateami ne A starting at 4, 8. 12, or 16 hours post- infection. The amount of virus present in the supernatants of each sample at 24 hours post infection was then determined by plaque assay. Untreated samples were also harvested at the time of drug addition. As 3 ί shown in figure 9Bthe addition of DMDA-pateamine A to rn&eted cultures at. any time post infection prevented any further accumulation ofHSVl particles. These data

demonstrate that much like HCMV infection, HSV1 replication is potently inhibited by Pateamine A treatment.

Exa mpfe 7: The Role of the EJ€ as a Host Bet r mi ami of HC Replica lion

[0001061 The EJC. a critical host complex that facilitates the nuclear export of spliced mRNAs, is not an important host determinant of ^' HCMV replication. While the eIF4AIH protein, is critical for efficient, virus replication, depletion of another core EJC component, roagoh, had a minimal effect on virus replication. elF4AHt expression is required progression from the immediate earl to early stage of the lytic cycle and for the efficient production of progeny virus. Likewise, in the absence of el.F4 Al II the export of viral mRNAs from the nucleus is inhibited. In addition a chemical inhibitor of elF4All.I, pateamine A, potently inhibits the replication of HCMV and the related alpliaherpesvirus HSVl. Importantly Pateamine A could inhibit viral protein synthesis and the production of progeny virus at any stage of infection, indicating an ongoing requirement for e.!F4AI! l throughout the virus lytic cycle. Together, results demonstrate that HCMV utilizes specific components of the host EJC to promote efficient virus replication. In addition, efF4Alll ts a novel target for the de velopment, of new antiviral therapeutics,

10001071 HCM V has evol ved to utilize specific aspects of the host EJC complex to facilitate the cytoplasmic accumulation of viral messages. Depletion of eIF4AIH decreases HCMV replication, potentiall by limiting the export of viral mRNAs from the nucleus. As a core component of the EJC eIF4Al.Il facilitates the export of spliced host mRNAs from the nucleus. However several lines of evidence suggest that elF4Ai l.l-mediaied export of viral mRNA is sot dependent on a canonical EJC. Depieiion of other EJC core components had a minimal impact on HCMV replication, protein expression, or the export of viral mRNAs to the cytoplasm., in addition the EJC is thought to specifically promote the export of spliced mRNAs from the nucleus, (Bono, F. & Gerhing, ..H. (2 1.1.) RN Bio I., 8:24-9). Both spliced and. unspliced viral mRNAs require the presence of eIF4AIH for their export.. Together these data suggest that an E.1C- independent function of elF4Al!I is utilized to export HCMV mRNAs from the nucleus.

[000108] eIF4AIII may facilitate the export of viral messages by forming a no vel complex hi HCMV- fected cells, potentially containing viral proteins. A likely candidate would be the HC V UL6 protein (plJL69). plJL69 associates with uaspliced viral mRNAs and facilitates their export to the cytoplasm via its interaction with host mRNA export factors such as UAP56. (Zieike, B. et al. (20.12) J. Virol., 86:7448-53). Interestingly UAP56 is an auxiliary component of the EJC that facilitates the export of unspiiced

HCMV mRNAs from the nucleus. (Gattleld, D. et ). (2001 ) Cnr BloL 11 : 1.716-21). it is therefore plausible that pUL69 also recruits el.F4AIII into a novel, export-competent rihoiUicieoproteni complex. f 0001091 Alternatively, host cells may contain an additional eIF4AIII-dependent mK A export pathway that does not require an intact EJC complex. This idea is supported by the recent finding that ©IF4AIH binds some cellular mRNAs at sites other than splice junctions. (Sauiiere, .1., ei al. (2012) Not. Struct. Mai. Biol. 19: 1. 24-31), The consequence of elF4A0I binding to sites other than exon boundaries is currently unknown. This could reflect a role for eIF4AIH in a separate protein complex, as the other core EJC components are thought to direct elF4AIIi to newly ligated splice junctions.

JOOOilOj The EJC can also function to stimulate the translation of its associated mRNAs in the pioneer round of translation. (Lejenoe, F. , et al (2002) EMBO J., 21 :3536- 45 )/ The pioneer round of translation functions as a quality control, checkpoint that monitors the integrity of recently exported cytoplasmic mRNAs. This initial round of translation is thought to be a prerequisite for entry of spliced mR As into subsequent elF4F~depeadent. translation which results in robust protein synthesis. (Chin. S.Y., et a.l. (2004) Genes Dev. , 18:745-54). The results above demonstrating that HCMV mRNAs efficiently associate with polysomes when eSF4A!II is depleted demonstrates that elF4AO ^' i is not required for the translation of HCMV mR s. Additionally, viral mRNAs may not utilize the pioneer round of translation. Alternatively the fact that the EJC component magoh is not required for efficient virus replication indicates that HCMV mRNAs may enter the pioneer round of translation in a no vel manner that does not require the presence of the full complement of EJC core proteins,

{00011.1 j The critical role for eiF4Aiii in HCM replication is further supported by results showing that a chemical inhibitor of eIF4AIII possesses potent antiviral activity. The natural product pateamine A was found to completely prevent further progression through the lytic cycle at any stage post-infection. Our finding thai Pateamine A has a 3$ similar effect on HSVl replication suggests that eiF4Aiil may be required for the replication of herpesviruses in general, although experiments addressing the impact of elF4Alil depletion and Paieamine A treatment on gatnmahcrpesvims replication.are needed to test this hypothesis.

[fl HH 12J While Pateamine A can inhibit all three eIF4 A R ^' NA heticases found in mammalian ceils (Dang, Y. ₅ et al (2009) « Biol. Chem. _y 284:23613-2.1 ), several lines of evidence support the conclusion that the effects of Pateamine A on virus replication, are primarily due to inhibition of elF4Aili. First, depletion of eIF4AlII similarly reduced HCMV replication. Second, HC V replication and protein synthesis become resistant, to the effects of the el F4Ai/Il -specific inhibitor hippuristanoi during the later stages of infection (Lenarcic and Moorman, unpublished observation). In addition, during the later stages of infection both, viral protein synthesis and replication are not affected by disruption of the e!F4F translation initiation, complexes, whose activity is dependent on elF4Ai/II helicase activity. In contrast both the production of progeny virus and protein synthesis were completely inhibited by the addition of Pateamine A at any time following infection. The temporal requirement, for eJF4AI/H activity together with the ability of Pateamine A to inhibit HCMV at all. times post infection suggest that the relevant target for the antiviral activity of Pateamine A is eiF4A.01.

[OOOtfBl Pateamiiie A has a more pronounced effect on the synthesis of the HCMV IE! protein than depletion of eIF4AEI (compare figures i A and ?A). This could be due to residual ei.F4.Ali.I present in the ei.F4.AI II -depleted ceils. However, this result more likely reflects the mode of action of Pateamine A. Pateamiiie A inhibits eIF4AIII by stimulating its R A binding acti vity. In the presence of Pateamine A elF4AIH cannot be released from its target mR As. (Bordeleau, M.E., et ai. (2006) Chem, Mil., 13: 1287-95). The bound eIF4AIII prevents the ribosome from, traversing the mRNA, and thereby prevents its translation. While technical difficulties have hindered our efforts to determine if elF4AliI associates with viral mRNAs, our finding that Paieamine A potently limits viral proteiii synthesis suggests a direct interaction. However it is also possible that Pateamine A inhibits the expression of a host protein needed for HCMV IE I translation. l<HM)11.4j The ability of Paieamine A to inhibit herpesvirus replication at nanomolar concentrations suggests that Pateamine A may be an effective anti viral compound. At the doses used in the above studies, Pateamine A was not toxic to primary human fibroblasts, but potently inhibited virus replication, la related studies, Pateamine A lias exhibited antiviral activity against diverse RNA and DMA viruses, suggesting a potential use for Pateamine A as a broad spectrum antiviral (Ziehr and Moorman, unpublished

observation). Pateamine A. is bioavai!able in vivo _y and is well tolerated in mice for periods up to one month. (Kuznetsov, G., et al. (2009) Mo/. Cancer Ther., 8:1250-60). Therefore, Pateamine A or similar elF4A!ll inhibitors may be clinically relevant.

Example 8: Attenuation of Chik«ngnnva virus (C V) aad Ross River virus (RRV) Replication by Pateamine A

{900115} To determine the effect of pateamine A on CHKV and RRV replication, Hela cells were infected with eithe CHKV or RRV at a multiplicity of infection of 10. Ail experiments contained in this Example and the corresponding figure utilized DMDA- pateamine A, Following virus adsorption, DMDA-pateamme A (100 nM) or an equal volume of vehicle con trol (DMSO) was added to the cel ls. Twenty four hours later the amount of infectious viral progeny in the supernatant was q uantified by plaque assay. Based on these studies it was determined that treatment with DMDA-pateamine A decreased the replication of the RRV and CHK V by greater than 100 fold. (Figure 10).

Example 9: Attenuation of Dengue Virus Replication by Pateamine A

( ⁽ HHM 1(5} To determine the effect of pateamine A on dengue virus replication, Hela cells were infected with dengue virus (serotype 2} at a multiplicity of infection of 3. All experiments contained in this Example and the corresponding figures utilized DMDA- paieamine A. Following viral adsorption, DMDA-patearnine A _. (100 «M.) or an equal volume of vehicle control (DMSO) was added to the cultures. The amount of dengue virus in cell free supernatants was quantified by the tissue culture infectious dose 50 assay

(TQDS0) at 72 hours post infection. j ^' WMil 17} Based on these studies, DMDA-paieamine A was found to profoundly attenuate dengue virus replication, decreasing the yield of ceil free infectious virus by over 1 ,000 fold. (Figure 1 1 ). Pateamine A also inhibited the replication of respiratory syncytial virus (RSV) and the parainfluenza virus { Pi V) in both tissue culture cells and in primary human airway epithelial cell cultures. Similarly, DMDA-paieamine A decreased the replication of human rfrinovtrus (RRV) by several orders of magnitude in hela celts. ( Figure 12,

ExaBjiplc _; ^' ¾- _; 0 _; ? _; jtt¾O _; _ _ Respiratory Synctyial Virus (RSV) & Para nffot&oza Virus (PIV) Replication I Pateatnioe A MHii l Sj Primary human airway epithelial cells (HAEC) grown in raft cultures were infected with RSV or PIV viruses encoding the green fluorescent protein (GFP) at a multiplicity of infection of five. All experiments contained iu this Example and the corresponding figures utilized DMDA-pateamine A. The presence of infectious virus was ^• determined by washing the apical surface of the cells with media, and then titering the amount of virus present in the wash by the TCID50 assay. The number of infected, ceils in the presence and absence of drug on eac h day was measured by fluorescence microscopy. The effect of DMDA-pateamine A on RS replication in H.EP2G cells was also assessed. B.EP2G cells were infected with RS at a multiplicity of infection of five, and the yield of infectiou progeny was determined by the TCID50 assay.

ΪΘ0Θ1Ι9| DMDA-pateamine A treatment accelerated the clearance of RSV from HAE raft cultures. Viral titers i the apical wash decreased more quickly (Figures 13 & 14), as did the number of infected, cells infected with RSV (Figure 18), The -amount of PIV in the apical wash did not increase over the test period (Figure 15), suggesting that the virus present in the wash was from the inoculum. Consistent with this interpretation, no GFP positive ceils were observed following PIV infection of HAE ceils suggesting that pateamine A prevented infection (Figure 16). In addition pateamine A reduced RSV replication in HEP2G cells by greater than 1000 fold at all times tested (Figure 1.7).

Example 11 : Inhibition of Human Rhinovirus (HRV)

Replication bv Pateamine A

1000120] To determine the effect of pateamine on HRV replication. He la ceils were infected with HRV strain .16 at a multiplicity of infection of five. All experiments contained in this Example and the corresponding figure utilized DMDA-pateamine A.

Following virus adsorption, cells were treated with DMDA-pateamine A ( 100 nM.) or an equivalent volume of vehicle control (DMSO). Cell free virus was harvested and titered by the TCID50 assay on He!a cell indicator monolayers at the 24 hour post infection. 1800121) Based ό» these studies, D DA-pateamine A treatment ^' significantly limited HRV replication, decreasing the amount of cell free virus fay greater titan 1000 fold.

Example 12: Inhibition of Hiiman Immunodeficiency Virus (HIV) Replication y

Pateamiae A

1000122) To determine the effect of HIV replication, SupTl cells were infected with HIV strain NL4 or mock infected in the presence or absence of patearame A (.100 am), Viral particles were quantified by p24 E S A at different times post infection. (Fig, ISA). All experiments contained in this Example and the corresponding figure utilized DMDA- pateamine. A. S ^' apTl cells were infected with HIV strain R3A or mock infected in the presence or absence of DMDA-patearai.ee A (100 nm). Viral particles were quantified b p24 ELISA at different time post, infection. {Fig, S.8B), HIV strain 3A induces extensive syncytia formation. Pateamine A completely inhibited syncytia formation. Syncytia were observed by light microscopy at 2 dpi. Error bars are derived .from duplicate samples. (Fig. 18C).

Example 13: Inhibition of HCMV Replication bv the eIF4A Inhibitor

Hir>¾>»r¾sfanol in Primary Human Cells

1000123 _. ) To determine the effect of hippuristanol on HCMV replication, primary human foreskin fibroblasts (HFFs) were infected with tire AD16 strain of HCM V at a multiplicity of infection (MOI) of three. Cells were harvested at 24, 48, 72, 96, or 120 hours post infection, and frozen at -80 degree Celsius until analysis, infected ceils were either treated with vehicle control .(DMSO) or with 100 riM hippuristanol dissol ved in DMSO. The amount of infectious virus present in. the stipernatants was determined by the TCID50 assay. The effect of hippuristanol on cell viability was determined using the LDH assay as described above. This concentration of hippuristanol was not toxic to HFFs over a treatment period of 5 days (Figure i 9).

Example 14: Toxicity of Hippuristanol in Human Cells and Ability to Inhibit Protein

in Human Foreskin Fibroblasts

1000124) To determine whether eIF4Ai.lI activity was required during HCMV lytic replication, hippuristanol treatment was used to inhibit eIF4A. |04M)12f>i Hippuristano.1 has previously been shown to be non-toxic to primary human ceils. (Bordeieau, M. E. _f et al (2006) Nat Chem Bial 2:213-20; Ltndqvist, L., et ai. (2008) PL S One 3:et 583; Lindqvist, L., & Pellctier, i (2009) Future Med Chem 1 :1709- 22; Pesiova, T. V., & Heilen, C.U, (2006) Nat Chem Biol 2:1.76-7; Tsumuraya, T,. C. et al. (2011 ) iochem Pharmacol 81 :713-22). To determine if the viability of primary human fibroblasts was affected by ppuristanoi treatment, confluent serum starved fibroblasts were treated with increasing concentrations ofhippuristanol, and the viability of the cells was measured, by the LDH assay at 5 days post treatment Concentrations ofhippuristanol greater than 250 nM had some effect on viability over the lime course examined, however no toxicity was observed with concentrations .ofhippuristanol of 100 nM or less (fig. 25), Therefore the 1 0 nM concentration, was chosen for use in the remaining experiments.

[iKM)l2i»j Previous studies have shown that hippurisratio! potently inhibits global protein synthesis in multiple cell types. (Bordeieau, M. E., et al. (2006) Nat Chem Biol 2:213-20). To confirm that hippitristanol also inhibits protein synthesis in HPFs, the amount of radiolabeled amino acids incorporated into TCA-insofuble protein in the presence ofhippuristanol was measured. ρκ ⁾ Μ27 Primary human foreskin fibroblasts (HFFs _. ) were passaged in DMEM {Siaraa-Aldrich, ^' St. Louis. United States! contitinine 10% newborn calf serum and used between passages 7 and 14. Unless otherwise indicated cells were seeded at confluence and then serum starved for 48 hours prior to infection. Vero and He!a cells were

maintained in DMEM with 1 % fetal bovine serum.. HCMV infections were performed with the BADinGFP strain (ADGFP). The previously described BADinGFP virus contains the green fluorescent protein ( iFP) open reading frame under the control of the SV40 promoter inserted, in the non-essential UL21.5 locus. HCMV infections were performed at a multiplicity of infection (MOI) of three in serum free medi unless otherwise noted. HSVl infections used the KOS strain at a multiplicity of 10 unless otherwise noted. Cell free HCMV -was titered by the tissue culture infectious dose 50 (TCTD50) method on primary uman fibroblasts, while HSV ί was titered by plaque assa on Vero cells as described previously. l<MM)128j At the indicated time post infection, the media was removed and replaced with 1 ml serum free DMEM lacking methionine and cysteine (cat#D0422, Sigma- Aldrich. St. Louis, United States). Following incubation for fifteen minutes, 125 μθ ' ^" - labeled methionme/eysteine was added directly to each well (Eas Tag Express Protein Labeling Mix; cat # EG772, Perkin Elmer, Waitham, Massachusetts). Where the effect of inhibitors was tested, the inhibitors were included in both the met cys free incubation and labeling periods. Ceil were labeled for 30 minutes, at which time the media was removed and the cells washed three times with ice cold PBS. Ceils were pelleted and stored at -BifC until analyzed.

1000129) Cells were infected with HCMV at a .multiplicity of three, and the inhibitors were added when the inoculum was removed. Cyclohextmide, an inhibitor of peptide elongation, was used as a positive control for die inhibition of protein synthesis. As an additional control, cells were Created with aetmomycin D, which also decreases protein synthesis by inhibiting raRNA transcription. To compare the results to previous studies, infected cells were treated with Torinl, an inhibitor of the mTOR kinase. Consistent with previous reports, Torinl inhibited the rate of total protein synthesis by 50% while actiaomycin D reduced protein synthesis by 60% (fig, 20D). Treating the cells with cycloheximide reduced total protein synthesis to 15% of the level in the untreated controls. Rippuristanol was as effective as cycloheximide, reducing the rate of protein synthesis to 15% of that observed, in the untreated control Together these results demonstrate that hippuristanol effectively inhibits the global rate of protein synthesis in HCMV infected ceils at a concentration that, is not toxic in primary human fibroblasts.

Example IS; Effect of Hippwr istan o¾ o Host Sign aling Path ways

1 ^' t!tMi!Mj To determine ifhippuris.tano.1 treatment inhibited eiF4F activity without affecting host signaling pathways, particularly mTOR activity, during HCMV infection., infected cells were treated with Torinl or hippuristanol at the time of infection, and mTOR activity was assessed six hours later by measuring the phosphorylation of the ribosomal protein 86 (rpS6>. Phosphorylation of rpS6 ha previously been shown to be entirely dependent on mTOR activity in. HCMV-infected cells. Consistent with previous studies, Torinl treatment decreased rpS6 phosphorylation in infected cells (tig. 2 A), in contrast hippurisianol treatment did not affect. rpS6 phosphorylation. This data demonstrates that the effect of ippuristanoi on protein synthesis is not due t off-target effects on. the mTOR signaling pathway.

Example 16: Effect of iltppnristanol on HCMV immediate Early (IE) Protein

Expression (000131) ϊο determine whether hippitristaool affected HCMV immediate early (IE) protein expression., IE. I protein expression was analyzed using quantitative reverse- transcriptase real-time PGR as described previously. Briefly, frozen ceil pellets were directl re-suspended in Triasol (Invittogen, Carlsbad, United Slates). The mixture was extracted with chloroform, and the RNA was precipitated from the aqueous phase with isopropanol. RNA pellets were resuspended in RNAse-free water and the amount of RN A was quantified using a NanoDrop spectrometer. One microgram of RNA was reverse transcribed with RT Master Mix (Roche, Basel Switzerland) using random hexarners as primers. Two microliters of the resulting cDN A was mixed wit gene specific primers SYBR. Green Master Mix and amplified in a using a Roche LightC eler 4800. The amount of viral transcript in each sample was determined using die AACt method., with GAPDH as the reference sample,

1000132) As previously shown, both aciinomycin D ami cyelohexiniide prevented ΪΕ1 protein expression when added following the removal of the inoculum. Consistent with our previous studies, the m ^' TOR inhibitor Tori.nl did not limit IB protein expression (Fig. 20B). Surprisingly hippuristanol treatment also did not limit the expression of TB I protein when added at the time of infection _* despite decreasing the global, rate of protein synthesis as efficiently as cyclohexi ide. Inhibiting elF4Al H activity may result in a compensatory increase in IE! mRNA transcription. However quantitative real-time PGR (qPCR) analysis of IE I mRNA levels revealed that hippuristanol did not affect the amount of IE i raR A present (fig. 20C). As similar amounts of IE I mRNA and protein were made in the presence of hippuristanol this result demonstrates that inhibiting elF4Al/II does not decrease the efficiency at which the IE. I mRNA is translated.

1000133) Previous studies examining the effect of m ^' TOR inhibitors on HCMV protein expression, demonstrated that IE and early protein were expressed despite significant disruption of the elF4F complex. To determine if both IE and earl HCMV proteins could be synthesized when e!F4Al, ^; il is inhibited with hippuristanol, HFFs were infected with HCMV at a multiplicity of three, and hippuristanol was added to the cultures when the inoculum was removed. Samples were harvested at 24 hour intervals following infection, and the expression of a representative IE (IE !), early (UL44), and late (pp28) protein was measured by Western blot. (0001341 Cell pellets were iyscd in RIPA buffer (50 inM Tns-HCL pH 7.4, 1% NP- 40, 0.25% sodium deoxycholate, 50 mM NaCl, im BDTA) containing protease inhibitors (complete EDTA-free; Roche, Basel Switzerland) and phosphatase inhibitors (NaF and NaVCti, ImM each). Cells were incubated on ice in RIPA buffer for 15 minutes and then cleared of debris by cen.rifugati.on for 5 minutes at 14,000 x g. The protein concentration of each sample was determined using the Bradford reagent. Equal amounts of protein, from each sample were resolved on 10% poly aery iarnide gels, transferred to Protran nitrocellulose membranes (Whatman, Springfield. Mills, England) and. then blocked for at least one hour in iris buffered saline containing 0.05% Tween 20 (TBS-T) containing 5% non-fat milk. .Primary mouse .monoclonal antibodies were diluted in TBS-T containing 1 %BSA and incubated with the membrane for one hour at room temperature. Rabbit polyclonal, antibodies were incubated overnight at 4 C in TBS-T plus 5% BSA. Following washing with PBS-T or TBS-T, blots were incubated with HRP-conjugated secondary antibodies for one hour at room temperature. Blots were again washed, in TBS- T and proteins visualized by treatment with ECL reagent (Amersham, Amersham England) and exposure to film. Antibodies specific for the following proteins used in this study include: IE! < f; I00), UL44 (1 : 100), eIF4Al ( 1 : 1000, Cell Signaling, Danvers, United States ), tubulin. (1 :5000, Sigma-Aldrieh, St. Louis, United States), pp28 (1 TOO), rpS6 (1 : 1000, Ceil Signaling, Danvers United States).

[00013. ! As shown in figure 21 A, the 1E.1 protein, was expressed throughout the HCMV lytic cycle in the presence of hippuristanol. The expression of the HCMV early protein UL44 (pUL44) was delayed and reduced by hippittistano! treatment, however the amount of pUT4 continued, to increase as infection progressed.

[O0O1 J In contrast, the expression of the pp28 late protein was significantly diminished in hippuristanol treated cells at all times post infection. To determine if the decrease in p28 protein levels was due to decreased transcription or translation of the pp28 mRNA, quantitative real-time PCR was used to analyze pp2S mRNA levels. Results revealed that hippuristanol significantly decreased the transcription of pp28 mRNA (fig. 21B). Together these data demonstrate that ciF4A. iI is not required for the translation of t} HCMV IE protein throughout infection, and suggest a partial requirement for eIF4AI/H from the translation of an HCMV early protein, in contrast, inhibiting eIF4Ai/ll from the start of infection significantly reduced the transcription of a late gene .mRNA. 4!

Example 17: impact of Inhibiting eIF4A Activity on Viral DNA Replication

1000137! Transcription of HCMV late TOR AS is dependent on viral DNA replication (Chua, C.C., Carter, TJ-1.. & St ieor, S. (1981 ). J Gen Virol 56: 1-1 I), and .bippuristan l reduced the transcription of pp28 Sate mRNA. To determine whether inhibiting eiF4Al/ll activity decreases vital DNA replication, the abundance of viral DNA in. cells treated with hippuristanol was measured, infected cells were treated with bippuristanol at the time of infection, and total DNA was isolated ninety-six hours later and quantified by qPCR, As a control, the level of viral DNA in cells treated with phosphonoaeteie acid (PAA) was measured, a well-described inhibitor of HCMV DNA replication. (Bird, R. M, et aL (19863/ Anti icrob Chemother 18 Suppl B:201-5).

( 01 81 HCMV DNA accumulation was measured essentially as described in

Moorman, NX, et al. (2008) Cell . Host Microbe,, 3:253-62, Briefly, HFPs were seeded at confluence in six well plates, and then serum starved for forty eight hours prior to infection. The cells were infected at an MO! of 0.05, and inhibitors (hippuristanol ^::: 100 nM; phosp ^' hotioacetic acid (PAA) ~ 200 pg'ral) were added following the removal of the inoculum. Control infected cells were treated with DMSO vehicle. Ninety six hours post infection the cells were scraped from the dishes, pelleted, and frozen at -80 degrees. Frozen ce!f pellets were resuspended m 100 lysis buffer and then digested overnight with Proteinase . (IGrng rn!). DNA was extracted front the !ysates with two phenoi cftloroform extractions, and the DNA was precipitated by the addition of an equal volume of isopropanoi. The precipitate was pelleted in a microcentrifuge (30 minute at I4.000g) and then resuspended in distilled water. The DNA was quantified using a Nano ^' Drop spectrometer. Five hundred nanograms of DN A was analyzed by quantitative real-time PGR (qPCR). The number of HCMV genomes in each sample was determined by comparing the results to a series of DMA standards containing from 10 ^s to 10* HCMV genomes. To control for variation in pipetting the results were normalized to the amount of total DNA in each samples by qPCR using primers specific for GAPDH.

(000139! Both Mppuristarjoi and PAA reduced viral DNA accumulation by two orders of magnitude (fig. 22A), This data show that inhibiting eIF4AI/il activity from the onset of infection prevents HCMV DN A replication. Thi data also suggests that the reduction in pp28 mRNA levels in the presence of bippuristanol was due to decreased viral DNA replication. fOOOi -tO] To determine whether reduced accumulation of viral DNA and a viral late protein in the presence of hippuristanol suggested that inhibiting eIF4Al¾I from the start of infection would inhibit HCMV -replication, the production of cell-free virus from hippuristanol-treated ceils over a single round of virus replication. Primary fibroblasts were infected at a multiplicity of three, and hippuristanol was added when the inoculum was removed. Supernatants were harvested at twenty four intervals following infection and titered on fresh fibroblast cultures. As shown in figure 22B, treating cells with hippuristanoi. at the time of infection reduced die y ield of cell-free progeny virus to the limit of detection of the assay (6.7 X \ ⁽ TODx/ml). This data shows that inhibiting elF4AMI from the onset of infection reduces viral DNA replication and the production of progeny virus, despite the expression of HCMV IE and early proteins.

Example 18: Off-Target Effects of Hipp«ristaao¾ on Other Host Enzymes

[000141] A concern with the use of chemical inhibitors is the potential for off-target effects. While hippuristanoi has been extensively characterized as a specific inhibitor of el.F4Al/U (Bordeleau, M. E., et ai (2006) Nat Chem Biol 2:213-20.), it remained possible that the phenotypes observed were due to an off-target effect of hippuristanoi on other host (or viral) enzymes. Therefore, shRNA-mediated depletion of eIF4A l as an additional approach was used to confirm that the observed phenotypes were specific to the inhibition 0:felF4A i.

[000142] JLentivirus shRNA. expression constructs targeting el.F4Al were obtained from the University of North Carolina Lentivims Core Facility. JLentivirus stocks were prepared by transacting 293T cells with icntivsrus shRNA construct together with packaging vectors. Three days post transfcetion eel! free supernatants were harvested and filtered through a 0.45mm filter. One day prior to transduction HFFs were seeded at confluence into 6 well plates. The ceils were transduced with lentivims overnight in the presence of 5 mg ral polybrene. The next morning the media was removed and the cells were washed three times with PBS. Serum free media was added to the wells, and. the ceils were incubated for an additional 48 hours prior to infection with HCMV as described above. 1000143} Confluent fibroblast monolayers were transduced with leutiviruses expressing either e!F4A ! -specific or scrambled shR As. The transduced cells were then, infected with HCMV and the expression of representative IE, early, and late proteins was measured throughout a time course of infection. Similar to hippuristanol treatment, an IE artd early protein were expressed in efF4Al -depleted ceils, while the expression of a viral Sate protein was diminished (fig. 23A). Limiting e!F4AI expression prior to infection also, decreased the yield of ceil free virus by approx. 150 fold (Fig, 23B). shRNAs also may have non-specific effects due to the presence of similar target sequences in other mRNAs, To ensure that the specificity of the above results, the experiments were repeated using a different e.lF4A~speci£ie sh.RNA, and similar results were obtained (Fig. 23.8 and data not shown). These data confirm the specificity of the results obtained in hippuristanol. treated cells, and support the conclusion, that eIF4AI/H activity is required at the start of infection for efficient HCMV lytic replication.

Example 19: The Role of eiF4A in the Translation of HCMV Late mRNAs

[0001 4} Recent studies of the role the m ^' TOR kinase revealed that viral protein synthesis becomes increasingly resistant to mTOR inhibition as infection progresses. (Ciippingei A,J„ Magsire, T.G., & Alwine, 3.C. (201 i ) ../. Virol., 85:3930-9). Therefore, additional experiments were performed to address the temporal requirement for eIF4AS i! activity for the translation, of HCMV mRNAs, f 0001 5} To determine if the requirement for elF4Al 0 activity for viral rnRNA translation changes during different temporal stages of infection, viral protein expression was measured when hippuristanol was added to cells at different times post infection. Hippuristanol was added to infected ceils for a twenty four period at different times post infection (e.g from 24 to 48 hours, 48 to 72 hours, etc.). For comparison, vehicle treated cells were harvested at the time of hippuristanol addition, and at the end of the 24 hour test period, HCMV protein expression was then measured by Western blot. At no time did hippyrisianol affect the expression of the HCMV IE I protein (fig. 24A), This is consistent with the results in figure 20B showing that ΪΕΙ protein synthesis occurs when eiF4Ai/ll activity is inhibited from the start of infection. 600146} When hippuristanol was added to cells beginning at 24 hours post infection, expression of the HCMV early protein pUL44 increased during the 24 hour hippuristanol treatment (fig. 24A), demo.r¾stratmg that translation of the p UL 4 RNA occurred when eIF4A was inhibited. However pUL44 expression was reduced in hippuristanol-treated ceils compared to untreated cells, suggesting that both el ^' F4A-depcndent aad independent rnRNA translation contribute to pUL44 expression at this stage of infection. A more modest effect on pUL44 expression was seen when cells were treated with hippuristanol. from 48 to 72 hours post infection, and no effect on the steady state levels of pUL44 were observed when hippuristanol was added after 72 hours post infection. >iH47j A similar trend was seen for the HCMV late protein pp28. Hippuristanol ^• delayed the onset of pp28 expression when added at 24 hours post infection (fig. 24 A). pp28 accumulated in the presence, of hippuristanol between 48 and 72 hours, although its expression was reduced compared to ^' untreated cells. Much like with pUL44, adding hippuristanol after 72 hours of infection had minimal effect on the steady state levels of pp28 observed at 96 hoars post infection (fig. 24A). Together these data demonstrate that elP4Ai -independent mRNA translation contributes to the synthesis of HCMV proteins. These data suggest that the ability to be translated in the absence o eIF4AI/H activity is a general property of viral mRNAs during the later stages of infection, as representative viral proteins from each kinetic class accumulated in the presence of hippuristanol. These data also suggest that the elF4AI<ll- independent translation of viral mRNAs becomes more prominent as infection progresses.

JCHW148) The above results suggest that significant degree of HCMV protein synthesis occurs when e!F4A! II activity if inhibited during the later stages of infection. However an alternative explanation could be that the viral proteins observed at the end of the treatment period were made prior to addition of the drug. To distinguish between these possibilities, we compared the amount of viral protein, present following a 24 hour treatment with either cyeloheximide or hippuristanol. As the addition of cyeloheximide prevents new protein synthesis, the levels of viral proteins are expected to decrease over ^• rime in accordance with the ^' half-life of the viral protein. 00eJ 9 ^' j To determine the degree of nascent viral protein synthesis when eIF4AI/SI activity was inhibited, the amount of viral proteins present following a 24 hour hippuristanol treatment was compared to the amount present following an identical cyeloheximide treatment. At forty eight hours post infection infected cells were treated with cyeloheximide or hippuristanol for 24 hours, and the expression of representative immediate early, early, and late HCMV protein was determined b Western blo ^' at the end of the 24 hour period. Vehicle treated cells harvested at the time of dru addition and at the end of Ac 24 hour period were used as controls for the amount of each protein synthesized during the treatment period. The levels of the IE I protein did not change appreciably in cycloheximide treated cells, suggesting that IEI has a half-life greater than approximately 18 hours (fig. 24B). In contrast, 1E2 levels were significantly reduced following 24 hours of eyclohexircride treatment. However IE2 was clearly present in the hippuristanoS treated samples, indicating that nascent 1.E2 was synthesized when elF4Ai/ll activity was inhibited, Similarly the abundance of the pUL44 early protein and the late proteinpp2H increased over the 24 hour incubation, in cells treated with hippuristanol, although the increase was less than that observed in untreated cells (fig. 24B). These results support the conclusion that HCMV proteins can be translated in the absence of eIF4A activity during the later stages of infection, and that e1F4A!/H-independent tn NA translation contributes to the expression of viral proteins from each kinetic class. Howeve maximal expression of HCMV proteins during the early stage of infection requires eiF4A activity. j uti!S ] As an additional confirmation of the above results, the amount of

etabolieaily labeled viral proteins synthesized in the presence of hippuristanol was measured. j 000151) Ceils were infected with HCM V ^" for seventy two hours, and then treated for sixteen hours with hippuristanol or cycloheximide. Radiolabeled amino acids were included in the media during the final thirty minutes of drug treatment to label newly synthesized proteins.

|0tRSl$2j Immune complexes specific for the indicated viral proteins were isolated from the infected cell lysates, and the newly synthesized protems were visualized by autoradiography. ^' 'S-!abe!ed ceils were iysed in 1 0 μί RIP A buffer, and the protein concentration of the iysate determined by Bradford assay (cat# 500-0205, BioRad, Hercules, United States). Fifteen microliters of ceil extract was mixed with 0.1 ml of 1 nig/mi BSA. containing 0.02% (w/v) sodium azide (NaN3). One milliliter of 20%

trichloroacetic acid (TCA) was then added to the sample, and the mixture was vortexed and incubated on ice for 30 minutes. The precipitate was then vacuum filtered onto 2.5 cm glass microfiber filters (cat# 1820 024, Whatman, Springfield Mills, England). The filters were washed twice with 20% TCA, once with 100% ethanol, and then allowed to air dry for 30 minutes. Filters were transferred to vials containing 5 ml scintillation fluid and radioactivity measured using a scintillation counter. The amount of precipitated

radioactivity in each sample was normalized to the total amount of protein present.

[880153! Radiolabeled amino acids were visualized by resolving 10 ig of protein on 1 % SDS-PAGE gels. Gels were fixed for 30 minutes in 7% acetic acid/25% methanol, and then dried for one hour in a BioRad gel dryer. (BioRad, Hercules, United States). The dried gels were exposed to film to visualize radiolabeled proteins.

[000154! As shown in figure 24C, similar amounts of radiolabeled IE1 , pUL44, and pp28 were recovered from hippuristanoi treated as from untreated cells, In contrast cyeioheximide completely prevented the synthesis of each viral protein. This demonstrates that viral proteins continued to be synthesized during the later stages of infection when eiF4Ai/lf activity was inhibited with hippuristanoJ.

[000155 _. ! To determine the effect of hippuristano! on virus yield over a twenty four hour treatment window throughout a time couxse of infection, HCM V infected cells were treated with hippuristanoi at 2 , 48, 72, or 96 hours post infection, and cell free infectious virus was quantified 24 hours later. Samples were also collected from untreated, ceils at both the beginning and end of the treatment period. The amount of virus present in the supernatant of all samples at 4$ hours was below the limit of detection (fig. 25A).

However by 72 hours the amount of virus in the supernatant of untreated ceils had increased, by greater than .100 fold. Whe hippuristano! was added, at 48 hours post infection, the amount of virus present in the supernatant at 72 hours was decreased by 5 fold. When compared to the amount of virus present at the time hippuristanoi was added to the culture it was clear that HCMV replication had. occurred, albeit to reduced levels, in the presence of .hippuristanoi. In addition, when hippuristanoi was added at later times in infection (e.g. 72 or hours post infection) no effect on HCMV replication was observed (fig. 25 A). These data suggest that there is a minimal requirement for eIF4Ai/iI activity for the production of infectious HCMV virions after forty eight hours post infection.

[000156] One explanation for the above results could be that infected cells are rreessiissttaanntt to to tthhee eeffffeeccttss ooff hhiippppuurriissttaannooii o onn to tottaall pprrootteeiinn ssyynntthheessiiss.. T Too ^' d deetteerrmmiinnee tthhee e effffeecctt: ooff eelIFF44AAII//llii i innhhiibbiittiioonn o o»n ttoottaall pprrootteeiinn ssyynntthheessiiss aatt d diiffffeerreenntt ttiimmeess ppoosstt i innffeeccttiioonn,, i innffeecctteedd cceelillss wweerree ttrreeaatteedd w wiitthh hhiippppuurriissttaannooii ffoorr aa t twweennttyy ffoouurr h hoouurr ppeerriioodd a ass aabboovvee,, aanndd n naasscceenntt p prrootteeiinnss wweerree m meettaabboolliiccaalllly llaabbeelleedd dduuririnngg tthhee ffiinnaall tthhiirtrtyy m miinnuutteess o off h hiippppuurriissttaannooii t trreeaattmmeenntt. TThhee pprrootteeiinn c coonntteenntt ooff eeaacchh ssaammppllee wwaass q quuaanntitiffiieedd,, aanndd. tthhee llaabbeelleedd pprrootteeiinnss wweerree pprreecciippiittaatteedd ffrroomm t thhee s saammpplleess wwiitthh t trriicchhlloorrooaacceettiicc aacciidd,, AA --ppoorrttiioonn ooff tthhee s saammppllee wwaass rruunn oonn aann SSDDSS--PPAAGGEE ggeeil ttoo vviissuuaalliizzee t thhee pprrootteeiinnss s syynntthheessiizzeedd iinn tthhee pprreesseennccee ooff

hhiippppuuririssttaannooii.. TThhee rreemmaaiinnddeerr ooff eeaacchh ssaammppllee wwaass ttrraannssffeerrrreedd ttoo ggllaassss ffiilltteerrss aanndd tthhee t toottaall aammoouunntt o off r raaddiiooaaccttiivvee aammiinnoo aacciiddss iinnccoorrppoorraatetedd iinntoto p prrootteeiinn wwaass ddeetteerrmmiinneedd iinn aa s scciinnttiillllaattiioonn ccoouunntteerr.. AAss oobbsseerrvveedd pprreevviioouussllyy,, h hiippppuurriissttaannooii ttrreeaattmmeenntt i innhhiibbiitteedd ttoottaall p prrootteeiinn ssyynntthheessiiss iinn u unniinnffeecctteedd cceeiillss b byy ggrreeaatteerr tthhaann 7700%% ((ftiigg.. 2255BB)),,

[[iiKKMM))l15577jj H Hiippppuurrtissttaannooii aallssoo eefffefeccttiivveellyy iinnhhiibbiitteedd t tootteal! pprrootteeiinn ssyynntthheessiiss wwhheenn aaddddeedd ttoo iinnffeecctetedd cceellllss 2244,, 88,, oorr 7722 hhoouurrss ppoosstt i innffeeccttiioonn.. AAtt 9966 hhoouurrss aanndd bbeeyyoonndd,, ttoottaall pprrootteeiinn s syynntthheessiiss bbeeccaammee iinnccrreeaassiinnggllyy rreessiissttaanntt ttoo iinnhhiibbiittiioonn bbyy hhiippppuuririssttaannooii,, wwiitthh aapppprrooxxiimmaatteellyy 5500%% ooff pprrootteeiinn s syynntthheessiiss iinn iinnffeecctteedd cceellllss bbeeiinngg rreessiissttaanntt ttoo h hiippppuurriissttaannooii b byy 112200 hhoouurrss ppooss i innffeeccttiioonn.. HHoowweevveerr tthhiiss mmaayy r reefflleecctt tthhee o obbsseerrvvaattiioonn tthhaatt ttoottaall pprrootteeiinn ssyynntthheessiiss iinn uunnttrreeaatteedd iinnffeecctteedd cceellllss ddeecclliinneedd ssoommeewwhhaatt aatt tthhiiss ttiimmee ppoosstt iinnfefeccttiioonn.. W Whheenn tthhee nneewwllyy ssyynntthheessiizzeedd pprrootteeiinnss wweerree vviissuuaalliizzeedd oonn aaccrryyiiaammiiddee ggeellss,, ddiiee ppaatttteernrn ooff pprrootteeiinnss

ssyynntthheessiizzeedd iinn i innffeecctteedd cceellllss wwaass cclleeaarrllyy ddiiffffeerreenntt ffrroomm tthhaatt oobbsseerrvveedd iinn u unniinnffeecctteedd cceellllss,, iinntteerreessttiinnggllyy,, iinn HHCCMMVV--iinnffeecctteedd c ceellllss aa ssuubbsseett ooff pprrootteeiinnss ccoonnttiinnuueedd t too bbee e efffificciieennttllyy ssyynntthheessiizzeedd iinn tthhee p prreesseennccee o off h hiippppuurriissttaannooii ((fifigg.. 225588)).. T Thheessee pprorotteeiinnss wweerree oonnllyy oobbsseerrvveedd iinn iinnffeecctteedd cceellllss,, ssuuggggeessttiinngg tthhaatt tthheeyy aarree vviirraall pprrootteeiinnss.. TThheessee ddaattaa ddeemmoonnssttrarattee tthhaatt eeiiFF44AAll ^'' iiXX--iinnddeeppccnnddeenntt mmRNNAA ttrraannssllaattiioonn ooccccuurrss tthhrroouugghhoouutt H HCCMMVV i innffeeccttiioonn,, aanndd ssuuggggeessttss tthhaatt vviirraall m mRRNNAA mmaayy aaccccoouunntt ffoorr tthhee m maajjoorriittyy ooff tthhee eeIIFF44AAII llII--iinnddeeppeennddeenntt ttrraannssllaattiioonn oobbsseerrvveedd..

(HSVI)

[000158] A previous study reported, that inhibition of elF4Ai/ll with hippuristanoi resulted in cell-type specific effects on herpes simplex virus (HSVI) protein synthesis. (Dauber, B., PeHeticr, J ^' ., & Smiley, J JR.. (201 1) . Virol, 85:5363-73), This study demonstrated that relatively low concentrations of hippuristanoi (less than IOC) nM) inhibit the expression afHSVl laic proteins in Hcla cells, but not in Vero cells. Surprisingly, both ΪΕ and early HSV 1 proteins were not affected by hippuristanol treatment in either ceil type (Id), suggesting that elF4Al/Il activity may be dispensable for the translation of at least some HSVl mRNAs. Based on the similarities between these results, the requirement for eIF4AI/!I activity tor protein synthesis during HSVl infection was further investigated.

[00015*1 First, the effect of hippuristanol on the production of HSVl progney virions in Vero cells, was determined, which was not assessed in the previous study. Treatment with 100 nM .hippuristanol immediately following infection had no effect: on HSV 1.

replication in Vero cells. This is consistent with the previous finding that concentrations of hippuristanol in excess of 1000 nM are necessary to inhibit HSV late protein expression in these ceils. Consistent with this previous study, we found thai 100 nM hippuristanol had no effect on the accumulation of HSV 1 in the culture supernatant of infected Vero cells (fig. 26 A). However in primary human fibroblasts a 100 nM dose of hippuristanol potently suppressed HSVl replication, decreasing the yield of virus by over 1000 told (fig. 26B).

[000160] To determine whether hippuristanol does not inhibit protein synthesis in Vero cells, but is effective at limiting protein synthesis in human fibroblasts (fig. 19D), the rate of protein synthesis was measured in Vero cells treated with either hippuristanol or the peptide elongation inhibitor cyclohexirnidc. Uninfected Vero cells were treated with the drugs for 5.5 hours and then radiolabeled amino acids were added to label newl synthesized proteins. Surprisingly, 100 nM hippuristanol did not affect th . rate of protein synthesis in Vero cells (data not shown), in contrast cycloheximide decreased the total rale of protein synthesis by greater than 85%. This result provides a _. potential explanation fo the cell type specific effects of hippuristanol previously described for HSV, namely that Vero cells are resistant to the effects of hippuristanol on total protein synthesis.

Examp e 22: elF4A Sabuntt is Critical for HCMV Replication

[0001611 The elF4Al/0 subunit of the eIF4F complex plays a critical rol in HCM V ^• replication. When elF4AI/lI is inhibited from the start of infection, HCM V replication is significantly decreased. The inhibition of virus replication is due primarily to a block in the immediate early to early transition. While the HCMV lEi and IE2 immediate early proteins are efficiently expressed when elF4Al/ii is inhibited, the expression of a viral earl protein ίρϋΐ.44) was delayed and reduced when hippuristanol. was added at the time of infection, although its expression was diminished. Consistent with Ms finding, when elF4AI/Il was ta bited viral DMA synthesis and late protein expression was significantly reduced. Similar results were observed when el ^" F4Al was depicted prior to infection demonsfcafcmg the specificity of the observed phenotypes for inhibition of eIF4AI IL A. temporal analysis of the role of elF4ALTl throughout HCMV infection revealed an increasing degree ofeIF4AI n-independent translation of viral mRNAs as infection progresses. An HCMV IE, early, and late mRNAs could each be translated when

elF4A]/ii was inhibited after their transcription had commenced. Total protein synthesis in infected cells remained sensitive to elF4AI/K inhibition until the final 48 hours post infection, at which time an increasing degree of eiF Al/ii-inde/pendent mRNA translation was observed. In addition, viral replication became resistant to eIF4Ai/Il inhibition as infection progresses. Togethe these data indicate that elF4Ai/lI activity is required at the start of infection for the efficient progression through the HCMV lytic cycle. However, HCMV mRNAs can be translated in the absence of eiF4Al ll activity, especially during the later stages of infection. Together these resuits suggest that a novel, mode of translation initiation occurs daring the later stages of herpesvirus infection cells that allows for the continued translation of viral mRNAs in the absence of host translation factor activity.

[<XH)162j The studies began by examining the requirement for ei.F4Al/TI activity for the translation of an HCM V immediate early mRNA. Given the important role of el.F4Al/H in host protein synthesis (Cencic, R., Gahcia-Vazquez, O., & Pelletier, (20! 2) J. Methods Emymol 51 S :437-61), it was surprising to ind that the HCMV IE I protein was efficiently expressed when eIF4AI/il was inhibited. The le vels of!El mRNA and protein were equivalent in hippuristano!-treated and untreated cells (fig. 20B, C), demonstrating that the expression of IE i was not the resul of increased transcription, of IE I -mRNA when e!F4A iwas inhibited. Depletion of elF4AI prior to infection confirmed the eIF4A- independent translation of IE 1 (fig, 2.3) . The efficient expression ofl ^' El was particularly surprising as hippuristanol inhibited total protein synthesis as effectively as

cycloheximide, indicating that the drug efficiently disrupted cap-dependent translation (fig. 2 D). While it is possible that the observed IE! protein was expressed prior to the full effect of the drug on protein synthesis, this seems unlikely for three reasons. First, the expression of !El is still increasing at six hours following infection. Therefore even a partial inhibition of elF4A i activity should result in a reduction in. IE! levels if translation of the ΪΕ1 mRNA required elF4Al/H ^" activity. Second, similar results were obtained when ceils were pre-treaicd with hippuristanol for 1 hours prior to infection (not shown). Third, prior depletion of eIF4A I did not limit HCMV IE I protein expression (fig. 23). Together with oar previous data demonstrating that HCMV IE! is efficiently

expressed wh n the eIF4F complex is disrupted by mTOR inhibitors (Moorman, N. J., & T, Shenfc, T (2010) J Virol 84:5260-9), these data strongly suggests that the HCMV IE1 mR A utilizes a mode of translation initiation that is distinct from that used by most host niRNAs. This may reflec a role for viral proteins delivered in the tegument in the translation of the HCMV 1E1 m ^' RNAs. Alternatively this may reflect an inherent property of the IE I raRNA itself. Additional studies are ongoing to address these possibilities. 0Θ363] ^' When eIF4Al/H is inhibited from the start of infection the expression of early proteins ts delayed and reduced, and the expression of HCMV late proteins is decreased (fig. 2.1). Similar results were obtained when eiF4A was inhibited with hippuristanol or depleted with eIF4A.l shRNAs (fig. 20A and 28, respectively),

demonstrating the specificity of the observed pheuotype for eIF4AI H. The continued expression of HCMV early proteins may be due to incomplete inhibition of elF4AI/li in the ease of hipporistanol, or incomplete depletion o IF4AI by the shRNAs. However in both cases HCMV early proteins are expressed, albeit at reduced levels and with delayed kinetics. These data suggest that during the early stage of infection both elF4A] ^' /Il- dependent and independent mechanisms exist for the translation of viral proteins,

| (HW164) ^' inhibiting eiF4AITI. with hipporistanol from the start of infection prevents viral DNA replication (fig. 22). riippuristanol was as potent at inhibiting viral DNA accumulation as PAA, a well-characterized inhibitor of viral DNA replication. (Bird, R. M., et al. ( i 986) J Antimicroh Chemother 18 Suppl 5:201-5; Chiou, H. C, Weiier, S.K., & Coen, D.M. (1985) Virology 145:213-26; Coen., Ό. M., et al. (1984) J Viral 49:236-47). How eIF4AI/U acti vity contributes to viral DNA replication is currently unclear, although several explanations exist. Perhaps eI.F4Ai/ll activity is required for the expression of a subset of viral proteins needed for viral DNA replication. Nine HCMV proteins arc required for viral DNA replication in vitro. ( agele, D„ et al. (2012) Virology 424: 106-14; Pari, G. S. (2008) Curr Top Microbiol Immunol 325:153-66; Pari, G. S., acica, M.A. & Anders, D.G.. (1993) J Virol 67:2575-82). While the essential DN replication protein p ^" UL44 is expressed when eiF4AI<Ti is inhibited, it is possible that one or more of the other essential proteins requires eIF4AI/ll activity for their expression.. While beyond the scope 5 ! of the current study, the generation of antibodies to additional essential viral DNA replication factors will allow for the determination of the requirement for eIF4AI/iI in the expression of critical viral ^" DMA replication proteins. Another explanation could be that inhibiting eIF4AI II prevents the expression of a host protein required for viral DNA

.replication. When e!F4Ai II is inhibited, expression of the hypothetical host factor would be lost, resulting in diminished viral DNA replication. A global analysis of the translation efficiency of host proteins when elF4Al/ii is inhibited may identify host factors critical for viral DNA replication whose expression depends on eIF4AI/D. activity for their

expression. 0036 j hi th initial experiments we s tudied the effect of inhibiting elF4Ai lI from the start of infection. However this approach was not amenable to defining the role of CIF4AMI during the late stage of infection, as inhibiting elF4Al/fl ^' activity blocked the early/late stage transition by inhibiting viral DNA replication. Therefore, a series of experiments in which we examined the requirement for el.F4Al/II activity for HCMV protein synthesis and replication during discrete temporal stages of HCMV infection were performed. The results indicate tha the requirement for el.P4Ai/II activity for HCM V replication changes as infection progresses. Prior depletion of e!P4AI or inhibiting elF4A l at the start of infection significantly reduced the production of progeny virus (figs. 2 1 B & 238). However the effect ofelF4Af/H inhibition on viral replication is minimal following the onset of viral DNA replication (fig. 25A). This suggests that viral proteins necessary for replication, assembly, and egress can be translated through an alternative mechanism that is eiF4A~independent. The finding that host synthesis relies on ei ^' F4F activity throughout infection suggests that the factors controlling assembly and egress are primaril of a viral nature. However host factors that are translated in an e!F4A l-independent manner or that have long protein half-lives may also be involved.

The decreasing requirement for eIF4AI/Il activity for HCMV replication correlates with an increasing level of elF A Wl-indcpendent protein synthesis as infection progresses. The eiF4 At/li-mdepeadent translation of viral mRNAs is evident both, at the level, of total protein synthesis (fig. 258) and in the expression of specific viral, proteins (fig. 24). Our results indicate that later in infection representative viral proteins from each kinetic class (IE, early, and late) can be translated when eIF4Ai/il is inhibited. This suggests that a general mechanism exists for the translation of viral mRNAs that is distinct from the canonical ei.F4F-dependent pathway. Our data suggest that ctF4At/H- independent translation is restricted to viral raRNAs, as total protein synthesis is significantly inhibited by elF4Al/Ii inhibition at all times post infection (fig. 25B). While this ma reflect some unique feature common, to viral raRNAs, we find this explanation unlikely as HCMV raRNAs are not known, to contain common 5' or 3' untranslated regions. However the 5' and 3' ends of the majority of HCMV mRNAs are unmapped, allowing for the possibility that post-transcriptionai processing of viral. mRNAs results in the presence of common sequences on HCMV messages. A global analysis of HCMV message structure is needed to rule out a role for a common sequence element in the eif4Ai.THndependeiif translation of viral mR As.

1090167} Studies of the effect of eiF4AI/0 inhibition on HSVI replication and protein synthesis revealed similarities between HSV I and HCMV. Both viruses require elF4AI/II activity at the onset of infection in order to efficiently progress through the virus lytic cycle. When considered in. light of recent studies demonstrating HSVI protein, synthesis in the presence of potent niTO inhibitors (Walsh, D. _s & Mohr, I. (2004) Genes Dev 18:660-72), these results suggest that elF4Al/ff-independent translation of viral m ^' RNAs ma ^' be a common feature of alpha and beraherpesviras infection.

{000168} These data also provide an explanation for the recent observation that, in some cell types hippuristanol did not affect HSV I protein synthesis. (Dauber, B., Pelietiet, J. & Smifcy, J.R. (2011 ) Virol 85:5363-73). Much higher doses (ca. 10 fbld)o.f hippuristanol were required for limit HS I late protein synthesis in Vero cells than in Heta. cells. We find that hippuristanol does not. inhibit total protein synthesis in uninfected. Vero cell at a dose (! 00 nM) which potently limits protein synthesis in HFFs and other cell types. This correlates with the inability of hippuristanol to inhibit HSVI replication in Vero cdls. These data, suggest thai the differential effects of hippuristanol on HSV I replication is due to an intrinsic difference in the host cell, rather than a difference in the biology of HSV 1 in different ceil types. j 00169} These results raise the larger question of how HCMV mR As translated when eiF4AI/II is inhibited. As discussed above, perhaps a common feature of HCMV raRNAs allows for their translation in an eIF4AI. II and eIF4F-½dependenfc manner. Host mRNAs that have short or unstructured 5' UTRs are thought to have a decreased requirement for the eIF4F complex for their translation. Perhaps HCMV mRNA on the whole have very short ^' and/or unstructured 5' UTRs, or use alternative 5' UTRs during the later stages of infection. The 5 'UTRs of the viral mRN As examined in this study have previously been defined during the late stage of infection. Based on in vitro folding algorithms, both the IE1. and pp28 mRNAs have 5 'UTRs with moderate to significant potential to form secondary structure (IE1 ^~ 13$ bp, AG -43.7 keal mol; pp2S ^:::: 476 or 232 hp with AG of -175 or -76 kcal/mo , respectively as determined using the mFOL ^' D algorithm). In vitro studies have demonstrated a requirement for eIF4A activit for the translation of mRN As with even weak secondary structure (>AG -6.7 keal/mol). Based on these studies both viral messages should require the helicase acti vity of elF4AI/S S for their translation. Again, a better understanding of the structure ofHCMV mRN A 5' UTRs is needed to address this issue.

1000170] One possible explanation for the translation ofHCMV mRNAs when elF4Al/II is inhibited could be the existence of a host dF4AI/H-radependent translation pathway. Several recent studies in mammalian cells have suggested that the eIF4F complex may only -regulate the translation of a subset ^' host mRNAs. (Hsieh, A. C, et al. {2012} Nature 485:55-61 ; Thoreen, C. C, et al. (2009 J Biol Ch in 284:8023-32).

Surprisingly these studies found no correlation, between 5UTR length or sequence and the requirement for the elF4 ^' F complex for translation. Rather the requirement for elF4F closely mirrored the presence of a TOP (t ract of pyrimidienes) motif in the eap-proxima ^■ nucleotides of the mRNAs. (Hsieh, A. C, et al (2012} Nature 485:55- 1 ; Thoreen, C. C, et al (2012) Nature 485: 1 9-13), TOP or TOP-like motifs have not been described in HCMV transcripts, although relatively few HCMV mRNA 5 ^' ends have been precisely mapped. The study of HCMV mR A translation may provide a tractable system for identification of novel host translation initiation complexes.

[0(1 { 7! j An alternative explanation may be that viral proteins exist that functionally substitute the eIF4F complex or the components of the elF4F complex in the initiatio of translation on viral messages. HCMV encodes greater that} 200 predicted open reading frames (Murphy, E. & Shenk. T. (2008) Curr Top Microbiol Immunol 325: 1- 19), and the molecular function of the majority of the encoded proteins remains undefined, it is possible that one or more viral proteins facilitate the recruitment o f host ribosomal subuiitts to viral mR As. Viral proteins encoded by HSV I and KSHV have been found to associate wit mRNAs (Swaminathan, S. (2005) Cell Bi chem 95:698-711; Taddeo, B„ Esc ne, A & Roiz an, B. (2004) Biochem Soc Tram 32:697-701), mRNA caps (Walsh, D. & Mote, i (2006) Genes D v 20:461 -72), and ribosomes (Mulvey, ML, ei at (1999) J Virol 73:3375-85), suggesting the existence of ^' herpesvirus proteins that promote translation initiation on viral m NAs. One HCMV protein, pUL69, has been shown to associate with mRHA m ^' G caps, viral mRNAs, and poiysaraes in infected cells. (Aoyagi, M, Caspar, M <& Shenk, T. ^' E. (2010) Proc NatiAcad Scl USA 107:2640-5). These properties of pU.L69 suggest that it could participate in translation initiation on viral mRNAs. However HCMV likely encodes additional proteins that interact with and influence the translation machinery in infected ceils. A more comprehensive

understanding _. of HCMV proteins that interact with the host translation, machinery will be required determine how or if viral proteins gover the translation of HCMV mRNAs during intcciion.

[©001721 The current study was conducted to directly assess the role of the host elF4AI/H RNA helkase in HCMV mR A translation. Using a combination of

phannacological and biological manipulations to inhibit eIF4Al/H, an essential subunit of the clF4P complex, we find that elF4Ai 0 activity is not an absolute prerequisite for

HCMV protein synthesis. This suggests that our current understandin of the factors that control m ^' RNA translation during HCMV infection is incomplete. However they also demonstrate that at least some thai viral mRNAs can use an alternative elF4Ai/ff~ independent mechanism for their translation. This suggests die existence of additional host or viral factors that provide functions critical for the translation of viral proteins fO©eS73 ^' | These example demonstrate that the natural compound, pateamioe A, an inhibitor of all isoforms of the cellular UNA he!icase eSF4A, inhibits the replication of diverse human viruses. Pafeamine A is effective at limiting virus replication at concentrations that do not affect cell viability (low nanomolar range) in primary human cells. The data demonstrates that pateamine A inhibits the replication of human cytomegalovirus (HCMV), herpessimpiex vims (H ^' SV), Ross River virus (RR ),

Chikungunya virus (CROC), respiratory syncytial vims (RSV), parainfluenza virus (PIV), dengue virus, and human rbinovirus 16 (RRV 16) in tissue culture. The financial and public health burdens imposed by these viruses suggest a wide applicability of patearaine A, and rel ated compounds that inhibit eIF A, to the treatment o f viral disease.

Additionally, patea ^' mine A has been shown to be tolerated in mice tor extended periods of time (e.g., greater than 30 days), demonstrating that pateamine A is tolerated throughout the therapeutic window for acute viral infections. uznetsov et. al. Mo! Cancer Ther. 2009 May;8(5): 1250-60. Epub 2009 May 5. This is especially important as many viruses do not currently .have effective antiviral therapeutics. In addition, the currently available antiviral therapeutics for many viruses can onl be administered for limited periods of time due to toxic side effects. j 000 I 74) in addition, the invention is not intended to be limited to the disclosed embodiments of the in v ention. I t should he understood that, the foregoing disclosure emphasizes cenain specific embodiments of the invention and that all modifications or alternatives equivalent thereto are within the spirit and scope of the invention as set forth in the appended claims.