Master’s Thesis Proposal

A Cell Based Assay for the Discovery of West Nile Virus Protease Inhibitors

Jason Machula

Wolkowicz Lab

April 2012

Aims:

The goal of this project is to develop an assay to monitor the activity of West Nile Virus (WNV) protease in HEK293T cells. We have previously developed an assay for HIV-1 protease in a Gal4 fusion context in non-adherent T-cells1. This assay exploits the autocatalytic properties of HIV-1 protease (PR), and Gal4, a transcription factor that is only active when the terminal DNA binding and Trans-activating domain (DBD, TAD) are linked1–3. This assay for WNV will provide a platform for the high-throughput screening for novel inhibitors of WNV protease utilizing flow cytometry and/or plate reader based technologies. In addition it will provide the proof of concept for the utility of the Gal4 protease fusion system, for a different viral type of protease, which, in contrast to HIV-1, requires a cofactor for full activity. Moreover, the assay will be developed in adherent cells rather than in T-cells. While the main goal of the assay is to drastically facilitate drug discovery it could be used to study the interaction of protease, cofactor, host and other viral factors required for protease activity.

Background:

WNV, a member of the Flaviviridae, is a (+) single stranded RNA virus with a genome of about 11Kb encased in an icosahedral capsid surrounded by an envelope. Viral entry occurs via endocytosis facilitated by receptor binding. The capsid is released into the cytoplasm following membrane fusion. The entire genome consists of one 3300 amino acid open reading frame and can serve as mRNA. Upon translation of the single open reading frame it is then processed by both viral and cellular proteases into structural and non-structural proteins (Figure 1)4–7. WNV infections are usually asymptomatic however infection can be febrile, or even neuroinvasive causing meningitis or encephalitis8. WNV infection has been found in a wide variety of mammals and even in some reptiles9. These hosts are considered dead-end hosts although birds are an amplifying host providing viral loads high enough to be transmitted by mosquitos, which is the vector responsible for human infection10,11. WNV was first identified in 1937 in Uganda and was first reported in the US in 199912,13. By 2004 it had spread throughout the continental US and has since been found in Alaska, Hawaii, and Canada14. It was originally found to be spread by only one type of mosquito however it is now spread by three different known species in the US alone15. Overwintering mosquitos have also been found that continue to carry viral loads through the winter16–

18. The wide variety of hosts from mammals, including rodents, humans, and killer whales, to birds, combined with the processive RNA-dependent RNA polymerase prone to errors provide an environment for the emergence of new strains of WNV7,19,20. The lack of effective vaccines or FDA approved inhibitors against WNV reiterates the need for new methods that facilitate the discovery of antivirals against WNV21. The assay described in this proposal will provide a platform for the high throughput screening of random peptide or compound libraries for the discovery of novel WNV protease inhibitors.

Figure 1. Genome, Polyprotein Precursor, and Mature Processed Proteins of WNV. (Murray KO, Mertens E, Desprès P. West Nile virus and its emergence in the United States of America. Veterinary Research. 2010;41(6):67.)

The assay is based on the inducible expression of the viral protease as a fusion within the DBD and TAD of Gal4, and the activation of the reporter green fluorescent protein (GFP) driven by an Upstream Activation Sequence (UAS) which serves as the promotor for Gal41. Gal4 is a well characterized cis-acting transcription factor from yeast with two required domains, the DBD binds to the UAS and the TAD activates expression of downstream genes. The two required domains can be spatially separated however they must be present in a single polypeptide to remain active2,3. The engineered fusion, where the viral protease is inserted between the Gal4 domains, will be cleaved by protease if active thus preventing the induction of GFP driven by the UAS. Induction of GFP, via the UAS thus occurs only when protease is inhibited, acting as a biosensor for protease activity1. To avoid possible cytotoxic effects of protease expression, the Gal4/protease fusion will be expressed in an off/on inducible manner regulated by a Tetracycline Responsive Element (TRE). When Tetracycline (Tet) or Doxycycline (Dox) is present it will bind to and change conformation of the reverse tetracycline trans-activator (rtTA) allowing it to bind to TRE and activate transcription22. Without Tet or Dox, the Gal4/protease fusion will not be expressed (Figure 2).

Figure 2: Assay Overview. A. Wild type Gal4 as control, no Dox. Without Dox, rtTA cannot bind to the Tet-responsive element (TRE) thus Gal4 is not expressed and, consequently, neither is GFP. B. In the presence of Dox (blue diamond), Gal4 is expressed and binds the Upstream Activating Sequence (5xUAS), activating GFP expression. C. The PR/Gal4 fusion-based system. In the presence of Dox PR/Gal4 is expressed; however, its catalytic activity results in the separation of the Gal4 domains and thus, no GFP expression. D. Same scenario as in C but in the presence of Protease inhibitors (yellow circles). PR/Gal4 fusion remains intact, resulting in the induction of GFP expression.

WNV like HIV rely on the processing of their proteome however the WNV protease, like other flaviviral proteases, requires a cofactor for activity 1,23,24. In order for members of the Flaviviridae to become infective the cofactor NS2B is autocatalytically cleaved from the protease NS3 (Figure 1)25–27. The boundary between NS2B and NS3 and the NS2B/NS3 protease are thus an attractive target for the discovery of protease inhibitors or competitors because this site is only cleaved by viral, not host, proteases, as is the case with other NS proteins21,26,28,11. The assay, previously developed in T-cells for HIV, is being adapted to adherent mammalian cells, specifically Human Embryonic Kidney (HEK) 293T cells, providing a better model for flaviviral infections.

Experimental Design:

Many studies have been performed regarding the analysis of the structure and functions of the NS3 protease and its cofactor (NS2B). It has been shown that only the central hydrophilic domain of cofactor NS2B is required for proteolytic activity of NS3 and that the N-terminal NS3 protease domain does not require the C-terminal NS3 helicase domain for autolytic activity23,25,28,11,29–32. Shiryaev et al. have built a construct with the required portion of NS2B linked by a GGGGSGGGG linker to the protease portion of NS333. This construct was generously supplied as the source of WNV protease. The construct provided contains a K48A mutated cleavage site rendering the site un-cleavable (Figure 3). Many other flaviviral protease studies use the same constructs, as they allow the study of the activity of the entire NS2B/NS3 protein avoiding cleavage at the NS2B/NS3 boundary.

Figure 3. West Nile Virus Proteinase Based Constructs The autolytic site deficient construct with the K48A mutation without the helicase domain was generously provided by Sergey Shiryaev of the Burnham Institute. The wild type construct was made by reverting the mutation with overlapping PCR. The catalytically inert mutant will be made in a similar manner (See body for details).

(Shiryaev SA, Kozlov IA, Ratnikov BI, et al. Cleavage preference distinguishes the two-component NS2B–NS3 serine proteinases of Dengue and West Nile viruses. Biochemical Journal. 2007;401(3):743)

Gal4/NS2B/3 based constructs were introduced into pcDNA for transient expression in HEK293T and Huh cells. In this plasmid, expression is driven by a minimal Cytomegalovirus (mCMV) promoter thus allowing for expression in mammalian cells, at least transiently. Two main pcDNA constructs were engineered: One with the provided protease bearing the mutated cleavage site and a second one with the wild type sequence obtained by reverting the A48K mutation present in theNS2B/3 cleavage site by overlapping PCR. These protease sequences were then ligated into a pcDNA-mCMV-Gal4 vector cut with EcoRI and dephosphorylated using calf intestinal phosphatase (Figure 4).

Figure 4. The WNV Genome Consisting of a Single Polyprotein Precursor and Derived Constructs. The protease NS3 and required cofactor NS2B in red within the Gal4 transcription factor. A. Also depicted the Gal4/NS2B/NS3 fusion with wild-type cleavage site (green) B. Construct with mutated, non-cleavable, cleavage site (crimson). C. Gal4 without a protease will be used as a positive control.

Lentiviral packaging vectors (pH) were used to make HIV based viral particles for the stable expression of assay components in HEK293T cells. Virus was produced in HEK293T cells with the envelope glycoprotein from the Vesicuar Stomatitus Virus and the Gag, Pol, and Vpr proteins of HIV. Four different viral particles were produced in this manner. Two of them include basic assay components for development of the HEK293T cell lines; the reporter vector pH-5xUAS-GFP and the vector providing rtTA, pBMN-rtTA-i-Lyt2. The latter construct contains an internal ribosome entry site (IRES) allowing expression of two genes from the same mRNA and Lyt2, a receptor from mouse CD8a that acts as a selectable marker. Two viruses were also packaged for Gal4/NS2B/3 constructs (wild type and mutant, referred to PR and PRm respectively, in Figure 6. For that purpose the Gal4 fusion constructs were cut out of the pcDNA Gal4 construct with NotI and XbaI then ligated into pcDNA-Hygro, cut out with BamHI and XbaI and finally ligated into the pH-7xTRE vector.

Figure 5. Retroviral Constructs for Development of HEK293T Cell Lines The retroviral 5xUAS-GFP and rtTA constructs for the inducible system. The original rtTA construct in the HIV assay contained an IRES-mCherry, which is replaced by the Lyt-2 receptor (mouse CD8a) to free the channel for mCherry for future studies.

Figure 6: Retroviral Gal4/Protease Fusion Constructs. Schematics of the retroviral constructs containing the wild type Gal4/PR (pH-TRE-PR), and mutated Gal4/PRm (pH-TRE-PRm) and Gal4 controls (pH-TRE-Gal4) used in the HIV-1 assay. Each construct contains 7xTRE, mCMV promoter, and 5’ and 3’ Long Terminal Repeats (LTR).

HEK293T cells have been successfully infected and express both 5xUAS GFP and rtTA. The next step will be to infect these cells with pH-TRE-Gal4 fusion constructs. Selection of the cells successfully infected should be straightforward with the mutated construct as they have all of the components required to express GFP in the presence of Dox. This is not the case with the wild type fusion-expressing cells as they will not be distinguishable by GFP expression. In order to select these cells an inhibitor must be used. Several protease inhibitors have been shown to be successful in-vitro including, aprotinin, palmatine, and various peptide inhibitors4,24,30,34,35. If inhibitors are proven to not successfully inhibit protease, additional Gal4 fusion constructs will be needed.

In order to determine whether the protease is actually being expressed and the expected GFP pattern expression is not due to non-specific factors, western blots will be performed utilizing monoclonal antibodies against WNV protease and/or Gal4. Western blotting will confirm expression of the protease whether there is GFP expression or not. Uncleaved Gal4 fusion should be around 56 KDa and when cleaved the DBD and cofactor 2B (approximately 23KDa) will be separated from the NS3 protease and TAD which together is approximately 33 KDa.

Once a stable cell population has been raised expressing all of the necessary elements of the assay, clonal selection will begin. The cells expressing the non-cleavable control fusion can be directly sorted based on GFP expression. The cells expressing the wild type fusion will involve more rounds of sorting. These cells will be sorted for no GFP expression with Dox but no inhibitor eliminating background and then sorted for GFP expression when Dox and an inhibitor are present. Selection will be performed to obtain the best clones, lowest background and highest GFP expression, in the assay. After a yet to be determined number of rounds of sorting, cells will be individually plated in a 96-well plate and amplified to obtain clonal cell lines. These cell lines will then be tested in order to find a cell line that will show robust GFP expression when Dox and inhibitor are present and little to no background when Dox is present without inhibitor.

After generating stable cell lines expressing all of the components of the assay, optimization of enzyme kinetics can be performed. This will involve titration experiments with Dox to determine the optimal conditions for the maximum induction of Gal4.

Preliminary Data:

Gal4 WNV protease fusions, wild type and mutated, were constructed in pcDNA vectors with a CMV promoter and confirmed by sequencing. Transfection experiments were performed with these constructs in both naïve HEK293T cells and Huh 7.5.1 cells, a human hepatocytic cell line proving a second model, and in both Huh 7.5.1 and HEK293T cells that expressed 5xUAS GFP. In all these experiments, Gal4 alone positive control inducted GFP expression in approximately 10-15% of cells. Similar level of GFP induction was seen with transfection by Gal4/NS2B/3m, which serves as a positive control for Gal4 fusion protein (and negative for protease activity). Little to no GFP expression was observed with naïve cells as with cells transfected with Gal4/NS2B/3wt, as expected (data not shown).

A population of HEK293T cells engineered to stably express rtTA and 5xUAS GFP has already been established. Preliminary transient expression experiments with these cells have shown that the Gal4/NS2B/3 fusion with a

mutated cleavage site (Gal4/PRm) activate GFP expression, proving that the fusion is indeed a single polyprotein and that the approximately 1500bp-long protease sequence does not disrupt the ability of the fusion to bind to 5xUAS and induce the downstream gene (GFP).

Transfection of HEK293T cells that express both 5xUAS GFP and rtTA also yielded similar results when transfected with pH-7xTRE constructs in the presence of Dox. The Gal4 control without protease resulted in 1.9% GFP expression in the absence of Dox and 10.3% with Dox. The Gal4/NS2B/3m construct resulted in 9% GFP in the presence of Dox, while the Gal4/NS2B/3wt construct showed no GFP expression (Figure 7). While the percentage of positive green cells is not very high, it is very close in both control and mutated fusion, demonstrating the expected trend (10.3% and 9% respectively). The cell samples analyzed by flow cytometry were transfected in these preliminary experiments for the purpose of determining the best vectors of choice for virus production and Gal4 fusion constructs. In addition, the non-specific effect of Dox will be determined in future experiments.

Figure 7. Gal4/NS2B/3 Fusion Transfection in HEK293T Cells Expressing 5xUAS GFP and rtTA. The Gal4 fusion constructs with a mutated cleavage site (Gal4/WNVm) showed similar GFP expression as the Gal4 with Dox positive control. The wild type Gal4 fusion (Gal4/WNVwt) showed almost no GFP expression indicating cleavage of the protease within the Gal4 domains. The Gal4 control without Dox showed 1.9% GFP as background.

Timeline:

The next step will be aimed at obtaining stably-expressing cell lines, where expression of protease will be verified based on GFP expression in the presence of Dox. In order to corroborate expression of the wild type construct western blotting will be performed. In this case, as wild type protease does not lead to GFP expression unless inhibited, selection will be performed in the presence of a putative protease inhibitor. Upon selection of two cell populations stably expressing rtTA and harboring the 5xUAS GFP construct, one expressing the mutated version and the second, the wild type Gal4/NS2B/3 fusion, clonal selection will be utilized to obtain a cell line with optimal sensitivity. We expect to obtain a cell population expressing rtTA, 5xUAS GFP, and Gal4/NS2B/NS3m by the end of Spring 2012. Next, we intend to obtain the Gal4/NS2B/NS3wt. Upon successful inhibition of Gal4/NS2B/3wt, a stable population of cells can be selected and clonal selection can begin. The clonal cell lines should obtained by the summer of 2012, allowing to optimize the assay Fall 2012. The assay will ultimately be used to screen libraries in search for novel inhibitors as well as for the study of protease and cofactor activity.

1. Hilton BJ, Wolkowicz R. An Assay to Monitor HIV-1 Protease Activity for the Identification of Novel Inhibitors in T-Cells. PLoS ONE. 2010;5(6):e10940.

2. Johnston SA, Zavortink MJ, Debouck C, Hopper JE. Functional domains of the yeast regulatory protein GAL4. Proc Natl Acad Sci U S A. 1986;83(17):6553–6557.

3. Murray MG, Hung W, Sadowski I, Das Mahapatra B. Inactivation of a yeast transactivator by the fused HIV-1 proteinase: a simple assay for inhibitors of the viral enzyme activity. Gene. 1993;134(1):123–128.

4. Robin G, Chappell K, Stoermer MJ, et al. Structure of West Nile virus NS3 protease: ligand stabilization of the catalytic conformation. J. Mol. Biol. 2009;385(5):1568–1577.

5. Ebel GD, Fitzpatrick KA, Lim P-Y, et al. Nonconsensus West Nile Virus Genomes Arising During Mosquito Infection Suppress Pathogenesis and Modulate Virus Fitness In Vivo. J. Virol. 2011;85(23):12605–12613.

6. Stiasny K, Fritz R, Pangerl K, Heinz F. Molecular mechanisms of flavivirus membrane fusion. Amino Acids. 2011;41(5):1159–1163.

7. Murray KO, Mertens E, Desprès P. West Nile virus and its emergence in the United States of America. Veterinary Research. 2010;41(6):67.

8. Sejvar JJ, Haddad MB, Tierney BC, et al. Neurologic Manifestations and Outcome of West Nile Virus Infection. JAMA. 2003;290(4):511–515.

9. Farfán-Ale JA, Blitvich BJ, Marlenee NL, et al. Antibodies to West Nile virus in asymptomatic mammals, birds, and reptiles in the Yucatan Peninsula of Mexico. Am. J. Trop. Med. Hyg. 2006;74(5):908–914.

10. Campbell GL, Marfin AA, Lanciotti RS, Gubler DJ. West Nile virus. The Lancet Infectious Diseases. 2002;2(9):519–529.

11. Chappell KJ, Stoermer MJ, Fairlie DP, Young PR. Insights to substrate binding and processing by West Nile Virus NS3 protease through combined modeling, protease mutagenesis, and kinetic studies. J. Biol. Chem. 2006;281(50):38448–38458.

12. Pesko KN, Ebel GD. West Nile virus population genetics and evolution. Infect. Genet. Evol. 2012;12(2):181–190.

13. Hayes CG. West Nile virus: Uganda, 1937, to New York City, 1999. Ann. N. Y. Acad. Sci. 2001;951:25–37.

14. Hayes EB, Komar N, Nasci RS, et al. Epidemiology and transmission dynamics of West Nile virus disease. Emerging Infect. Dis. 2005;11(8):1167–1173.

15. Molaei G, Andreadis TG, Armstrong PM, Anderson JF, Vossbrinck CR. Host Feeding Patterns of Culex Mosquitoes and West Nile Virus Transmission, Northeastern United States. Emerg Infect Dis. 2006;12(3):468–474.

16. Armstrong PM, Vossbrinck CR, Andreadis TG, et al. Molecular evolution of West Nile virus in a northern temperate region: Connecticut, USA 1999–2008. Virology. 2011;417(1):203–210.

17. Ciota AT, Drummond CL, Drobnack J, et al. Emergence of Culex pipiens from overwintering hibernacula. J. Am. Mosq. Control Assoc. 2011;27(1):21–29.

18. Bugbee LM, Forte LR. The discovery of West Nile virus in overwintering Culex pipiens (Diptera: Culicidae) mosquitoes in Lehigh County, Pennsylvania. J. Am. Mosq. Control Assoc. 2004;20(3):326–327.

19. Steinman A, Banet-Noach C, Tal S, et al. West Nile Virus Infection in Crocodiles. Emerg Infect Dis. 2003;9(7):887–889.

20. Klenk K, Snow J, Morgan K, et al. Alligators as West Nile virus amplifiers. Emerging Infect. Dis. 2004;10(12):2150–2155.

21. Shiryaev SA, Cheltsov AV, Gawlik K, Ratnikov BI, Strongin AY. Virtual Ligand Screening of the National Cancer Institute (NCI) Compound Library Leads to the Allosteric Inhibitory Scaffolds of the West Nile Virus NS3 Proteinase. Assay Drug Dev Technol. 2011;9(1):69–78.

22. Weyler M, Morschhäuser J. Tetracycline-inducible gene expression in Candida albicans. Methods Mol. Biol. 2012;845:201–210.

23. Radichev I, Shiryaev SA, Aleshin AE, et al. Structure-Based Mutagenesis Identifies Important Novel Determinants of the NS2B Cofactor of the West Nile Virus Two-Component NS2B–NS3 Proteinase. J Gen Virol. 2008;89(3):636–641.

24. Chappell KJ, Stoermer MJ, Fairlie DP, Young PR. Mutagenesis of the West Nile Virus NS2B Cofactor Domain Reveals Two Regions Essential for Protease Activity. J Gen Virol. 2008;89(4):1010–1014.

25. Shiryaev SA, Strongin AY. Structural and functional parameters of the flaviviral protease: a promising antiviral drug target. Future Virol. 2010;5(5):593–606.

26. Nall TA, Chappell KJ, Stoermer MJ, et al. Enzymatic Characterization and Homology Model of a Catalytically Active Recombinant West Nile Virus NS3 Protease. J. Biol. Chem. 2004;279(47):48535–48542.

27. Aleshin AE, Shiryaev SA, Strongin AY, Liddington RC. Structural evidence for regulation and specificity of flaviviral proteases and evolution of the Flaviviridae fold. Protein Sci. 2007;16(5):795–806.

28. Chappell KJ, Nall TA, Stoermer MJ, et al. Site-directed mutagenesis and kinetic studies of the West Nile Virus NS3 protease identify key enzyme-substrate interactions. J. Biol. Chem. 2005;280(4):2896–2903.

29. Shiryaev SA, Aleshin AE, Ratnikov BI, et al. Expression and purification of a two-component flaviviral proteinase resistant to autocleavage at the NS2B–NS3 junction region. Protein Expression and Purification. 2007;52(2):334–339.

30. Johnston PA, Phillips J, Shun TY, et al. HTS Identifies Novel and Specific Uncompetitive Inhibitors of the Two-Component NS2B-NS3 Proteinase of West Nile Virus. ASSAY and Drug Development Technologies. 2007;5(6):737–750.

31. Erbel P, Schiering N, D’Arcy A, et al. Structural basis for the activation of flaviviral NS3 proteases from dengue and West Nile virus. Nat. Struct. Mol. Biol. 2006;13(4):372–373.

32. Chernov AV, Shiryaev SA, Aleshin AE, et al. The Two-Component NS2B-NS3 Proteinase Represses DNA Unwinding Activity of the West Nile Virus NS3 Helicase. J. Biol. Chem. 2008;283(25):17270–17278.

33. Shiryaev SA, Kozlov IA, Ratnikov BI, et al. Cleavage preference distinguishes the two-component NS2B–NS3 serine proteinases of Dengue and West Nile viruses. Biochemical Journal. 2007;401(3):743.

34. Shiryaev SA, Ratnikov BI, Chekanov AV, et al. Cleavage targets and the D-arginine-based inhibitors of the West Nile virus NS3 processing proteinase. Biochemical Journal. 2006;393(2):503.

35. Stoermer MJ, Chappell KJ, Liebscher S, et al. Potent Cationic Inhibitors of West Nile Virus NS2B/NS3 Protease With Serum Stability, Cell Permeability and Antiviral Activity. J. Med. Chem. 2008;51(18):5714–5721.