Characterization of TGV-dependent double band

FIGURE 1. HCV NS5B polymerase

Crystal structure of HCV NS5B polymerase. The subdomains of this canonical polymerase structure are shown: palm (green), fingers (red) and thumb (blue). Cys 366, located in the protein active site, is indicated by the arrow.

Mass spectrometry of intact NS5B. (A) Deconvoluted spectrum from Mass Spectrometry of control (non-treated), (B) TGV treated, (C) treated with compound containing no halide, (D) treated with compound containing Chlorobenzene, (E) treated with compound containing Fluorobenzene, and (F) treated with Rigel R803. In each spectrum, the 64 kD peak corresponds to NS5B protein. For each compound-treated sample, an additional peak is observed. The spacing between the 2 peaks is equivalent in mass to the covalent addition of compound plus glutathione. Please see Fig. 3 for proposed chemical mechanism.

FIGURE 4. Proposed Chemical MOA for TGV

FIGURE 5. Data analysis workflow

FIGURE 6. Novel peptide identification

(A) Ratio plot from Elucidator. (B) Mass spectrum of peptide unique to TGV-treated sample. (C) Isotope cluster from peptide unique to TGV-treated sample. (D) Bar graph showing relative intensities of unique peptide in drug treated vs. control samples.

Identification of TGV binding site

FIGURE 8. Confirmation of addition of TGV at NS5B Cys 366

Point mutations in Cys 366 were engineered by site-directed mutagenesis to the NS5B region of HCV replicon and subcloned into pFasBacMam. Recombinant virus was generated using the Bac-to-Bac system (Life Technologies, Carlsbad, CA). Protein was immunoprecipitated and subsequently analyzed +/- TGV treatment. No doublet was observed in any mutated NS5B protein. These data support the binding of TGV at NS5B Cys 366.

Gilead Sciences, Inc. 333 Lakeside Drive

Foster City, CA 94404 Tel: (650) 522-1890 Fax: (650) 522-5166

kathy.brendza@gilead.com

Mass Spectrometry as a Tool to Uncover a Novel Mechanism of Action for the HCV

NS5B Polymerase Non-nucleoside Inhibitor Tegobuvir Katherine M. Brendza1, Cheryl F. Lichti2,5, Randall W. Vivian1, Christy M. Hebner1, Petra Erdmann-Gilmore2, Maisoun Sulfab1, James

Tayler1, James Trenkle1, Kyla Bjornson1, John Link1, Uli Schmitz1, Bin Han1, Steven Bondy1, Xiaohong Liu1, Magdeleine Hung1, Wanchi Fung1, Todd Appleby1, Swami Swaminathan1, Johan Neyts3, Weidong Zhong1,4, Roman Sakowicz1, R. Reid Townsend2

1Gilead Sciences, Inc., 333 Lakeside Drive., Foster City, CA 94404, USA, 2Proteomics Center, Washington University School of Medicine, 660 Euclid, Ave., St. Louis, MO 63110, 3Rega Institute, KU Leuven, Belgium, 4present address: Novartis, 4560 Horton St., Emeryville, CA 94608 , 5present address: UTMB, 301 University Blvd., Galveston, TX 77555.

60th Annual ASMS May 20-24, 2012 Vancouver, BC Canada

Overview

Methods

Introduction

Sample Preparation Purification of NS5B for Mass Spectrometry Analysis. Cell pellets were resuspended and lysed in lysis buffer (50 mM Tris pH 7.5, 350 mM sodium chloride, 2 mM PMSF, 5 mM DTT plus Complete Mini protease inhibitor tablet). Following lysis, debris was pelleted by centrifugation at 14,000 RPM for 5 minutes at 4 °C. Supernatant was incubated overnight with protein A Dynabeads (Life Sciences, Carlsbad CA) previously conjugated to anti-NS5B antibody (Axxora ; Clone 12B7). The next day, beads were washed three times with lysis buffer. Captured protein was eluted in 100 mM Glycine (pH 2.4)/4M Urea and immediately analyzed by mass spectrometry. To prepare peptides, the proteins were concentrated by precipitation, reduced (1 mM TCEP), and alkylated (20 mM iodoacetamide). The reduced and alkylated proteins were digested in 8M urea with 1 µg of endoproteinase Lys-C (Roche, Basel, Switzerland) and incubated overnight at 37 °C. The digests were then diluted to 2 M for digestion with 5 µg of trypsin overnight at 37 °C. Peptides were acidified with formic acid and extracted sequentially with NuTip wedge tips (200 µL; Glygen, Columbia, MS) containing C4 and porous graphite carbon stationary phases, respectively. Peptides were eluted into autosampler vials with 60% acetonitrile in 1% formic acid. The remaining sample was dried, dissolved in 35 µL of aqueous acetonitrile/formic acid (1%/1%), and analyzed by high resolution nano-LC-MS. LCMS To investigate the mechanism of action for tegobuvir, the intact molecular mass of NS5B from a tegobuvir-treated cell-based heterologous expression system was determined by mass spectrometry. Immunoprecipitated NS5B proteins were analyzed on an Agilent 6210 Time of Flight Mass Spectrometer with an Agilent 1200 Rapid Resolution HPLC. To identify the tegobuvir modification site on the NS5B protein, immunoprecipitated samples were reduced, alkylated, and digested with Lys-C and trypsin. Resulting samples were analyzed on an LTQ-Orbitrap with an Eksigent nanoLC 1D Plus to acquire high-resolution tandem spectra (MS2). Data Analysis Intact protein data was processed via Agilent Masshunter B.03 Qualitative Analysis, with the Bioconfirm upgrade, allowing for protein deconvolution. For modification site ID, the MS2 spectra were analyzed both by searching a customized protein database that contained the sequence of NS5B (DDA) and expert manual interpretation (high resolution targeted analysis). The exact masses of the fragmentation ions were calculated using the MS-Product utility within Protein Prospector (http://prospector.ucsf.edu). For database searches, the LC-MS files were processed using MASCOT Distiller (Matrix Science, version 2.3.0.0) with the settings previously described (7). All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK; version Mascot). Mascot was set up to search a custom, in-house database (215 entries) assuming the digestion enzyme trypsin with 4 missed cleavages. Mascot was searched with a fragment ion mass tolerance of 0.800 Da and a parent ion tolerance of 50 PPM. Iodoacetamide derivative of cysteine was specified in Mascot as a fixed modification. Oxidation of methionine was specified in Mascot as a variable modification. Theoretical values for tegobuvir adducts were added to Mascot and were specified as variable modifications.

Hepatitis C Virus (HCV) is a major cause of morbidity and mortality with an estimated 170 million people infected worldwide and 3-4 million new infections acquired annually (1). HCV, a member of the Flaviviridae family, is a positive-strand RNA virus with seven major genotypes each divided into a series of subtypes. Additionally, each subtype exists as a quasispecies population within an infected individual (2). Current treatment, a combination of pegylated interferon (PEG) and ribavirin (RBV), has significant side-effects and is of limited efficacy against HCV genotype 1, the most prevalent in the United States and Europe (3). Viral polymerases are attractive targets for drug discovery, yielding approved drugs for HIV, HBV, herpes simplex virus and cytomegalovirus. HCV NS5B polymerase is an RNA-dependent RNA polymerase containing the canonical palm, finger, and thumb subdomains (Fig 1; 4,5,6). Tegobuvir (TGV) is a novel non-nucleoside inhibitor of HCV RNA replication and has demonstrated potent antiviral activity in genotype 1 HCV patients. However, the detailed mechanism of action for TGV has not been clearly defined despite the fact that resistance mutations to TGV have been mapped to the NS5B polymerase region. Here we demonstrate that TGV exerts anti-HCV activity through a unique mechanism, directly interacting with the NS5B protein. We further identify the site of interaction through high resolution mass spectrometry.

Identification of TGV binding site

Conclusions

References

C366A C366G C366S WT

9190 - + - + - + +

C366A C366G C366S WT

9190 - + - + - + +

FIGURE 3. Mass Spectrometry of Intact NS5B protein.

Counts vs. Deconvoluted Mass (amu)

(A)

(B)

(C)

(D)

(E)

(F)

FIGURE 7. Sequence of Peptide with Chemical adduct.

G*

G*

G*

(A)

(C)

(B)

(A)

(C) (B) FIGURE 2. TGV induces double band formation, visualized by Western analysis

(A) NS5B, NS5A, and NS3 Western blots of lysates from 1b (Con-1) replicon cells treated with 50 nM TGV for various periods of time. Note NS5B doublet formation at 8 hours, not observed with other HCV proteins. (B) NS5B Western blot of lysates from 1b replicon cells treated for 24 hours with known HCV inhibitors at 50x EC50 concentrations. Note NS5B doublet formation specific to TGV treated sample, and not observed with other known NS5B inhibitors.

NS5B

NS5A

Hours Post-treatment w/ 50nM TGV

0 2 4 8 12 24

NS3

0 8 12 24

A) B)

-TGV

+TGV

-TG

V

-TG

V

+T

GV

+T

GV

+T

GV

-TG

V

(1) Wasley, A., and Alter, M. J. 2000. Epidemiology of hepatitis C: geographic differences and temporal trends. Semin Liver Dis 20:1-16. (2) Simmonds, P. 2004. Genetic Diversity and evolution of hepatitis C virus: 15 years on. J Gen Virol 85:3713-88. (3) Manns, M. P., McHutchison, J. G., Gordon, S. C., Rustgi, V. K., Shiffman, M., Reindollar, R., Goodman, Z. D., Koury, K., Ling, M.-H. and Albrecht, J. K. 2001. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 358:958-965. (4) Lesburg, C. A., Cables, M. B., Ferrari, E., Hong, Z., Mannarino, A. F., and Weber, P. C. 1999. Crystal structure of the RNA-dependent RNA polymerase from hepatitis C virus revealse a fully encircled active site. Nat Struct Biol 6:937-43. (5) Behrens, S. E., Tomei, L., and DeFrancesco, R. 1996. Identification and properties of the RNA-dependent RNA polymerase of hepatitis C virus. EMBO J 15:12-22. (6) Yamashita, M. P., Kaneko, S., Shirota, Y., Qin, W., Nomura, T., Kobayashi, K., and Murakami, S. 1998. RNA-dependent RNA polymerase activity of the soluble recombinant hepatitis C virus NS5B protein truncated at the C-terminal region. J Biol Chem 273:15479-86. (7) Nittis T, Guittat L, LeDuc RD, Dao B, Duxin JP, Rohrs H, Townsend RR, Stewart SA. “Revealing novel telomere proteins using in vivo cross-linking, tandem affinity purification, and label-free quantitative LC-FTICR-MS.” Mol Cell Proteomics 9 (6), 1144-56 (2010). (8) Strohalm M, Kavan D, Novák P, Volný M, Havlíček V. mMass 3: A Cross-Platform Software Environment for Precise Analysis of Mass Spectrometric Data. Anal Chem 82 (11), 4648-4651 (2010). (9) Strohalm M, Hassman M, Košata B, Kodíček M. mMass Data Miner: an Open Source Alternative for Mass Spectrometric Data Analysis.Rapid Commun Mass Spec 22 (6), 905-908 (2008).

Fingers

Palm

Thumb

Cys 366

Here we show the application of High Resolution Mass Spectrometry in elucidation of the chemical mechanism of action of the HCV NS5B inhibitor Tegobuvir (TGV). Through a combined approach utilizing investigation of intact proteins and detailed proteomics, we were able to determine TGV is covalently bound to NS5B protein, and the site of binding is Cys 366.

• Through SDS-PAGE, an aberration of the NS5B protein was detected. This aberration was specific to the HCV non-structural protein 5B, and only observed upon treatment with TGV. • Through Mass Spectrometry of intact NS5B protein, we have shown that the HCV NS5B polymerase inhibitor TGV binds to the protein covalently • Using tool compounds to examine SAR, we have confirmed our proposed chemical MOA for TGV. Consistent with our observations by LCMS, the TGV is oxidized via Cyp-mediated reaction and the halogen is eliminated. • Through high resolution MS analysis, we have determined that the site of TGV-binding is Cys 366. •Site-directed mutagenesis confirmed Cys 366 as TGV point of attachment to NS5B

(D)

High resolution MS/MS spectrum (A) and supporting ions (B, C) for 346YSAPPGDPPKPEYDLELITSC(TGV)SSNVSVAHDASGK379 (theoretical [M+5H]5+ = 809.758, observed [M+5H]5+ = 809.758). The spectrum is dominated by fragment ions that arise from fragmentation on the N-terminal side of proline. The spectrum is contaminated with fragment ions from a co-eluting, doubly charged peptide of similar m/z, as seen in the MS1 spectrum (inset).

Three pairs of samples (+/–TGV) were analyzed by nanoLC-MS (1). Resulting data files were imported into Rosetta Elucidator™ software for retention time and m/z alignment (2). A scatter plot of Log10 intensities (+/–TGV) was generated within Elucidator. Isotope groups > 20 fold higher in +TGV samples were examined manually (3). A theoretical digest of NS5B protein, modified with knowledge of drug protein target sites, was performed using mMass (version 3.2.0). The m/z values were matched to the list of detected isotope groups using a 20 ppm mass tolerance, and resulting isotope groups examined manually (4). Isotope groups markedly increased in drug-treated sample having theoretical m/z values matching ion signals from the total aligned list of isotope groups were used for directed mass spectrometry analysis (DMSA) to obtain MS/MS spectra of the potential TGV-modified peptide(s) (5).

TGV is likely to be oxidized to form epoxide intermediate (2) which can equilibrate with epoxide-opened form (3). The latter can eliminate the fluoride atom to yield unsaturated ketone (4). The imidazopyridine ring system adjacent to the 2-fluorophenyl uniquely allows for a highly stabilized resonance structure conjugated through a double bond between the rings. This creates three different possibilities for a nucleophilic attack by GSH or another reactive cysteine residue (labeled A, B and C on structure 4). In the case of GSH attacking positions A or B, the ensuing adducts 5a and 5b are capable of tautomerizing to the aromatic, disubstituted phenol analogs 6a and 6b. Both 6a and 6b exhibit molecular weights of 820 daltons and correspond to the putative TGV metabolites mentioned above. Pathway C, however, involves an attack on the ipso carbon which leads to adduct (7) which cannot tautomerize and remains an activated ketone prone to a second nucleophilic attack by a reactive cysteine residue yielding double adducts (8a) and (8b).