Structural and antiviral studies of the human norovirus GII.4 protease

Subscriber access provided by United Arab Emirates University | Libraries Deanship

Article

Structural and antiviral studies of the human norovirus GII.4 protease Kendall M. Muzzarelli, Benjamin D. Kuiper, Nicholas Spellmon, Joseph S. Brunzelle, Justin Hackett, Franck Amblard, Shaoman Zhou, Peng Liu, Iulia A Kovari, Zhe Yang, Raymond Felix Schinazi, and Ladislau C. Kovari Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b01063 • Publication Date (Web): 03 Jan 2019 Downloaded from http://pubs.acs.org on January 4, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease Kendall M. Muzzarelli1, Benjamin Kuiper1, Nicholas Spellmon1, Joseph Brunzelle2, Justin Hackett1, Franck Amblard3, Shaoman Zhou3, Peng Liu3, Iulia A. Kovari1, Zhe Yang1, Raymond F. Schinazi3, Ladislau C. Kovari1* 1

Wayne State University School of Medicine, Department of Biochemistry, Microbiology and Immunology, Detroit, MI 48201, USA;2Synchrotron Research Center, Life Science Collaborative Access Team, Northwestern University, Argonne, Illinois, USA ; 3Center for AIDS Research, Laboratory of Biochemical Pharmacology, Department of Pediatrics, Emory University School of Medicine, Atlanta, GA 30322, USA *To whom correspondence should be addressed: Ladislau Kovari: Department of Biochemistry, Microbiology & Immunology, Wayne State University School of Medicine, Detroit MI, 48201; [email protected]

Abstract Norovirus is the leading cause of acute gastroenteritis worldwide with a yearly reported 700 million cases driving a $60 billion global socioeconomic burden. With no FDA approved therapeutics and the chance for severe chronic infection and life-threatening complications, researchers have identified the protease as a potential target. However, drug development has focused on the norovirus GI.1 strain despite its accounting for less than 5% of all outbreaks. Our lab aims to change focus for norovirus drug design from GI.1 to the highly infective GII.4; responsible for more than 50% of all outbreaks worldwide. With the first published crystal structure of the norovirus GII.4 protease, we have identified several significant differences in the structure and active site that have hindered development of a potent inhibitor targeting the norovirus GII.4 protease. With these new insights, we have begun designing compounds that demonstrate increased inhibition of the clinically most relevant norovirus GII.4 strain. Introduction Chronic gastroenteritis propagated by norovirus exposure leaves immunocompromised patients vulnerable to debilitating and life-threatening complications due to the lack of FDA approved antiviral drugs and vaccines. A recent cohort study of 8.5 years was conducted at Texas Children’s Hospital which showed a 60-87% decrease in rotavirus infection post vaccine licensure (1). This study helps confirm that noroviruses are quickly replacing rotaviruses as the leading cause of acute gastroenteritis in both immunocompetent and immunocompromised patients owing to the recent development and universal distribution of a rotavirus vaccine (2,3). According to a report from the Centers for Disease Control and Prevention (CDC), it is now estimated that noroviruses are responsible for 685 million cases each year with an estimated $60 billion medical and socioeconomic cost (4). Specifically, the GII.4 norovirus genotype accounts for more than 50% of all outbreaks worldwide (3,5,6). This overwhelming global burden drives the need for identification and characterization of potential norovirus targets to exploit for drug design and development (7). Norovirus is a genus of positive sense, single-stranded RNA viruses belonging to the Calciviridae family. They consist of 6 genogroups (GI-VI), from which the prototypic GI.1 genotype is the most well characterized while the predominant GII.4 genotype accounts for the majority of outbreaks in humans (8). The ~7.7 kb genome contains three open reading frames (ORF); ORF2 and ORF3 encode structural proteins and ORF1 encodes a ~200 kDa nonstructural polyprotein essential for viral replication. The translated polyprotein encoded by ORF1 undergoes self-cleavage into mature non-structural proteins by a 3C-like cysteine protease (9-13). This norovirus 3C-like cysteine protease has been identified as belonging to the chymotrypsin-like protease superfamily which displays a chymotrypsin-like serine protease fold as well as

ACS Paragon Plus Environment

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Structural and antiviral studies of the human norovirus GII.4 protease a nucleophilic cysteine residue in the catalytic triad (13-15). The norovirus protease is pertinent for production of mature viral proteins which are necessary for viral replication and therefore makes the proteolytic mechanism an appealing target for antiviral therapeutics (16). Characterization of the norovirus 3C-like protease is currently limited to the GI.1 Norwalk virus protease while the more relevant GII.4 Minerva virus protease structure remains unsolved (17). The first GI.1 structure was solved at a 2.8 Å resolution (PDB 1WQS) while the highest resolution structure is 1.5 Å (9,17-19). However, a crystal structure of the GII.4 protease structure was, until this manuscript, unavailable resulting only in homology model predictions for the GII.4 protease. Additionally, our current GII.4 homology model is based on an alignment of the Norwalk virus GI.1 protease (3UR6.pdb) and GII.4 protease sequences which results in a sequence identity of 67%. This high genetic diversity among human noroviruses indicates the need for an improved model of the GII.4 protease (18,19). In this report, we describe for the first time the determination of the crystal structure of norovirus GII.4 (Minerva virus) protease through X-ray crystallography. To further investigate the GII.4 protease structure and mechanism, a comparative analysis of the GI.1 and GII.4 protease strains was conducted by evaluating both enzymatic and proteolytic activity using the fluorescence resonance energy transfer (FRET) assay as well as computational analyses such as molecular docking and binding energy estimations. Additionally, we describe the design and characterization of a novel set of covalent inhibitors using a FRET based assay as well as a thermal shift assay. The elucidation of the GII.4 norovirus protease structure we anticipate to greatly improve the structure-based design of potent therapeutic inhibitors for future studies. Materials and Methods Expression and Purification. Plasmids expressing norovirus GI.1 and GII.4 protease were constructed by cloning into the XhoI and BamHI sites of a pET28a vector (GeneArt). The DNA fragment encoding the NS6 protease (Genbank: EF684915) was amplified by PCR using the following primers: 5’GAATAAGAAGACATAGGTGCCCCACCAAGCATC-3’ (forward); 5’GATACGCTCGAGTTATTCAAGTGTAGCTTCC-3’ (reverse). The active site of the protease was mutated from Cys139 to Ala by site-directed mutagenesis following the provided protocol to generate a catalytically inactive construct. The inactive Minerva virus protease (iNoV GII.4 PR) containing the C139A mutation was cloned and purified (unpublished data). Briefly, the inactive Minerva virus protease C139A gene was shuttled into a pCDF-SUMO vector containing an N-terminal His6-SUMO tag, and the clone was transformed into BL21 (DE3) cells for protein expression. Cells were grown in LB media supplemented with 34 µg/mL streptomycin until an optical density reached 0.5. Cultures were induced with 0.1 mM IPTG and grown 3 hours at 37°C. Harvested cells were lysed using the French press method, and the supernatant was collected for purification. His6-SUMO-iNoV GII.4 PR was captured by passing the supernatant through a HisTrap column (GE Healthcare), and an imidazole gradient was applied to the column to elute His6-SUMO-iNoV GII.4 PR. The SUMO-tag was digested out using Ulp1, and the digested products were separated by a second passage through the HisTrap column. The native inactive Minerva protease was further purified using Superdex 200 column (GE Healthcare) and homogenized into gel filtration buffer (20 mM Tris pH 7.5, 150 mM NaCl, 5% glycerol). Purified inactive Minerva protease was pooled, concentrated to 28 mg/mL, and flash frozen into the 80°C. Thermal Shift Assay. Thermal shift assays were performed using an Applied Biosystems 7500 Real-time PCR system to identify thermal stable conditions for protein crystallization. Purified Minerva GII.4 protease inactive (iNoV GII.4 PR) and SYPRO Orange (5000x stock, Invitrogen) concentrations were optimized to 0.02 mg/ml and 5x, respectively, to generate a well-defined melt curve. Following concentration optimization, iNoV GII.4 PR was screened against 70 buffer conditions consisting of a range of common buffers, varying pH values, NaCl concentrations, as well as confirmed conditions used to crystallize the GI.1 protease (9, 19). Inactive GII.4 protease was loaded in a 96 well plate at 20 uM with the calculated concentration of buffer condition at a ratio of 1:1, protein to buffer condition, and a final SYPRO Orange concentration of 5x. The buffer screen assay was run as a standard curve from 2 ACS Paragon Plus Environment

Page 2 of 14

Page 3 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease 25-95˚C with a ramp speed of 0.5 degrees/min. Raw data were normalized to generate a melting and first derivative curve (7500 Software v2.0.1). For evaluation of inhibitors to be used for co-crystallization, the structure-based compounds, A and B (synthesis of compound A and B in Figure S2 and S3, respectively), were assayed at serial saturations of 1:12.5, 1:25, 1:50, 1:100, and 1:200 protein to inhibitor ratios. The protein buffer was exchanged from gel filtration buffer (20 mM Tris pH 7.5, 150 mM NaCl, 5 % glycerol) to optimized GII.4 protease buffer (20mM MES pH6.0) with Micro Bio Spin Chromatography Column (Bio Rad). SYPRO Orange was plated at a 5x final concentration. The compound screen assay was run as a standard curve from 25-95˚C with a ramp speed of 0.5 degrees/min. Data were analyzed using the normalized melt curve and derivative curve (7500 Software v2.0.1). Crystallization. Purified Minerva C139A protease was buffer exchanged from gel filtration buffer to 20 mM MES pH 6 using Micro Bio Spin Chromatography Columns (Bio Rad) and subsequently concentrated using Ultrafree-MC concentrators (Millipore, 10kD) to 20 mg/ml. 5% (v/v) glycerol was added to the protein samples prior to crystallization screens. Crystals were produced using the sitting drop method by mixing 1:1 volume ratio of 20 mg/mL MVpro and reservoir solution (100 mM Tris pH 8.9, 16% PEG 8K, 600 mM Li2SO4) and incubated at 20°C. Full sized crystals were harvested after one month. Crystals showing a hexagonal morphology were transferred into cryoprotectant consisting of mother liquor and 20% (v/v) glycerol then directly flash-frozen in liquid nitrogen. Samples were shipped to the Advanced Photon Source (APS) at Argonne National Laboratory for X-ray data collection. Data Collection, Structure Determination and Refinement. Table 1 contains relevant data collection and refinement statistics. Diffraction data was collected at the Advanced Photon Source (APS) at the Argonne National Laboratory (Lemont, IL) using the Life-Science Collaborative Access Team (LS-CAT) 21-ID-F beamline. The NoV GII.4 PR crystals were indexed into P31 space group with eight molecules per asymmetric unit. Observable reflections were integrated and scaled using XDS and AIMLESS via Xia2, respectively. Phases were solved by molecular replacement using PHASER within PHENIX and employing monomeric Norwalk virus protease (PDB ID = 3UR6) as the search model. Automated model building was performed using PHENIX autobuild (30). Multiple rounds of model building and refinement were carried out in WinCoot and PHENIX refine, respectively (25-33). Computational Analysis. Docking studies as well as ADME and toxicity screens were carried out using QPLD (Quantum Polarized Ligand Docking) and Qikprop in Maestro (Schrodinger, v2017-2). Crystal structures were analyzed for polar contacts and interactions through Pymol (PyMolWin). Covalent Docking of Compounds A and B was performed using CovDock in Maestro and Cys 139 was specified as the reactive residue undergoing nucleophilic addition to a double bond. Kinetic Enzyme Assay and IC50 Assay. Purified NoV GII.4 PR was tested for enzymatic activity with a FRET-based fluorometric enzyme assay. The NoV GII.4 PR was diluted in reaction buffer (50 mM HEPES, pH 8.0, 120 mM NaCl, 0.4 mM EDTA, 20% glycerol, and 4 mM DTT) to a final concentration of 128 nM. Each reaction was initiated by addition of FRET substrate (HiLyte Fluor 488 – DFELQGPK (QXL520), purchased from Anaspec, Inc.). The FRET substrate is a peptide that represents the Minerva virus NS2/NS3 cleavage site. In order to determine kinetic parameters, the FRET substrate was serially diluted with DMSO to final concentrations of 100 µM to 49 nM and added to the reaction. The fluorescence emitted by substrate cleavage was monitored by a microplate reader (SpectraMax M5, Molecular Devices, Sunnyvale, CA, USA) at a 488 nm excitation wavelength and a 520 nm emission wavelength. Readings were taken every minute for 20 minutes, and the reactions was performed at 37°C. To convert relative fluorescence units (RFU) into molar concentrations, a standard curve was created by measuring fluorescence of free HiLyte Fluor 488, which was serially diluted from 250 nM to 3.9 nM. All data were plotted and analyzed with GraphPad Prism v. 6.07. 3 ACS Paragon Plus Environment

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 14

Structural and antiviral studies of the human norovirus GII.4 protease Compound IC50 experiments were designed based on enzymatic activity previously described(34). GII.4 Minerva virus protease and GI.1 Norwalk virus protease were diluted in buffer (50 mM HEPES pH 8.0, 120 mM NaCl, 0.4mM EDTA, 20% glycerol (v/v), 4mM DTT) to a final well concentration of 128 nM and incubated at 37°C for 10 minutes. Compounds A and B were serially diluted in DMSO for a final well concentration of 64 µM-62.5 nM and incubated with diluted protein for 90 minutes at 37 oC. FRET substrate (HiLyte Fluor 488 – DFELQGPK (QXL520), Anaspec, Inc.) was then added at a final well concentration of 15 µM and a final reaction volume of 100 µL. Fluorescence was measured every minute for 20 minutes at 37oC on a SpectraMax M5 microplate reader (Molecular Devices, Sunnyvale, CA, USA) with an excitation wavelength of 488 nm and an emission wavelength of 520 nm. GraphPad Prism v6.07 was used to analyze the data and determine anti-proteolytic activity against GII.4 Minerva and GI.1 Norwalk virus proteases. Accession number for protein structure. The native protein crystal structure coordinates and structure factors have been deposited under accession number 6B6I in the Protein Data Bank (www.pdb.org). Results Stabilization of purified NoV GII.4 protease leads to diffraction quality crystallization. Initial Minerva C139A protease crystals were produced using 20% PEG 4000, 0.1M Tris pH 7.5, 200 mM Li2SO4 and 10 mM Mn(II)Cl2 using the sitting drop method. Unfortunately, these crystals were limited to a 6 Å diffraction with a thin, hollow needle morphology. Inactive norovirus GII.4 protease (iNoV GII.4 PR, C139A) was screened against a variety of buffers with varying parameters using the thermal shift assay, as described in the methods section. The most noteworthy condition was determined to be a buffer of 20 mM MES and 5% glycerol which increased the protein melting temperature compared to the original gel filtration buffer (data not shown). Moving forward, a five-plate screen expanded from the original needle condition was carried out following the protein buffer exchange to optimal conditions. This screen facilitated the formation of a hexagonal crystal morphology in the crystallization condition: 100 mM Tris pH 8.9, 16% PEG 8K, 600 mM Li2SO4. These crystals, described in Figure S1, diffracted at a resolution of 2.4 Å. Table 1. X-ray data collection and refinement statistics of norovirus GII.4 protease Values in parentheses reflect the highest resolution shell. Data collection Space group P31 Unit cell (a, b, c, Å) 112.94, 112.94, 153.37 90, 90, 120 (, , , ) NoV GII.4 protease molecules/ ASU 8 Wavelength (Å) 0.97872 Resolution range (Å) 52.99-2.44 (2.50-2.44) Completeness (%) 100.0 (99.9) Multiplicity 6.4 (6.5) No. unique reflections 81454 (6060) Wilson B-Factor (Å2) 19.15 10.4 (1.2) Mean I/I Rmerge 0.094 (1.355) CC1/2 0.998 (0.571) Refinement Rwork/ Rfree 0.1796/ 0.2220 No. of atoms 10640 Model Quality R.m.s deviations Bond lengths (Å) 0.008 Bond angles (º) 1.22 PDB code 6B6I Rmerge=∑|Ihkl-{Ihkl}|/ ∑|Ihkl| with I running over the number of independent observations of reflection hkl. represent intensity of an individual reflection and the mean intensity, respectively.

4 ACS Paragon Plus Environment

Ihkl and {Ihkl}

Page 5 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease

Overall Structure of the Norovirus GII.4 Protease. The Advanced Photon Source at Argonne National Laboratory produced X-ray diffraction data at a 2.4 Å resolution. The structure was solved through molecular replacement in Phaser-MR and subsequent refinements were carried out through phenix.refine (Phenix 1.9-1692)(30). The resulting refined model had an overall B-factor of 19.15 Å2 allowing proper main chain and side chain modeling. The phased and refined structure resulted in Rfree value of 22.20% and Rwork value of 17.96%. The iNoV GII.4 PR crystal structure was defined as a P31 space group with 8 molecules (Chain A-H) per crystallographic asymmetric unit. Alignment of the 8 chains (A-H) shows four (B, D, F, H) of the eight chains containing electron density for the C-terminal tail that can extend into the active site of a neighboring chain. Examination of the crystal model electron density shows Chains B and F extending into the active site of Chains E and A, respectively. Additionally, the other four chains lack electron density for the C-terminal tail (A, C, E, G). Figure 1A illustrates the C-terminal tail of chain B extending into the active site of chain E. The active site of norovirus GII.4 protease is defined by the catalytic triad consisting of His30 (base), Glu54 (acid), and Cys139 (nucleophile mutated to Ala139 in the crystal structure) (Figure 1D) and is conserved across norovirus proteases according to sequence alignments(7). Chain B H[Grab

Glu54

His30 Chain B [Grab Ala139

Chain E E[Grab

Chain A H[Grab

Chain E E

Glu54 H[Grab His30

Ala139

Figure 1. A) Pymol cartoon representation of the three chain positions. Chain B (blue) and chain E (green) are the asymmetric dimer with binding activity. Chain A is lacking electron density for the Cterminal tail. B) Pymol cartoon representation of the asymetric dimer, chain B and E (blue and green, respectively), with the C-terminal tail from chain B (blue) extending into the neighboring active site of chain E (green). C) Stick model of the C-terminal tail of chain B extending into the active sire of chain E (green). D) Cartoon model of chain A and the defined catalytic triad consisting of Glu54, His30, and the mutated Cys139Ala. E) Pymol stick representation of the electron-density map for the C-terminal tail of chain B (grey) extended into the active site of the neighboring chain E (green).


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Structural and antiviral studies of the human norovirus GII.4 protease Structure analysis reveals a chymotrypsin-like protease fold. The N-terminal domain begins with an α-helix leading to a twisted β-sheet which is then partnered with the C-terminal domain consisting of a β-barrel. Further chain analysis shows the N-terminal domain beginning with an α-helix containing two turns (residues 3-8) followed by the five stranded β-sheet (residues 9-11, 16-21, 24-28, 50-52, and 55-59) containing two additional single turn α-helices (residues 29-31 and 44-46). The N-terminal five stranded βsheet is then connected by an extended loop region (residues 60-79) connected to the six stranded β-barrel (residues 80-89, 92-109, 112-121, 140-147, 150-157, and 166-170). Between the first three strands and the last three strands of the β-barrel lies a loop region containing an α-helix (residues 127-130). As described above, both the N-terminal domain and the C-terminal domain are composed of loop regions that are missing electron density. Binding energy was calculated between the chains based on protein interface surface area using an online server (ExPASY). It was determined that the ΔG (kcal/mol) was best for the pair of chains H:C > E:B > G:D > F:A (-16.1, -15.6, -15.4, -14.5, respectively). These binding energies represent the chains containing the C-terminal tail (B, D, F, H) and the chains lacking the electron density of the C-terminal tail (A, C, E, G) with an average ΔG = -15.4 kcal/mol. The next best surface interface binding energy is between neighboring chains with an average ΔG = -10.0 kcal/mol. The NoV GII.4 PR sequence has been defined as 181 amino acids. Chains A, C, E, and G lack electron density at the C-terminal end of the peptide, resulting in 173, 173, 174, and 173 amino acids solved for in the model, respectively. Alternatively, three of the chains (B, D, F) lack electron density at the N-terminus of the peptide resulting in a sequence of 179, 178, and 178 amino acids defined in the model, respectively. This is represented in the crystal structure model where the C-terminal tail extends into the active site of the neighboring chain (Figure 1E). It should be noted that only chain H showed electron density for the full 181 amino acid sequence. Crystal structure substrate binding site, oxyanion hole, and catalytic triad analysis enhances drug design and optimization. Hydrogen bonding was viewed for the last 5 residues of the C-terminal tail in the adjacent active site. Additionally, the P1-P4 sidechains of NoV GII.4 PR were compared to that of the NoV GI.1 PR using the residues 178-181 of the NoV GII.4 PR sequence (ATLE) and docked ATLE into the active site. The comparison of the binding pocket interactions with the ATLE ligand shows a high similarity in structure binding conformation. Glu 54 and His 30 are situated in such a way as to stabilize and promote the ideal conditions for binding of the substrate and catalytic activity of Cys 139 (Figure 2). The active site cleft consists of the catalytic triad and an electrophilic oxyanion hole. The electrophilic oxyanion hole has a role in the stabilization of the tetrahedral intermediate of the ligand covalently linked to the protease. The electrostatic potential surface model of the NoV GII.4 PR crystal structure compared to the NoV GI.1 PR crystal structure (pdb 4INH) is shown in Figure 2. According to previous work the S1 pocket consists of residues 134-138, and 157, the S2 pocket consists of residues 30, 109, 112-113, and 158-159, and the S4 pocket consists of residues 107, 118, 161, 166, and 168. Additionally, substrate backbone residues include 110, 158, and 160 which contribute to the hydrogen bonding network(20). Substrate backbone binding residue, Ala 160, is located in a loop region in the twisted beta-sheet near the N-terminus in the S1 pocket and binds to Glu 177 of the C-terminal tail substrate (Figure 2). Comparative analysis of the hydrogen bonding network for the solved NoV GII.4 and GI.1 PR structure with the ATLE ligand confirms interactions at the S1 pocket (residues 128, 134 and 137), S2 pocket (residue 158-159), and the substrate backbone (residue 158 and 160). As mentioned in previous work, the S3 pocket is not present and T179 is exposed to solvent in the ligand. However, differences in the pockets can be seen when comparing the overall model in Figure 2. GII.4 has an overall smaller binding site than GI.1. However, the S4 pocket appears to have increased solvent exposure.


Page 6 of 14

Page 7 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry


A)

P2

P4

P3

P1

B)

P2

P4

P3 P1

Figure 2. Norovirus GI.1 and GII.4 PR structures were aligned at the α-Carbon of the catalytic triad (Cys139Ala, His30, Glu54) using the Structure Alignment Tool in Maestro (Schrodinger 2017-1). The natural NoV GII.4 PR C-terminal ligand, ATLE, was then aligned in the active site of the two structures and the binding pocket visualization was carried out individually in Pymol. A) Left. Electrostatic potential surface model of NoV GII.4 PR crystal Chain E created with Poisson-Boltzmann ESP by Maestro with stick model of the ATLE ligand in green. Right. Representation of the ATLE ligands spatial distribution in the NoV GII.4 PR binding pockets using Surface Model; P1:blue, P2:yellow, P3:red, P4:green. Appropriate pockets are indicated in yellow. B) Left. Electrostatic potential surface model of NoV GI.1 PR crystal Chain E created with Poisson-Boltzmann ESP by Maestro with stick model of the TALE ligand in green. The ATLE ligand was mutated to the natural GI.1ligand, TALE, using Maestro Mutate Residue function. Right. Representation of the TALE ligands spatial distribution in the NoV GI.1 PR binding pockets using Surface Model; P1:blue, P2:yellow, P3:red, P4:green. Appropriate pockets are labeled. 7 ACS Paragon Plus Environment

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Structural and antiviral studies of the human norovirus GII.4 protease FRET Assay confirms the need to design inhibitors based on the NoV GII.4 PR crystal structure. Compounds A and B were designed as previously described and tested for inhibitory activity against NoV GI.1 and GII.4 PR using a FRET assay (Table 2)(21,35). Compound A had an IC50 of 0.077 µM and 0.074 µM for GI.1 and GII.4, respectively. Compound B had an IC50 of 0.23 µM and 2.41 µM for GI.1 and GII.4, respectively. These results coincide with our recently published data of compounds designed against NoV GI.1 PR (21). The inhibition of NoV GI.1 PR is nearly 10-fold greater when compared to the NoV GII.4 PR against our synthesized compounds. Table 2. Compound Kinetic Results Anti-norovirus activity (µM) Compound

IC50 (GI.1)

IC50 (GII.4)

A

0.077

0.074

B

0.225

2.409

Covalent docking was carried out for Compound B that was initially designed for the homology based on the NoV GI.1 PR structure. The inactive Ala139 was mutated to the active Cys139 to more accurately predict compound binding against NoV GII.4 PR. Compound B docked with greater accuracy for the predicted conformation to NoV GI.1 PR than to to NoV GII.4 PR. The difference in compounds is the P3 side chain of Compound A contains a naphthalene as well as a modified cap. The oxyanion hole in conjunction with either the hydroxyl or thiol side chain traditionally substantiate tight binding and stabilization of tetrahedral transition state formation in both serine and cysteine proteases(22). Zeitler et al. state that the oxyanion hole of the NoV GI.1 PR is formed sequentially by residues 137-140(17). Comparative analysis of the NoV GII.4 PR oxyanion hole with that of the GI.1 sequence confers conservation among the catalytic triad and oxyanion hole with Gly137-Asp138-Cys139Gly140 as the determined amino acid sequence. Additionally, there is high similarity in positioning of our compound compared to the substrate relative to the oxyanion hole (Figure 3). Despite the similarity in the overall docking, there are several structural differences observed that show an effect in docking modes. The first is in relation to the positioning of the S2 sub-pocket. In the GI.1 structure, Arg112 points outward allowing for proper positioning of the P2 moiety in the S2 sub-pocket. However, in the GII.4 structure, Arg112 points inward and clashes with the P2 moiety of the peptidomimetic. This is shown in Figure 3B where the P2 moiety is occluded from the P2 sub-pocket. The second observation is in the S4 sub-pocket. For GI.1, residue 162 is an outward pointing lysine whereas for GII.4 it is an inward pointing arginine. This Lys162Arg mutation interferes with the positioning of the P4 moiety. A third observation looks at the S118M mutation from GI.1 to GII.4, respectively. In the GI.1 structure, residue 118 is a serine whereas in the GII.4 structure residue 118 is a methionine. In the GII.4 structure, Met107 rotates to avoid a steric clash with Met118 which causes the P4 sub-pocket to shrink. Finally, with the above residue mutations and positioning, Compound B struggles to covalently bind and modify the GII.4 active site. However, the guanidium group in the Lys162Arg mutation will allow for additional hydrogen bonding to be exploited.

Discussion We report here the elucidation of the X-ray crystallographic structure, computational analyses, and the enzymatic inhibition of norovirus GII.4 Minerva virus protease. Initially, expression and purification of the iNoV GII.4 PR revealed the presence of both a monomer and dimer in solution (34). Crystallographic studies support the formation of the protease dimer in solution. As described previously, norovirus GII.4 protease consists of eight molecules per asymmetric unit subsequently composed of four asymmetric dimers. Each dimer consists of a chain containing the C-terminal tail that extends into the neighboring active site, and a chain lacking the electron density for the C-terminal tail. The missing electron density is 8 ACS Paragon Plus Environment

Page 8 of 14

Page 9 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease likely due to the solvent exposure of the C-terminal tail and the inability of the beam to detect them. The N-terminal domain contains a twisted β-sheet partnered with the C-terminal domain containing a β-barrel with a substrate binding channel located between them. Binding energy calculations reveal higher values for the partnered chains impacting the overall ability for the dimer to maintain its conformation in solution. These structural insights were previously observed in the GI Norwalk 1968 strain (23). Crystallographic analysis also revealed conserved Van der Waals interactions and hydrogen bonding among the dimers. Extensive efforts had been made in our laboratories to crystallize the NoV GII.4 protease at a resolution capable of structural analysis and characterization. According to recently published work, successful crystallization can be made possible in part by employing differential scanning fluorimetry to successfully stabilize the protein of interest through buffer optimization (24). Optimizing the protein buffer and conditions facilitated the formation of hexagonal crystals with a diffracted resolution of 2.4 Å (Accession Number: 6B6I). 6B6I is the first crystal structure of the norovirus GII.4 protease. Extensive efforts had been made in our laboratories to crystallize the NoV GII.4 protease at a resolution capable of structural analysis and characterization. According to recently published work, successful crystallization can be made possible in part by employing differential scanning fluorimetry to stabilize the protein of interest through buffer optimization (24). Optimizing the protein buffer and conditions facilitated the formation of hexagonal crystals with a diffracted resolution of 2.4 Å (Accession Number: 6B6I). 6B6I is the first crystal structure of the norovirus GII.4 protease. Previous research shows the NoV GII.4 PR as a cysteine protease with a catalytic triad containing Cys139 with the catalytic thiolate anion, His30, and Glu54 to complete the triad. Additionally, Glu54 has been shown to play an important role in the proteases catalytic activity (13). The inactive C139A mutated construct that was generated portrays the catalytic triad through the binding of the C-terminal tail to the neighboring active site. Our analyses also reveal that the positioning of our compound in comparison to the natural ligand relative to the active site and binding pockets retains a high conformational similarity. Additionally, docking experiments for two potential inhibitors as well as confirmation by FRET assay data provide support that the initial homology model of the GII.4 Minerva virus PR based on the GI.1 Norwalk virus PR crystal structure does not allow optimal design of inhibitors. The inhibition of NoV GI.1 PR is nearly 10-fold greater for most of the designed compounds and supports the need for additional modifications that fit the NoV GII.4 PR model more accurately. Additionally, the efficacy of Compound A provides a good starting point for modifications despite its high toxicity (21). Through protein purification, stability optimization, and crystallographic studies of the solubilized enzyme a 2.4 Å crystal structure was solved and utilized in computational studies to assist in the design of lead compounds for the inhibition of human norovirus. Despite the overwhelming prevalence of norovirus GII.4 outbreaks, this strain remains highly underrepresented in the literature. Up until now, the GII.4 homology model based on the GI.1 structure was inadequate for designing structure-based compounds. Solving the crystal structure of the most infectious and widespread strain of norovirus will aid in the design of more potent and efficacious protease inhibitors. Future studies will focus on comparative analyses of the three norovirus strains responsible for the majority of outbreaks (GII.4, GI.1, GIV.1, respectively) as well as the design and optimization of inhibitors to target norovirus proteases through structure-based drug design. Utilization of these analyses provides atomic level structural detail to further enhance the design and synthesis of effective protease inhibitors.


Biochemistry


1 2 3 NoV GI.1 PR + Compound B 54 107 B) 4 A) 109 30 5 112 31 110 S2 6 114 7 P2 134 8 S3 P1 9 135 10 P3 136 11 12 137 13 138 14 158 139 157 160 159 15 16 161 17 S1 18 P4 162 168 19 167 20 166 21 22 23 NoV GII.4 PR + Compound B D) 24 C) P4 25 26 S2 27 P3 139 28 29 138 S3 30 109 31 157 137 110 P2 32 136 158 33 112 34 135 159 35 S1 30 114 134 36 P1 160 128 54 37 38 162 39 40 41 42 Figure 3. A) Electrostatic potential surface model of norovirus GI.1 PR covalently docked with 43 Compound B (stick format in green) using Maestro CovDock (Schrodinger 2017-1). Pockets are labeled 44 S1, S2, S4. B) Ligand Interaction Diagram of A. using Maestro. Residues are numbered accordingly. R45 groups are labeled P1, P2, P3, P4. C) Electrostatic potential surface model of NoV GII.4 PR covalently 46 docked with Compound B (stick format in purple). Pockets are labeled S1, S2, S4. D) Ligand Interaction 47 Diagram of C. using Maestro. Residues are numbered accordingly. R-groups are labeled P1, P2, P3, P4. 48 49 Legend for B and D: Distance 50 Charged (negative) Polar H-bond 51 Charged (positive) Unspecified Residue Metal coordination 52 Glycine Water Pi-Pi stacking 53 Hydrophobic Salt bridge 54 Pi-cation Metal Solvent Exposure 55 56 57 58 10 59 ACS Paragon Plus Environment 60

Page 10 of 14

Page 11 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease Acknowledgements We would like to thank Wayne State University School of Medicine for supporting the Ph.D. studies of KMM. We would like to thank Wayne State University School of Medicine for a Rumble Fellowship supporting the Ph.D. studies of BDK. LCK is supported by the Wayne State University Grants Boost Program. RFS is supported in part by NIH grant AI-129607 and by the Center for AIDS Research grant 2P30AI-050409. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. Use of the LS-CAT Sector 21 was supported by the Michigan Economic Development Corporation and the Michigan Technology Tri-Corridor (Grant 085P1000817). Conflict of Interest The authors declare no conflict of interest. Supporting Information Crystal photographs and synthesis of Compounds A and B.


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References 1. Koo, H. L., Neill, F. H., Estes, M. K., Munoz, F. M., Cameron, A., DuPont, H. L., and Atmar, R. L. (2013) Noroviruses: The Most Common Pediatric Viral Enteric Pathogen at a Large University Hospital After Introduction of Rotavirus Vaccination. J Pediatric Infect Dis Soc 2, 57-60 2. Lopman, B. A., Steele, D., Kirkwood, C. D., and Parashar, U. D. (2016) The Vast and Varied Global Burden of Norovirus: Prospects for Prevention and Control. PLoS Medicine 13, e1001999 3. Jia, L., Zhang, Y., Liu, L., Dong, H., Zhao, L., and Qian, Y. (2017) High prevalence of GII norovirus in hospitalized children with acute diarrhea, in Beijing. PLoS ONE 12, e0179839 4. Bartsch, S. M., Lopman, B. A., Ozawa, S., Hall, A. J., and Lee, B. Y. (2016) Global Economic Burden of Norovirus Gastroenteritis. PLoS ONE 11, e0151219 5. Bok , K., and Green , K. Y. (2012) Norovirus Gastroenteritis in Immunocompromised Patients. New England Journal of Medicine 367, 2126-2132 6. Dai, Y.-C., Zhang, X.-F., Xia, M., Tan, M., Quigley, C., Lei, W., Fang, H., Zhong, W., Lee, B., Pang, X., Nie, J., and Jiang, X. (2015) Antigenic Relatedness of Norovirus GII.4 Variants Determined by Human Challenge Sera. PLOS ONE 10, e0124945 7. Nakamura, K., Someya, Y., Kumasaka, T., Ueno, G., Yamamoto, M., Sato, T., Takeda, N., Miyamura, T., and Tanaka, N. (2005) A Norovirus Protease Structure Provides Insights into Active and Substrate Binding Site Integrity. Journal of Virology 79, 13685-13693 8. Deng, L., Muhaxhiri, Z., Estes, M. K., Palzkill, T., Prasad, B. V., and Song, Y. (2013) Synthesis, Activity and Structure-Activity Relationship of Noroviral Protease Inhibitors. Medchemcomm 4 9. Kim, Y., Lovell, S., Tiew, K. C., Mandadapu, S. R., Alliston, K. R., Battaile, K. P., Groutas, W. C., and Chang, K. O. (2012) Broad-spectrum antivirals against 3C or 3C-like proteases of picornaviruses, noroviruses, and coronaviruses. J Virol 86, 11754-11762 10. Donaldson, E. F., Lindesmith, L. C., LoBue, A. D., and Baric, R. S. (2010) Viral shape-shifting: norovirus evasion of the human immune system. Nat Rev Micro 8, 231-241 11. Prior, A. M., Kim, Y., Weerasekara, S., Moroze, M., Alliston, K. R., Uy, R. A., Groutas, W. C., Chang, K. O., and Hua, D. H. (2013) Design, synthesis, and bioevaluation of viral 3C and 3C-like protease inhibitors. Bioorg Med Chem Lett 23, 6317-6320 12. May, J., Korba, B., Medvedev, A., and Viswanathan, P. (2013) Enzyme kinetics of the human norovirus protease control virus polyprotein processing order. Virology 444, 218-224 13. Hussey, R. J., Coates, L., Gill, R. S., Erskine, P. T., Coker, S. F., Mitchell, E., Cooper, J. B., Wood, S., Broadbridge, R., Clarke, I. N., Lambden, P. R., and Shoolingin-Jordan, P. M. (2011) A structural study of norovirus 3C protease specificity: binding of a designed active site-directed peptide inhibitor. Biochemistry 50, 240-249 14. Takahashi, D., Kim, Y., Lovell, S., Prakash, O., Groutas, W. C., and Chang, K. O. (2013) Structural and inhibitor studies of norovirus 3C-like proteases. Virus Res 178, 437-444 15. Chang, K. O., Takahashi, D., Prakash, O., and Kim, Y. (2012) Characterization and inhibition of norovirus proteases of genogroups I and II using a fluorescence resonance energy transfer assay. Virology 423, 125-133 16. Someya, Y., Takeda, N., and Miyamura, T. (2005) Characterization of the norovirus 3C-like protease. Virus Res 110, 91-97 17. Zeitler, C. E., Estes, M. K., and Venkataram Prasad, B. V. (2006) X-ray crystallographic structure of the Norwalk virus protease at 1.5-A resolution. J Virol 80, 5050-5058 18. Robilotti, E., Deresinski, S., and Pinsky, B. A. (2015) Norovirus. Clin Microbiol Rev 28, 134-164 19. Takahashi, D., Hiromasa, Y., Kim, Y., Anbanandam, A., Yao, X., Chang, K. O., and Prakash, O. (2013) Structural and dynamics characterization of norovirus protease. Protein Sci 22, 347-357


Page 12 of 14

Page 13 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Structural and antiviral studies of the human norovirus GII.4 protease 20. 21.

22.

23.

24.

25. 26. 27. 28.

29. 30.

31. 32.

33. 34.

35.

Zeitler, C. E., Estes, M. K., and Venkataram Prasad, B. V. (2006) X-Ray Crystallographic Structure of the Norwalk Virus Protease at 1.5-Å Resolution. Journal of Virology 80, 5050-5058 Amblard, F., Zhou, S., Liu, P., Yoon, J., Cox, B., Muzzarelli, K., Kuiper, B. D., Kovari, L. C., and Schinazi, R. F. (2018) Synthesis and antiviral evaluation of novel peptidomimetics as norovirus protease inhibitors. Bioorganic & Medicinal Chemistry Letters 28, 2165-2170 Shokhen, M., Traube, T., Vijayakumar, S., Hirsch, M., Uritsky, N., and Albeck, A. (2011) Differentiating Serine and Cysteine Proteases Mechanism by New Covalent QSAR Descriptors. Chembiochem : a European journal of chemical biology 12, 1023-1026 Muhaxhiri, Z., Deng, L., Shanker, S., Sankaran, B., Estes, M. K., Palzkill, T., Song, Y., and Prasad, B. V. V. (2013) Structural Basis of Substrate Specificity and Protease Inhibition in Norwalk Virus. Journal of Virology 87, 4281-4292 Reinhard, L., Mayerhofer, H., Geerlof, A., Mueller-Dieckmann, J., and Weiss, M. S. (2013) Optimization of protein buffer cocktails using Thermofluor. Acta Crystallographica Section F: Structural Biology and Crystallization Communications 69, 209-214 Evans, P. R. (2006) Scaling and assessment of data quality. Acta Cryst. D62 Tickle, I. J., Flensburg, C., Keller, P., Paciorek, W., Sharff, A., Vonrhein, C., & Bricogne, G. (2017) STARANISO. Vonrhein, C., Flensburg, C., Keller, P., Sharff, A., Smart, O., Paciorek, W., Womack, T. & Bricogne, G. . (2011) Data processing and analysis with the autoPROC toolbox. Acta Cryst. D67, 293-302 Afonine, P. V., Grosse-Kunstleve, R. W., Echols, N., Headd, J. J., Moriarty, N. W., Mustyakimov, M., Terwilliger, T. C., Urzhumtsev, A., Zwart, P. H., and Adams, P. D. (2012) Towards automated crystallographic structure refinement with phenix.refine. Acta Crystallogr. Sect. D Biol. Crystallogr. 68, 352-367 McCoy, A. J., Grosse-Kunstleve, R. W., Adams, P. D., Winn, M. D., Storoni, L. C., and Read, R. J. . (2007) Phaser crystallographic software. J. Appl. Crystallogr. 40, 658-674 Terwilliger, T. C., Grosse-Kunstleve, R. W., Afonine, P. V., Moriarty, N. W., Zwart, P. H., Hung, L.W., Read, R. J., and Adams, P. D. . (2008) Iterative model building, structure refinement and density modification with the PHENIX AutoBuild wizard. Acta Crystallogr. Sect. D Biol. CrysTallogr 64, 61-69 Kabsch, W. (2010) XDS. Acta Crystallographica Section D 66, 125-132 Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J., Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie, A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N., Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin, A., and Wilson, K. S. (2011) Overview of the CCP4 suite and current developments. Acta Crystallographica Section D: Biological Crystallography 67, 235-242 Evans, P. R., and Murshudov, G. N. (2013) How good are my data and what is the resolution? Acta Crystallographica Section D: Biological Crystallography 69, 1204-1214 Kuiper, B.D., Keusch, B.J.., Holcolm, J., Muzzarelli, K.M., Amblard, F., Liu, P.,Zhou, S., Kovari, I.A., Yang, Z., Schinazi, R.F., and Kovari, L.C. (2018) Expression, Purification and Characterization of a GII.4 Norovirus Protease from Minerva Virus. Infectious Disorders - Drug Targets 18, 1-9 Emmott, E., Sweeney, T.R., Goodfellow, I. (2015) A cell-based Fluorescence Resonance Energy Transfer (FRET) sensor reveals inter- and intragenogroup variations in Norovirus protease activity and polyprotein cleavage. J Biol Chem 290(46), 27841-27853


Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Structural and antiviral studies of the human norovirus GII.4 protease For Table of Contents Use Only Kendall M. Muzzarelli1, Benjamin Kuiper1, Nicholas Spellmon1, Joseph Brunzelle2, Justin Hackett1, Franck Amblard3, Shaoman Zhou3, Peng Liu3, Iulia A. Kovari1, Zhe Yang1, Raymond F. Schinazi3, Ladislau C. Kovari1* 1

Wayne State University School of Medicine, Department of Biochemistry, Microbiology and Immunology, Detroit, MI 48201, USA;2Synchrotron Research Center, Life Science Collaborative Access Team, Northwestern University, Argonne, Illinois, USA ; 3Center for AIDS Research, Laboratory of Biochemical Pharmacology, Department of Pediatrics, Emory University School of Medicine, Atlanta, GA 30322, USA


Page 14 of 14

Structural and antiviral studies of the human norovirus GII.4 protease

Recommend Documents