Heteroaryldihydropyrimidine (HAP) and Sulfamoylbenzamide (SBA) Inhibit Hepatitis B Virus Replication by Different Molecular Mechanisms

Heteroaryldihydropyrimidine (HAP) and sulfamoylbenzamide (SBA) are promising non-nucleos(t)ide HBV replication inhibitors. HAPs are known to promote core protein mis-assembly, but the molecular mechanism of abnormal assembly is still elusive. Likewise, the assembly status of core protein induced by SBA remains unknown. Here we show that SBA, unlike HAP, does not promote core protein mis-assembly. Interestingly, two reference compounds HAP_R01 and SBA_R01 bind to the same pocket at the dimer-dimer interface in the crystal structures of core protein Y132A hexamer. The striking difference lies in a unique hydrophobic subpocket that is occupied by the thiazole group of HAP_R01, but is unperturbed by SBA_R01. Photoaffinity labeling confirms the HAP_R01 binding pose at the dimer-dimer interface on capsid and suggests a new mechanism of HAP-induced mis-assembly. Based on the common features in crystal structures we predict that T33 mutations generate similar susceptibility changes to both compounds. In contrast, mutations at positions in close contact with HAP-specific groups (P25A, P25S, or V124F) only reduce susceptibility to HAP_R01, but not to SBA_R01. Thus, HAP and SBA are likely to have distinctive resistance profiles. Notably, P25S and V124F substitutions exist in low-abundance quasispecies in treatment-naïve patients, suggesting potential clinical relevance.

Supplementary Figure S1. Electron micrographs of compound-treated core protein assembly at different dimerto-compound ratios. Up left: 5 μM core protein dimer incubated with 5 μM HAP_R01. The same image in Figure 1d is used for comparison. Up right: 5 μM core protein dimer incubated with 5 μM SBA_R01. The same image in Figure 1d  Supplementary Figure S3. 2D diagrams of ligand interactions. Dotted boundary indicates the proximity contour of the ligand. Green and pink circles represent hydrophobic and polar residues within 4.5 Å cut-off distance, respectively. Blue dots represent exposed atoms in ligands. Interactions with main chain atoms or side chain atoms of core protein or water are shown in blue or green or brown dotted lines, respectively. Arrowheads point towards the hydrogen bond acceptor. (A) HAP_R01-core protein interaction. (B) SBA_R01-core protein interaction.  Figure S9. Antiviral activity of HAP or SBA against HBV core mutants determined in the HepG2 HBV transient transfection assay. Mean dose response curves from three independent experiments were plotted for selected mutants. Percent inhibition values for drug treated samples were normalized to values from HepG2 cells only treated with DMSO.
Supplementary Figure S10. Replication capacity of HBV core mutants compared to wild type was determined in the HepG2 HBV transient transfection assay based on DNA copies. Shown are mean values ± standard deviation (SD) from three independent experiments.

Supplementary Figure S11.
Supplementary Figure S11. Docking of HAP_R01 and its corresponding S-diastereomer HAP_R02 into the novel type 2 contact-to-spike interface. Light green represents the dimer in asymmetric unit. The symmetry mate is colored in dark green. Y88 is shown in space-filling model colored in magenta. Compounds are colored in cyan. Only the favorite docking poses are shown.
Supplementary Figure S12. Local conformational changes caused by HAP_R01 during energy minimization. Chain C and chain D' of icosahedral T4 capsid was extracted from 1QGT. The ligands and interface residues within 8 Å radius were regarded as flexible during energy minimization. (A) Light green and dark green ribbons represent chain C and chain D' from 1QGT, respectively. After forcefield minimization, HAP_R01 is highlighted in cyan stick and HBV core protein is shown in magenta. (B) Light grey and dark grey represent chain C and chain D' from 1QGT, respectively. After forcefield minimization, SBA_R01 is coloured in pink and HBV core protein is also shown in magenta. The red arrows indicate that only HAP_R01 can induce significant conformational changes on helix 5 of chain D' (S121-P129).
Supplementary Figure S13. The full-length blots for the DNA and native gels. The full length DNA gel (top row) and native capsid gel (middle row) were transfered to the whole membrane. Multiple images were taken under different exposures. The denatured gel (bottom row) electrophoresis was carried out to detect core protein and actin. After proteins were electrotransferred to a nitrocellulose membrane, the membrane was cut around the 30 KDa protein marker band. The upper parts of the membrane were used for β-actin (42 KDa) blotting and the lower parts were used for HBV core protein (~21 KDa) detection. After blotting, images were visualized by ChemiDoc TM imaging System under different exposures. Red boxes indicate the tested compounds HAP_R01 and SBA_R01 in this paper. Other unrelated compounds are not disclosed in this work.

Protein expression and purification
The recombinant protein was expressed in Escherichia coli strain BL21(DE3). IPTG was added to a final concentration of 0.1mM when OD600 reached 0.9. The culture was grown at 16 ˚C overnight. Cells were harvested by centrifugation at 8000 g for 10 min. About 14 g cells were lysed by ultra-sonication in190 ml lysis buffer (50 mM Tris pH 9.0, 2 M urea, 250 mM NaCl, 1 mM DTT). After centrifugation at 15000 g for 30 min, the supernatant was applied to Ni affinity column (Chelating Sepharose Fast Flow, 20mL, GE healthcare).
Tagged protein was eluted with lysis buffer containing 20-50 mM imidazole. The pooled fractions were dialyzed against 25 mM Tris pH 9.0, 2 mM DTT. The protein was further purified by anion-exchange chromatography (HiTrap Q HP, 20 ml, GE healthcare) with linear gradient of 0 to 1000 mM NaCl. The hexahistidine tag was removed using tobacco etch virus (TEV) protease digestion over night at 30 ℃ (500 U/mg). The cleaved sample went through a second affinity column and flow-through containing tagless protein (with additional C-terminal ENLYFQ) was collected. The protein was dialyzed into 25 mM Tris buffer pH 9.0 containing 2 mM DTT and 10 mM EDTA. Finally the product was concentrated to 17 mg/ml and stored at -

80˚C.
Capsid assembly fluorescence quenching assay 20 fold molar excess of maleimidyl BoDIPY-FL dye (Sigma) was added to Cp150 dimer overnight at 4°C.
DMSO and unreacted dye were removed by using G25 column in ice-cold 50 mM HEPES pH7.5 buffer. The modified protein is referred to as C150Bo. The concentration of bound dye and labeled protein were determined by formula as follow: [ and incubated in 37 ˚C, 5% CO 2 overnight. The next day, half-log series dilutions of testing compounds were prepared using medium and added 100 μL compounds into appropriate wells. For 3 days incubation, the media was replaced with fresh media (with compounds), then continued to incubate for two days. The cytotoxicity was measured using CCK-8 kit (Dongjindo)

Labeling of capsid with photoreactive HAP compounds
5 μM core protein dimer and 20 μM HAP compound were mixed in a buffer containing 50 mM HEPES, 250 mM NaCl, pH 7.4 to prepare sample with the dimer : probe ratio of 1:4. 1:2 and 1:10 ratios were also tested for the PL1s to determine the optimal labeling condition. Due to similar labeling efficiency with three tested concentrations, the dimer : probe molar ratio of 1:4 was adopted for PL2s and PL3s. The solution was then irradiated for 30 minutes with light from a mercury lamp (medium pressure, 500 W). The carbene group % assembly = 100 × F 25% assembly -F sample F 25%assembly -F full-assembly % assembly = 100 × F 25% assembly -F sample F 25%assembly -F full-assembly generated by UV light exposure is supposed to covalently modify amino acid residue(s) in its close vicinity.

Protease digestion
The irradiated sample was digested with four proteases, i.e. trypsin, chymotrypsin, proteinase K and pepsin, respectively. Briefly, about 20 μL of sample was mixed with 100 ul of 8 M urea in 50 mM ammonium bicarbonate buffer (ABC) in an Amicon® Ultra-0.5 centrifugal filter (10K Mw cutoff, Millipore). The preparation was centrifuged at 14000 g for 15 min. Then 200 μL of 50 mM ABC (0.06 M HCl was added for pepsin digestion) was added and centrifuged again. Proteases were added according to vendor's instruction. The preparation was incubated at 37°C overnight. The resultant peptide mixture was collected to a new tube by centrifugation with 100 μL 50% acetonitrile in 50 mM ABC and 100 μL of 50% acetonitrile in 50% formic acid, respectively. The reaction mixture was dried in a speed vac and reconstituted with 30 μL of 5% DMSO/5% formic acid/5% acetonitrile.

LC/MS
3 ul of irradiated sample was injected onto a reverse phase C4 column (BEH300, 150×2.1 mm, 1.7 μm, Waters) equilibrated at 75°C at a flow rate of 250 μL/min. Reverse phase chromatography was performed using an ultraperformance liquid chromatography (UPLC) system (Acquity UPLC, Waters). The gradient was generated by using 0.1% formic acid for mobile phase A and acetonitrile containing 0.1% FA for mobile phase B. After an isocratic elution at 20% B for 2 min, B was raised to 60% in 5 min and to 85% in an additional 1 min. The column was then washed for 2 min at 85% B and re-equilibrated for 5 min at 20% B, giving an overall run time of 15 min. The eluted species were then analyzed online by a Q-TOF mass spectrometer (TripleTOF™ 5600, AB Sciex) operating in the positive ion mode from m/z 400-3000 and calibrated according to the manufacturer's procedure.

LC/MS/MS
An aliquot of the digestion representing an amount of 1 μg core protein was injected onto a reverse phase C18 column (BEH300, 150×2.1 mm, 1.7 um, Waters) equilibrated at 35°C at a flow rate of 160 μL/min. The gradient was generated by using 0.1% formic acid for mobile phase A and acetonitrile containing 0.1% FA for mobile phase B. After an isocratic elution at 5% B for 2 min, B was raised to 45% in 29 min and to 90% in an additional 5 min. The column was then washed for 4 min at 90% B and re-equilibrated for 10 min at 5% B, giving an overall run time of 50 min. LC/MS/MS was performed in an information dependent mode (IDA) using the same system as above. For each cycle, one full MS scan was followed by up to 10 MS/MS for the most intense ions.

Energy minimization.
In capsid crystal structures in complex with HAP1 and AT-130, the C-D' interface is the shared ligand binding site (Fig. 5C) 1,2 . Since the compound binding pockets in Y132A hexamer structures are dominated by the concave from chain B, we simply superimposed the chain B of the Y132A-compound structures onto the chain C of T4 capsid 1QGT to generate the initial positions of the ligand. During energy minimization, ligands and residues within 8 Å or 20 Å radius were kept free of restraints. The local minimum of the molecular energy function was calculated using Amber12EHT forcefield till RMS gradient fell below 0.1 kcal/mol/Å 2 .

Rational design and synthesis of photoreactive HAPs
Diazirine has been widely used in photolabeling studies due to its small size and good stability 3 . HAP_R01 has a difluoroproline moiety at its 6-position, and the gem-difluoro group is critical to its high activities in both biochemical and cellular assays. Given the fact that diazirine and difluoromethyl have similar size and lipophilicity, we considered these two groups as bioisosteres, and prepared HAP_R01_PL1 where the original difluoromethyle was replaced by photoreactive group diazirine. To our delight, compound 1 did exhibit good activity in the biochemical assay and excellent anti-HBV activity in the cellular assay (Tab. S2). Since short alkyl esters were found tolerated at the 5th position of HAP, a butyl chain bearing a diazirine group was incorporated into that position to give HAP_R01_PL2. Phenyl ring at the 4th position of HAP is indispensable for its capsid targeting activity, and the small sized and linear substituent, cyano group, at the para-position of the phenyl ring was well tolerated in terms of potency. Due to the similarity of azido and cyano groups, azido group was envisioned as the best choice for photoreactive group at this direction. In addition, fluorine at the meta-position was beneficial to capsid assembly activity. Thus the third probe, HAP_R01_PL3, was prepared with a 4-azido-3-fluoro-2-chlorophenyl fragment at the 4th position of the molecule. The synthesis of all these probes was based on a three component reaction between an acetoacetate derived ester, a substituted benzaldehyde and thiazole-2-carboxamidine hydrochloride .
The residue was partitioned between ethyl acetate and ice-water, and neutralized to pH ~4 with 1 N sodium hydroxide solution. The organic layer was separated, and the aqueous phase was extracted with ethyl acetate three times. The combined organic phase was dried over Na 2 SO 4 , filtrated and concentrated to give 2.4 g of a crude product (LCMS (M+H + ) 352). Thus obtained crude material (505 mg, 1.4 mmol) was dissolved in N,Ndimethylmethanamide (2 mL), to it were added 2-(3-methyldiazirin-3-yl)ethanol (300 mg, 3.0 mmol), 1-ethyl-3-