Design and Synthesis of New GS-6207 Subtypes for Targeting HIV-1 Capsid Protein

HIV-1 capsid protein (CA) is the molecular target of the recently FDA-approved long acting injectable (LAI) drug lenacapavir (GS-6207). The quick emergence of CA mutations resistant to GS-6207 necessitates the design and synthesis of novel sub-chemotypes. We have conducted the structure-based design of two new sub-chemotypes combining the scaffold of GS-6207 and the N-terminal cap of PF74 analogs, the other important CA-targeting chemotype. The design was validated via induced-fit molecular docking. More importantly, we have worked out a general synthetic route to allow the modular synthesis of novel GS-6207 subtypes. Significantly, the desired stereochemistry of the skeleton C2 was confirmed via an X-ray crystal structure of the key synthetic intermediate 22a. Although the newly synthesized analogs did not show significant potency, our efforts herein will facilitate the future design and synthesis of novel subtypes with improved potency.


Design of Novel PF74/GS-6207 Molecular Hybrids 5 and 6
The design of the molecular hybrids is based on the shared binding mode of 1, PF74, and 2, GS-6207 (Figure 1B).Although vastly different in molecular complexity and functional group density, the two molecules bind to the same pocket at the interface of two adjacent CA promoters (CA1 and CA2).2, GS-6207, is well superimposed with 1, PF74, through the backbone and the R 2 and R 3 moieties, with shared key molecular interactions in the N-terminal domain of CA1 (CA1 NTD ) and the C-terminal domain of the adjacent CA2 (CA2 CTD ) (Figure 1B).The additional interactions conferred by 2, GS-6207, with CA2 NTD via the sulfone group of the R 4 moiety and CA2 CTD via the methanesulfonamide group of the R 1 moiety likely contribute substantially to its superior potency over 1, PF74, and will be retained in our design.Interestingly, from our previous chemical profiling, 1, PF74 analog 3 showed a drastically improved potency (>27-fold) and superb capsid stabilizing effect, as indicated by a large positive shift in the thermal shift assay (TSA) [6].Another PF74 analog 4, featuring a distinct indolone R 3 moiety, displayed a highly unusual destabilizing effect (negative shift in TSA) [6].An important aim of our redesign is to optimize both the stabilizer lead (3) and the destabilizer lead (4) through design and synthetic strategies aligned with improving the resistance profile of 2, GS-6207.Based on these, molecular hybrids 5 and 6 were designed, as shown in Figure 1C.To enhance the synthetic accessibility, the design also features a slightly simplified R 1 moiety for both 5 and 6 and undefined stereochemistry in the indolone ring of hybrid 6.Finally, the newly designed hybrids will contain the R 2 moiety of both 1, PF74 (R = H), and 2, GS-6270 (R = F).The main aim of this pilot design and synthesis is to develop a general and amenable synthesis to enable the future redesign and synthesis of structurally more elaborate subtypes of 2, GS-6207.

Synthesis of Hybrids 5 and 6
The general synthetic approach is depicted in Figure 1D.The synthesis is highly modular based on four synthetic components (C 1 -C 4 ) for installing R 1 -R 4 .C 2 is the core with the proper functional group handles to allow the installation of R 1 , R 4 , and R 3 via sequential reactions with C 1 (Suzuki), C 4 (Sonogashira), and C 3 (amide coupling), respectively.This modular synthesis will support future synthetic needs of structural diversification in all four structure-activity relationship (SAR) regions (R 1 -R 4 ), particularly the regions of R 1 , R 4 , and R 3 .
The synthetic route for the preparation of component C 1 is shown in Scheme 1. MnO 2 oxidation [38] of commercially available benzyl alcohol 11 afforded benzaldehyde intermediate 12.The subsequent conversion to nitrile 13 was effected via the standard method of oxime formation and dehydration [39].The nucleophilic aromatic substitution reaction [40] at the F site by hydrazine followed by the cyclization reaction produced aminoindazole 14, which was subjected to the palladium-catalyzed borylation [41] with bis(catecholato)diboron to yield the representative C 1 , compound 7.

Synthesis of Hybrids 5 and 6
The general synthetic approach is depicted in Figure 1D.The synthesis is highly modular based on four synthetic components (C 1 -C 4 ) for installing R 1 -R 4 .C 2 is the core with the proper functional group handles to allow the installation of R 1 , R 4 , and R 3 via sequential reactions with C 1 (Suzuki), C 4 (Sonogashira), and C 3 (amide coupling), respectively.This modular synthesis will support future synthetic needs of structural diversification in all four structure-activity relationship (SAR) regions (R 1 -R 4 ), particularly the regions of R 1 , R 4 , and R 3 .
The synthetic route for the preparation of component C 1 is shown in Scheme 1. MnO2 oxidation [38] of commercially available benzyl alcohol 11 afforded benzaldehyde intermediate 12.The subsequent conversion to nitrile 13 was effected via the standard method of oxime formation and dehydration [39].The nucleophilic aromatic substitution reaction [40] at the F site by hydrazine followed by the cyclization reaction produced aminoindazole 14, which was subjected to the palladium-catalyzed borylation [41] with bis(catecholato)diboron to yield the representative C 1 , compound 7.The synthesis of the core component C 2 began with the commercially available 2,5dibromopyridine 15 (Scheme 2).Deprotonative formylation [42] of 15 with TMPMgCl.LiCl in dry THF followed by the addition of DMF afforded aldehyde intermediate 16, which set the stage for the key asymmetric induction via a chiral auxiliary.The chiral auxiliary was introduced via condensation of aldehyde 16 with chiral non-racemic (S)-and (R)-tert-butanesulfinamides to produce (S)-tert-butanesulfinyl imine 17 and (R)tert-butanesulfinyl imine 18, respectively (Scheme 2) [43].The synthesis of the core component C 2 began with the commercially available 2,5dibromopyridine 15 (Scheme 2).Deprotonative formylation [42] of 15 with TMPMgCl.LiCl in dry THF followed by the addition of DMF afforded aldehyde intermediate 16, which set the stage for the key asymmetric induction via a chiral auxiliary.The chiral auxiliary was introduced via condensation of aldehyde 16 with chiral non-racemic (S)-and (R)-tertbutanesulfinamides to produce (S)-tert-butanesulfinyl imine 17 and (R)-tert-butanesulfinyl imine 18, respectively (Scheme 2) [43].The synthesis of the core component C 2 began with the commercially available 2,5dibromopyridine 15 (Scheme 2).Deprotonative formylation [42] of 15 with TMPMgCl.LiCl in dry THF followed by the addition of DMF afforded aldehyde intermediate 16, which set the stage for the key asymmetric induction via a chiral auxiliary.The chiral auxiliary was introduced via condensation of aldehyde 16 with chiral non-racemic (S)-and (R)-tert-butanesulfinamides to produce (S)-tert-butanesulfinyl imine 17 and (R)tert-butanesulfinyl imine 18, respectively (Scheme 2) [43].Both auxiliaries were used to determine the preferred auxiliary/nucleophile combination for producing the desired stereochemical outcome (S) for C 2 (8a and 8b).Specifically, in four different reactions shown in the table (Scheme 2), commercially available benzyl magnesium chloride 24 or (3,5-difluorobenzyl)zinc bromide 25 was used as the Both auxiliaries were used to determine the preferred auxiliary/nucleophile combination for producing the desired stereochemical outcome (S) for C 2 (8a and 8b).Specifically, in four different reactions shown in the table (Scheme 2), commercially available benzyl magnesium chloride 24 or (3,5-difluorobenzyl)zinc bromide 25 was used as the nucleophile to react with both sulfinylimines 17 and 18 to afford four pairs of diastereomers (19a/20a, 19b/20b, 21a/22a, and 21b/22b) in ratios of 1-2 favoring the undesired diastereomers.From these experiments, it was clear that (R)-tert-butanesulfinyl imine 18 is preferred over the (S)-enantiomer 17 for inducing the desired (S) stereochemistry in C 2 (8a and 8b).The structure and absolute stereochemistry of intermediate 22a were confirmed by single crystal X-ray diffraction analysis (Figure 3).The crystal selected for the study has chirality at C1 S. The chains of hydrogen bonds are parallel to the a-axis through the . ..O-S-N-H. . .fragment (Figure 3).Deprotection of the sulfinamide under HCl yielded intermediates 23a,b, which were Boc-protected to afford compounds 8a,b as the core C 2 .and 8b).The structure and absolute stereochemistry of intermediate 22a were confirm by single crystal X-ray diffraction analysis (Figure 3).The crystal selected for the stu has chirality at C1 S. The chains of hydrogen bonds are parallel to the a-axis through t …O-S-N-H… fragment (Figure 3).Deprotection of the sulfinamide under HCl yielded termediates 23a,b, which were Boc-protected to afford compounds 8a,b as the core C 2 .omers (19a/20a, 19b/20b, 21a/22a, and 21b/22b) in ratios of 1˗2 favoring the undesired diastereomers.From these experiments, it was clear that (R)-tert-butanesulfinyl imine 18 is preferred over the (S)-enantiomer 17 for inducing the desired (S) stereochemistry in C 2 (8a and 8b).The structure and absolute stereochemistry of intermediate 22a were confirmed by single crystal X-ray diffraction analysis (Figure 3).The crystal selected for the study has chirality at C1 S. The chains of hydrogen bonds are parallel to the a-axis through the …O-S-N-H… fragment (Figure 3).Deprotection of the sulfinamide under HCl yielded intermediates 23a,b, which were Boc-protected to afford compounds 8a,b as the core C 2 .Finally, the installation of component C 3 was achieved under standard peptide coupling conditions with HATU as the coupling agent [47], followed by the removal of one mesyl group.As such, commercially available acids 9a,b were incorporated into final compounds 5a,b and 6a,b.The final compounds, 6a,b, were produced as a mixture of diastereomers, which were separated by silica gel column chromatography to afford the desired diastereomer.Following through the purification process, the final compounds 5a,b and 6a,b were successfully crystallized from isopropanol with higher purity.
agent [47], followed by the removal of one mesyl group.As such, commercially available acids 9a,b were incorporated into final compounds 5a,b and 6a,b.The final compounds, 6a,b, were produced as a mixture of diastereomers, which were separated by silica gel column chromatography to afford the desired diastereomer.Following through the purification process, the final compounds 5a,b and 6a,b were successfully crystallized from isopropanol with higher purity.

Biological Analysis of Select Compounds
We tested compounds 5a, 5b, 6a, and 6b (using PF74 as a control) for their effect on the stability of covalently crosslinked HIV capsid (CA) hexamer.Of these compounds, 5b demonstrated a positive shift in the melting temperature of the CA hexamer, indicating some stabilization (Table 1).The other three compounds did not provide any stabilization of the CA hexamer.We further tested these four compounds in cell-based antiviral assays for the inhibition of HIV virus activity.Significant toxicity was visibly observed during the antiviral testing, suggesting that the compounds are cytotoxic.This made the EC50s

Biological Analysis of Select Compounds
We tested compounds 5a, 5b, 6a, and 6b (using PF74 as a control) for their effect on the stability of covalently crosslinked HIV capsid (CA) hexamer.Of these compounds, 5b demonstrated a positive shift in the melting temperature of the CA hexamer, indicating some stabilization (Table 1).The other three compounds did not provide any stabilization of the CA hexamer.We further tested these four compounds in cell-based antiviral assays for the inhibition of HIV virus activity.Significant toxicity was visibly observed during the antiviral testing, suggesting that the compounds are cytotoxic.This made the EC 50 s values difficult to determine reliably (Table 1).Overall, the compounds did not exhibit much biological activity.

Procedure for Synthesis of 12
To a solution of commercially available (3-bromo-2-fluorophenyl)methanol (11, 40 g, 1.0 equiv.) in 92 mL of DCM, was added MnO 2 (40 g, 1.05 equiv.)slowly under argon.The reaction mixture was stirred for 8 h at room temperature under an argon balloon.Upon completion, as confirmed by TLC, the reaction mixture was filtered through a pad of celite.The reaction mixture was washed by DCM five times.The combined organic layers were further washed with water and brine, dried over Na 2 SO 4 , and concentrated in vacuo to afford crude intermediate 12, 3-bromo-2-fluorobenzaldehyde as yellow solid (12, 4.34 g, 87%), which was directly used for the next step without further purification.Yield: 87%. 1

Procedure for Synthesis of 21
To a solution of MeSO 2 Na (26, 6 g, 29.26 mmol, 1.0 equiv.) in DMF (100 mL) was added copper(I) chloride (0.48 g, 2.4 mmol) slowly.The reaction mixture was heated to 40 • C and maintained at that temperature for 18 h.Upon completion, as confirmed by TLC, the reaction mixture was cooled to room temperature and concentrated under reduced pressure.The resulting residue was diluted with water and extracted with EtOAc to remove unreacted starting material.The aqueous layer was acidified with 0.5 M citric acid and followed by 1 N NaOH.The combined organic layers were washed with water, brine (1.0 L); dried over Na 2 SO 4 ; filtered and then concentrated in vacuo.The crude product was purified by Combi-flash on silica get using 0-5% Hexane/EtOAc to afford 3-methyl-3-(methylsulfonyl)but-1-yne (21, 2.92 g, 41%). 1 H NMR (600 MHz, CDCl 3 ) δ 3.04 (s, 3H), 2.58 (d, J = 1.5 Hz, 1H), 1.67 (s, 6H). 13

General Procedure for Synthesis of 5a,b and 6a,b
To a solution of desired acid (0.45 g, 14.7 mmol, 1.1 equiv.) in 2 mL of DMF was added HATU (0.51 g, 0.13 mmol, 1.05 equiv.) at 0 • C, and the reaction mixture was stirred for 30 min.Followed by addition of a solution of amine intermediate, (S)-N- methanesulfonamide (30b, 0.30 g, 1.0 equiv.) in 1 mL DMF and DIPEA (0.51 mL, 32.71 mmol, 3.0 equiv.).The reaction mixture was then slowly warmed to room temperature and stirred for 12 h.To the reaction mixture was added a solution of ammonia in MeOH (2 M, 10 mL), and the mixture was then stirred for 10 min.Upon completion, as confirmed by TLC, the reaction mixture was concentrated, diluted with water, and extracted with ethyl acetate and washed with aq 1 M HCl solution.The combined organic layer was further washed with water and brine, dried over Na 2 SO 4 , and concentrated.The crude product was purified by Combi-flash on silica gel using 20-100%

Modeling and Docking Analysis
Molecular modeling was performed using the Schrödinger small molecule drug discovery suite 2021-3 (Schrödinger Inc., New York, NY, USA) [48].We analyzed PF74-bound full length native HIV-1 CA (PDB ID: 4XFZ) by using Maestro (Schrödinger Inc.) [49].Standard docking protocols were following protein preparation, grid generation, ligand preparation, and molecular docking.Protein preparation conducted by using the Protein preparation wizard (Schrödinger Inc.) [50] and involved the refinement of the protein structure.By using prime, the missing hydrogen atoms, side chains, and loops were refined into the protein.The OPLS3e force field was used to minimize the hydrogen bonding network and readjusting the heavy atoms to a rmsd of 0.3 Å [51].The receptor grid generation tool in Maestro (Schrödinger Inc.) was utilized to standardize the binding site around the native ligand, surrounding all the key residues within the range of 12 Å.After sketching ligands in Maestro 2D Sketch tab, different conformers were generated in LigPrep [52] at pH of 7 ± 2 to serve as initial step for docking process.Finally, docking was conducted using the Glide XP (Glide, version 8.2, New York, NY, USA) [53] with a command as the van der Waals radii of nonpolar atoms for each of the ligands fixed by a factor of 0.8.After docking refinement and minimization, protein flexibility was also regarded under implicit solvent.All docked poses were subjected to analysis to cut off a small number of poses within the field of the receptor and binding pocket to generate better desired poses.Each docked pose was furtherance and presented in publication format using PyMOL (SchrodingerLLC) [54].The numbering of residues of HIV-1 CA used in this paper for description was based on the full-length native HIV-1 CA.

Figure 3 .
Figure 3. X-ray crystal structure of 22a.For this pilot design and synthesis, we used commercially available acids 9a,b as component C 3 .The component C 4 (compound 10) was prepared in a single step as described in Scheme 3. S-alkylation of sodium methanesulfinate[44] (MeSO 2 Na) with commercially available chloride 26 yielded compound 10 (Scheme 3).

Figure 3 .
Figure 3. X-ray crystal structure of 22a.For this pilot design and synthesis, we used commercially available acids 9a,b as component C3 .The component C 4 (compound 10) was prepared in a single step as described in Scheme 3. S-alkylation of sodium methanesulfinate[44] (MeSO2Na) with commercially available chloride 26 yielded compound 10 (Scheme 3).

a 3 . 3 .
Scheme 3. Synthesis of C 4 (compound 10).Reagents and conditions: (a) MeSO2Na, Cu(I)Cl, DMF, 60 °C, 18 h, 41%.With all four components in hand, the overall modular synthesis was carried out based on the core component C 2 (compounds 8a,b), as depicted in Scheme 4. The synthesis started with the Sonogashira coupling [45] of component C 2 (compounds 8a,b) with component C 4 (compound 10) to produce intermediates 27a,b.The subsequent Suzuki coupling [46] of 27a,b with component C 1 (compounds 7) under the catalysis of Pd(dppf)2Cl2 afforded intermediates 28a,b.Protection of the free NH2 group in component C 1 with mesylchloride gave bismesylated intermediates 29a,b, which upon Boc deprotection under TFA produced advanced intermediates 30a,b.Finally, the installation of component C 3 was achieved under standard peptide coupling conditions with HATU as the coupling Scheme 3. Synthesis of C 4 (compound 10).Reagents and conditions: (a) MeSO 2 Na, Cu(I)Cl, DMF, 60 • C, 18 h, 41%.With all four components in hand, the overall modular synthesis was carried out based on the core component C 2 (compounds 8a,b), as depicted in Scheme 4. The synthesis started with the Sonogashira coupling [45] of component C 2 (compounds 8a,b) with component C 4 (compound 10) to produce intermediates 27a,b.The subsequent Suzuki coupling [46] of 27a,b with component C 1 (compounds 7) under the catalysis of Pd(dppf) 2 Cl 2 afforded intermediates 28a,b.Protection of the free NH 2 group in component C 1 with mesylchloride gave bismesylated intermediates 29a,b, which upon Boc deprotection under TFA produced advanced intermediates 30a,b.Finally, the installation of component C 3 was achieved under standard peptide coupling conditions with HATU as the coupling agent[47], followed by the removal of one mesyl group.As such, commercially available acids 9a,b were incorporated into final compounds 5a,b and 6a,b.The final compounds, 6a,b, were produced as a mixture of diastereomers, which were separated by silica gel column chromatography to afford the desired diastereomer.Following through the purification process, the final compounds 5a,b and 6a,b were successfully crystallized from isopropanol with higher purity.

Table 1 .
Thermal shift and cell-based antiviral analysis of selected compounds.: melting point change of CA hexamer in presence of compound compared to DMSO control.Mean ± standard deviation (SD) from at least two independent experiments.b Half maximal effective concentration from at least two independent experiments.c Cytotoxic effects from compounds were visible during EC 50 determination; the results may be unreliable.
• C, followed by centrifugation for 40 min at 3500 rpm.The resulting viral-containing pellet was concentrated 10-fold by resuspension in DMEM without FBS and stored at −80 • C.