Design, synthesis, and mechanism study of dimerized phenylalanine derivatives as novel HIV-1 capsid inhibitors

Abstract

HIV-1 capsid (CA) plays indispensable and multiple roles in the life cycle of HIV-1, become an attractive target in antiviral therapy. Herein, we report the design, synthesis, and mechanism study of a novel series of dimerized phenylalanine derivatives as HIV-1 capsid inhibitors using 2-piperazineone or 2,5-piperazinedione as a linker. The structure-activity relationship (SAR) indicated that dimerized phenylalanines were more potent than monomers of the same chemotype. Further, the inclusion of fluorine substituted phenylalanine and methoxyl substituted aniline was found to be beneficial for antiviral activity. From the synthesized series, Q-c4 was found to be the most potent compound with an EC₅₀ value of 0.57 μM, comparable to PF74. Interestingly, Q-c4 demonstrated a slightly higher affinity to the CA monomer than the CA hexamer, commensurate with its more significant effect in the late-stage of the HIV-1 lifecycle. Competitive SPR experiments with peptides from CPSF6 and NUP153 revealed that Q-c4 binds to the interprotomer pocket of hexameric CA as designed. Single-round infection assays showed that Q-c4 interferes with the HIV-1 life cycle in a dual-stage manner, affecting both pre-and post-integration. Stability assays in human plasma and human liver microsomes indicated that although Q-c4 has improved stability over PF74, this kind of inhibitor still requires further optimization. And the results of the online molinspiration software predicted that Q-c4 has desirable physicochemical properties but some properties still have some violation from the Lipinski rule of five. Overall, the dimerized phenylalanines are promising novel platforms for developing future HIV-1 CA inhibitors with considerable potential for optimization.

1. Introduction

Acquired immunodeficiency syndrome (AIDS) is mainly caused by the pathogen of human immunodeficiency virus type 1 (HIV-1), a major cause of death globally, threatening human health [1]. HIV mainly attacks the human immune system, infects CD4⁺ T lymphocytes, and reduces cell-mediated immune response, resulting in increased individual susceptibility to infection and disease. By the end of 2019, 38 million people worldwide were infected with AIDS, including 1.8 million children under 15, and the annual death caused by AIDS was 0.69 million [2]. Combination antiretroviral therapy (cART) is the combination of several (usually three or four) antiretroviral drugs in the treatment of retroviral infection [3]. The application of cART can significantly reduce the viral load of patients or HIV carriers, improve the quality of life, and prolong the life expectancy of patients. Nevertheless, cART has shown mentionable disadvantages, such as drug resistance, neurotoxicity, and liver failure [4]. Therefore, there is still an urgent need (but equally challenging) to identify new targets for antiretroviral therapy and new HIV drug discovery with new and as yet unexplored mechanisms of action to achieve HIV infection cure [4-10].

HIV-1 capsid has drawn lots of attention for its critical roles in the viral life cycle. It comprises about 1500 copies of CA monomers forming 12 pentamers and about 250 hexamers (Fig. 1) [11-16]. The CA monomer is divided into the N-terminal domain (NTD, residues 1–145) and C-terminal domain (CTD, residues 150–231) connected by a flexible linker (residues 146–149). The pentamers and hexamers are assembled by NTD-CTD interaction and NTD-NTD interaction, while CTD-CTD interaction drives the formation of the complete CA core [17,18].

Fig. 1.

The monomers (PDB ID: 6WAP) are assembled into hexamers (PDB ID: 4XFX) and pentamers (PDB ID: 5MCY) and finally assembled into capsid (PDB ID: 3J3Q). The CTD is yellow, while NTD in monomer, hexamer, and capsid are blue and NTD in pentamer is red. The flexible linker in the monomer is highlighted in pink. The Figures are generated in PyMOL (www.pymol.org).

CA interacts with several cellular factors, including NUP153, CPSF6, CypA, Sec24C, and TRIM5α [19-24]. These interactions are essential to facilitate the escape of the virus from innate immune surveillance and the entry of virus pre-integration complexes into the nucleus. The CA-CPSF6 interaction promotes the entrance of the pre-integration complex into the nucleus, regulates nuclear localization, and contributes to integrating viral reverse transcripted DNA into the host gene transcriptionally active regions [25,26]. The CA-NUP153 is essential for nuclear transport. The complete structure, assembly kinetics, and stability of the capsid are related to the normal replication and infection of HIV-1. Interfering the normal disassembly in the early stage or assembly of CA in the late stage has been demonstrated to block the normal life cycle of HIV-1, thus leading to the loss of infectivity. Therefore, HIV-1 CA has become a promising new target in the field of anti-HIV-1 drugs [11,27-29].

Several inhibitors binding to different sites of HIV-1 CA have been reported, as exemplified by CAP-1, BM1, BD1, BI-1, CAI, PF74, 11l and GS6207 (Fig. 2) [14,30-37]. These compounds can perturb the stability of CA and show early-, late-, or dual-stage anti-HIV activity. PF74 is the most studied CA inhibitor binding to the NTD-CTD interface with micromolar antiviral activity. However, poor metabolic stability [38] and low antiviral activity hinder its clinical application. GS-6207, a PF74 derivative reported by Gilead Sciences, has increased the affinity to CA by adding multiple hydrogen bond donors and acceptors. The additional contacts present in GS-6207 tightly bind two adjacent capsid subunits and promote distal intra- and inter-hexamer interactions that stabilize the capsid [39,40]. Meanwhile, increased molecular volume bars the contact between adjacent CA subunits and then blocks the assembly of CA. At present, GS-6207 has been in phase II/III [41]. Previous reports showed that GS-6207 has strong metabolic stability as a long-acting inhibitor for subcutaneous dosing every three months or longer. However, drug resistance [42] and the arduous synthesis route provide rationale for discovering additional HIV-1 CA inhibitors [43-47].

Fig. 2.

Structures of representative HIV-1 CA inhibitors. Skeletons of PF74, 11l and GS-6207 were shown in magenta.

The co-crystallization of PF74 and GS-6207 with CA hexamers shows similar binding modes (Fig. 3). The two compounds form hydrogen bonds with Asn57, while GS-6207 forms extra polar contacts with Ser41 and Asn74, favouring the binding. Fluorine and more alkyl groups in GS-6207 improved the drug-like and lipophilic properties and enhanced the hydrophobic interaction with nearby residues such as Met66 and Ile73. Despite the bulky size of GS-6207, a substantial portion of the NTD-CTD is still not fully occupied and provides further potential for inhibitor design. Therefore, a novel compound with a large molecular volume and more hydrogen bond receptors and donors may improve antiviral activity. Despite cation-pi interactions with Arg173 and Lys70 indole of PF74, hydrogen bonds between the two residues and 2-piperazinone of 11l have a more significant effect on enhancing the antiviral activity. Based on the above analysis, we maintained the phenylalanine skeleton of PF74 and introduced a 2-piperazineone or 2,5-piperazinedione as a linker and the fluorine substituted phenylalanine as a privileged structure to obtain a series of dimerized phenylalanine derivatives (DPAs) to occupy the larger area in the NTD-CTD interface.

Fig. 3.

Binding modes of (A) PF74 (yellow, PDB ID: 5HGL) and (B) GS-6207 (pink, PDB ID: 6V2F). The CTD is shown in gray, and the NTD of the adjacent monomer is in green. (C) Design of DPAs as HIV-1 CA inhibitors. The Figures are generated in PyMOL (www.pymol.org).

Herein, we report the design, synthesis, and mechanism study of DPAs as novel HIV-1 capsid inhibitors. All synthesized compounds were tested for antiviral activity in MT-4 cells infected by HIV–1III_B or HIV ROD and subjected to structure-activity relationship analysis. Additionally, we performed surface plasmon resonance (SPR) experiments, action stage determination assays, and molecular dynamics (MD) simulations to provide insight into the mechanism of action of this series of compounds. The metabolic stability was determined in human plasma and human liver microsomes. Lastly, the physicochemical properties of Q-a2 and Q-c4 were predicted by Molinspiration Cheminformatics free web services (https://www.molinspiration.com, Slovensky Grob, Slovakia).

2. Chemistry

As shown in Scheme 1, starting from commercially available 4-methoxy-N-methylaniline or N-methylaniline (1), the target compounds were prepared via a concise and well-established synthetic route as outlined below. Treating of 1 with 4-methoxy-N-methylaniline or N-methylaniline and benzotriazol-1-yl-oxytripyrrolidinophosphonium hexafluorophosphate (PyBop) in N,N-diisopropylethylamine (DIEA) and dichloromethane afforded 2, followed by removal of tert-butyloxycarbonyl (Boc) protection resulted in the formation of free amine 3. The key intermediate 4 was obtained by acylation of 3 with bromoacetic acid in dichloromethane solution. The nucleophilic substitution of 2,5-piperazinedione with 4 in DMF results in the final products Q-a1-a3.

Scheme 1.

Preparation of Q-a1-a3.

Reagents and conditions:

(i) Boc-l-phenylalanine or Boc-3,5-difluoro-l-phenylalanine, PyBop, DIEA, dichloromethane, 0 °C to r.t.; (ii) trifluoroacetic acid, dichloromethane, r.t.; (iii) bromoacetic acid, HATU, DIEA, dichloromethane, 0 °C to r.t.; (iv) 2,5-piperazinedione, K₂CO₃, DMF, 55 °C.

As shown in Scheme 2, Q-b1-b3 and Q-c1-c9 were initiated from intermediate 4. Intermediate 4 reacted with 1-Boc-3-oxopiperazine by nucleophilic substitution to achieve 5, then removal of Boc group resulted in Q-b1-b3. The nucleophilic substitution of Q-b1-b3 with 4 in DMF results in the final products Q-c1-c9.

Scheme 2.

Preparation of Q-b1-b3 and Q-c1-c9.

Reagents and conditions: (i) 1-Boc-3-oxopiperazine, K₂CO₃, DMF, 55 °C; (ii) trifluoroacetic acid, dichloromethane, r.t.; (iii) intermediate 4, K₂CO₃, DMF, 55 °C.

3. Results and discussion

3.1. In vitro Anti-HIV assays and SARs analysis

All the target compounds were tested for antiviral activities and cytotoxicities using MT-4 cells infected by HIV-1 III_B or HIV-2 ROD. EC₅₀ and CC₅₀ values for each compound are shown respectively in Table 1. PF74 and 11l were utilized as the control drugs in the assay.

Table 1.

Anti-HIV-1 activity and cytotoxicity in MT-4 cells infected with HIV-1 III_B virus and HIV-2 ROD virus.

Compounds	R1	R2	R3	R4	R5	EC50a (μM)		CC50b (μM)	SIc
						HIV IIIB	HIV ROD		HIV IIIB	HIV ROD
Q-a1	OCH3	H	OCH3	H	CO	3.59 ± 1.00	>24.49	24.49 ± 18.14	6.82	<1.00
Q-a2	OCH3	F	OCH3	F	CO	0.69 ± 0.26	4.31 ± 0.87	13.34 ± 2.61	19.33	3.10
Q-a3	H	H	H	H	CO	4.24 ± 1.37	29.74 ± 6.56	71.78 ± 18.70	16.93	2.41
Q-b1	H	H	–	–	–	246.76 ± 46.36	265.24 ± 24.74	>316.87	>1.28	>1.19
Q-b2	OCH3	H	–	–	–	89.23 ± 78.85	32.93 ± 7.77	>294.46	>3.30	>8.94
Q-b3	OCH3	F	–	–	–	31.07 ± 9.17	≥46.50	>271.63	>8.74	≤5.84
Q-c1	OCH3	H	OCH3	H	CH2	≥3.67	>15.74	15.74 ± 3.95	≤4.29	<1.00
Q-c2	OCH3	H	OCH3	F	CH2	1.48 ± 0.74	>13.48	13.48 ± 3.55	9.11	<1.00
Q-c3	OCH3	H	H	H	CH2	>32.37	>32.37	32.37 ± 10.49	<1.00	<1.00
Q-c4	OCH3	F	OCH3	F	CH2	0.57 ± 0.13	>14.53	14.53 ± 2.56	25.49	<1.00
Q-c5	OCH3	F	OCH3	H	CH2	2.52 ± 0.45	>15.48	15.48 ± 2.33	6.14	<1.00
Q-c6	OCH3	F	H	H	CH2	2.84 ± 0.64	>17.47	17.47 ± 3.31	6.15	<1.00
Q-c7	H	H	H	H	CH2	>32.65	>32.65	32.65 ± 14.24	<1.00	<1.00
Q-c8	H	H	OCH3	F	CH2	4.19 ± 1.22	>12.80	12.80 ± 4.04	3.05	<1.00
Q-c9	H	H	OCH3	H	CH2	≥3.37	>31.29	21.73 ± 11.05	≤6.45	<1.00
11l	–	–	–	–	–	0.16 ± 0.03	0.034 ± 0.017	118.86 ± 2.50	742.88	3495.88
PF74	–	–	–	–	–	0.75 ± 0.33	4.16 ± 2.02	32.27 ± 2.94	43.03	7.76

^cSI: selectivity index, the ratio of CC₅₀/EC₅₀.

^aEC₅₀: the concentration of the compound required to achieve 50% protection of MT-4 cells against HIV-induced cytotoxicity effect, determined in at least triplicate against HIV in MT-4 cells.

^bCC₅₀: the concentration of the compound required to reduce the viability of uninfected cells by 50%, determined in at least triplicate against HIV in MT-4 cells; values were averaged from at least three independent experiments.

Generally, most newly synthesized dimerized compounds exhibited moderate to excellent activities against HIV-1 III_B virus with EC₅₀ values ranging from 0.57 μM (Q-c4) to 4.24 μM (Q-a3) (Table 1), but most of the compounds lost anti-HIV-2 activity. Additionally, the activities of DPAs were generally higher than that of phenylalanine derived monomers (Q-b1, Q-b2, Q-b3), indicating that the dimerization increases the antiviral activity.

As shown in Table 1, compounds with fluorine (Q-c2, Q-c4, Q-c5, Q-c6, Q-c8, Q-a2) on the benzene of phenylalanine displayed better antiviral activities than compounds without fluorine (Q-c1, Q-c3, Q-c7, Q-c9, Q-a1, Q-a3). Meanwhile, both sides replaced by fluorine (Q-c4, Q-a2) favoured the increase of antiviral activity. The p-methoxyl group of benzylamine was also a dominant group for phenylalanine derivatives (Q-c1, Q-c3, Q-c7; Q-c2, Q-c8; Q-c5, Q-c6). This conclusion was the same as that of fluorine substitution, and the methoxyl substitution on both sides was better than the methyl substitution on one side. For the compounds containing asymmetric 1-piperazinone, when the fluorine and methoxyl group substituted on the skeleton of the amide side, the antiviral activity was better than that of the substitute on the other side (Qa-6 > Q-c8). The carbonyl substitution of piperazine also affected the anti-HIV activity. Generally, compounds with two carbonyls show an increase in antiviral activity (Q-a1 > Q-c1, Q-a3 > Q-c7). Interestingly, Q-c4 and Q-a2 were a pair of notable examples. The contribution of the carbonyl group to antiviral activity seemed not apparent in consideration of the SD value.

Even if more fluoride and methoxyl groups were helpful to enhance the anti-HIV-1 activity, it seemed that they were detrimental to the toxicity. The cytotoxicities of these compounds were relatively higher than PF74, leading to low SI values, with Q-c4 having the highest SI value (25.49) due to its low EC₅₀.

The preliminary SARs analysis showed that: i) the DPAs proved to be much more effective than phenylalanine derivative monomers; ii) the compounds bearing more fluoride and methoxyl groups substituents were more potent; iii) two carbonyls on piperazine has proven to be more beneficial for antiviral activities; iv) introducing more fluoride and methoxyl groups could also increase the cytotoxicity.

3.2. Compounds interact with HIV-1 CA as determined by surface plasmon resonance (SPR) assay

Since most of the DPAs that exhibited anti-HIV activity had a similar skeleton, the most potent compounds (Q-c4 and Q-a2) were chosen to test their affinity to the CA monomer and hexamer using SPR with PF74 and 11l as in-line controls.

As shown in Table 2, Figs. 4 and 5, PF74 had the highest binding affinity with an equilibrium dissociation constant (K_D) value of 0.16 ± 0.04 μM (hexamer) and 3.41 ± 1.31 μM (monomer). Different from PF74 and 11l, the two DPAs (Q-c4 and Q-a2) exhibited slightly higher affinities to the monomer than hexamer (ratio <1). This could be explained by several reasons, the simplest being that they occupied more space at the NTD-CTD interface, but in the context of the disulphide-constrained hexamer construct used, making it difficult to enter the CA subunit interface.

Table 2.

SPR results of Q-c4, Q-a2, 11l and PF-74 binding to monomeric and hexameric CA constructs.

Compounds	KDa (μM)		Ratiob	koff(s−1)
	Monomer	Hexamer		Hexamer
Q-a2	6.42 ± 0.83	7.20 ± 1.41	0.98	0.091 ± 0.0089
Q-c4	7.79 ± 0.85	9.04 ± 1.22	0.86	0.11 ± 0.014
11l	13.88 ± 1.68	3.98 ± 0.67	3.48	0.22 ± 0.016
PF74	3.41 ± 1.31	0.16 ± 0.04	21.45	0.018 ± 0.00039

^aAll values represent the average response from at least three replicates. Error bars represent standard deviation (SD).

^bRatio = ${\mathrm{K}}_{\mathrm{D}}^{\text{Monomer}}∕{\mathrm{K}}_{\mathrm{D}}^{\text{Hexamer}}$.

Fig. 4.

SPR sensorgrams of Q-c4 and Q-a2 binding to two variants of the CA protein (monomer and disulfide-stabilized hexamer), respectively, with 11l and PF-74 as the reference.

Fig. 5.

SPR isotherms of Q-c4 and Q-a2 (A) binding to two variants of the CA protein (monomer and disulfide-stabilized hexamer), respectively, with 11l (B) and PF74 (C) as the reference. Isotherms are an average of 3 replicates with error bars represent standard deviation (SD).

Although Q-c4 and Q-a2 did not bind to CA as tight as PF74, they could suppress the virus replication to a similar extent compared to PF74. Again, this may be explained by the steric effect imposed by using the constrained hexameric CA construct used in the SPR, and the interaction with the native CA hexamer may be tighter. However, the new scaffold structure of the DPAs provides a larger space for chemical optimization and improvement of antiviral activity. Additionally, we noted that Q-c4 and Q-a2 displayed a slower off-rate to the CA monomer than 11l and PF74 (Table 2 and Fig. 4). Slower off-rates of inhibitors from their targets has been correlated better with potency than overall affinity [48] and is a parameter that offers room for optimization.

3.3. The DPAs bind to HIV-1 CA in the interprotomer pocket of the hexamer

Previous studies indicated that PF74 binds to the same interprotomer site in the hexamer as the host cell proteins, CPSF6 and NUP153. The DPAs were designed to also bind within this pocket, albeit with larger volumes and increased contacts. Therefore, we next sought to investigate this using SPR-based competition experiments with peptides demonstrated to interact within this region, derived from the CPSF6 and NUP153 proteins. We would like to know if the DPAs could also bind to the same site for large molecular volumes.

First, we demonstrated the ability of the immobilized disulphide-constrained hexameric CA to interact with the CPSF6 and NUP153 peptides. Fig. 6 shows the results and analysis of these experiments, and it verified that not only was our immobilized hexameric CA surface active, but the affinities obtained were in agreement with previously published values [26,49,50]. As the relative sizes of the compounds and the peptides were similar, the competition was evaluated by first saturating the CA surface with compounds, followed by an injection of the relevant peptide. Inhibition was then assessed by comparison to the response of the surface to peptide without prior compound injection. The ability of Q-c4 to compete with CPSF6 and NUP153 peptides was compared to that of PF74. Fig. 7 shows the results of this analysis. Both Q-c4 and PF74 couples impeded the binding of the CPSF6 and NUP153 peptides, indicating a shared binding site. Interestingly, the inhibition profile of Q-c4 was reversed compared to PF74, with Q-c4 competing with NUP153 more substantially, while PF74 inhibited CPSF6 better (Tables 3 and 4). To investigate the potential reason for this reversion of the inhibition profile between the two compounds, we performed rigorous docking calculations with PF74 and Q-c4 and compared them to the crystal structures of the peptides bound to hexameric CA. This analysis revealed that Q-c4 binds in a different orientation due to the larger volume compared with PF74 (Fig. 8). The conformation of Q-c4 is more similar to that of the NUP153 peptide than CPSF6, indicating that it might sterically block the binding pocket for NUP153 more efficiently.

Fig. 6.

SPR sensorgrams and isotherms of CPSF6 and NUP153 binding to HIV-1 NL4-3 CA hexamer. Error bars represent the standard deviation (SD) of three replicates.

Fig. 7.

CPSF6/NUP153 peptide competitive SPR experiments with Q-c4/PF74. Error bars represent the standard deviation (SD) of three replicates.

Table 3.

Binding affinities of CPSF6 and NUP153 peptide binding to hexameric CA. Error bars represent the standard deviation (SD) of three replicates.

Compounds	${\mathrm{K}}_{\mathrm{D}}^{\mathrm{a}}$ (μM)
CPSF6	35.5 ± 5.9
NUP153	94.4 ± 5.9

Table 4.

Results of CPSF6/NUP153 peptide competition assay based on SPR experiments with Q-c4 and PF74. Error bars represent the standard deviation (SD) of three replicates.

Compounds	IC50 (μM)	IC50 (μM)
	CPSF6	NUP153
Q-c4	26.31 ± 5.49	3.47 ± 0.92
PF74	0.22 ± 0.01	8.93 ± 3.70

Fig. 8.

Docking analysis of PF74 (yellow, PDB ID:4XFZ) and Q-c4 (red) compared with CPSF6 (blue) and NUP153 (orange). The co-crystal structures of the CPSF6 (PDB ID: 4WYM) peptide and the NUP153 (PDB ID: 4U0C) peptide are highlighted in blue and orange.

3.4. The DPAs exhibit both early and late-stage inhibition activity

Compounds that bind in the interprotomer pocket of the hexameric CA often have a dual-stage inhibition profile, inhibiting steps in both early and late stages of the viral lifecycle [36]. We have demonstrated that the DPAs also bind in this interprotomer pocket, we next sought to investigate whether they displayed this dual-stage inhibition profile. To achieve this, we performed the single-round infection assay. Table 5 and Fig. 9 show the results of this analysis. Compound 11l, as previously demonstrated, showed inhibition in both the early and late stages, with a more pronounced effect on the early stages. PF74 has also been demonstrated to have early, and late-stage inhibitory properties only inhibited in the early stages. Although a seemingly contradictory result, this is probably a reflection of the concentration of PF-74 used in the late-stage assay. We have demonstrated that PF74 interacts with the monomeric form of CA with a K_D of approximately 3.5 μM. In the late stages, the oligomeric form of CA is monomeric when it is part of the Gag polyprotein. Therefore, our use of 1 μM is likely not high enough to see the late-stage effect of PF74. Interestingly, Q-c4 displayed little to no early-stage inhibition but pronounced late-stage inhibition at the concentrations used (10-fold the EC₅₀). This finding, again, can be rationalized in the context of the concentrations used. We have demonstrated the Q-c4 interacts with hexameric CA, the oligomeric form present in the early stages of the HIV-1 lifecycle and disrupts the interaction of hexameric CA with the host cell factors CPSF6 and NUP153. However, the K_D for the interaction of Q-c4 with hexameric CA was approximately 8 μM, and its IC₅₀ value was even higher for disrupting the interaction of the host cell peptides. Therefore, we believe that Q-c4 will show early-stage inhibition but at concentrations higher than used in our analysis. Moreover, this finding indicates that its late stage effect dominates the EC₅₀ value derived from inhibiting multiple-round infectious virus.

Table 5.

Results of single-round infection assay.

Compounds	Concentration (μM)	% Infectiona
		Early Stage	Late Stage
Q-c4	5	86.6 ± 12.0	0.6 ± 0.1
PF74	1	−0.1 ± 0.2	152.9 ± 10.2
11l	1	7.1 ± 2.1	55.0 ± 19.1
DMSO	–	100.0 ± 8.4	100.0 ± 10.8

^aInfections are an average of 3 replicates with error bars indicating standard error of the mean (SEM).

Fig. 9.

Results of the single-round infection assay. (A) early stages; (B) late stages. Infections are an average of 3 replicates with error bars indicating the SEM.

3.5. Molecular dynamics (MD) simulations study

Q-c4, the most potent HIV-1 CA inhibitor in this study, was selected for the subsequent molecular dynamics simulation. Fig. 10A shows the root mean square deviation (RMSD) of the amino acids of HIV-1 CA during the MD simulation. The Figure shows that the protein structure has deviated from the starting structure, which is the X-ray structure after minimization. The RMSD pattern in Fig. 10A shows that there is more than one conformation for the protein structure. Also, it shows that there are different predominant protein conformations. To further explore the flexibility of the protein, root mean square fluctuation (RMSF) of amino acids of HIV-1 CA was calculated and represented in Fig. 10B. This Figure shows that most amino acids have deviated from the X-ray structure. The high deviation of amino acids supports the results of RMSD, which shows the presence of HIV-1 CA in different conformations. As a result of different conformational existence, Q-c4 could have different binding modes. To display the deviation of Q-c4 from the docked conformer, its RMSD was calculated and represented in Fig. 10C. The Figure shows that Q-c4 has deviated from the starting structure and clustered at different conformers.

Fig. 10.

(A) RMSD (heavy atoms) of amino acids in reference to the first frame of the MD simulation. (B) RMSF of the backbone Cα atoms for amino acids. (C) RMSD (heavy atoms) of the bound Q-c4 in reference to the docked conformer.

The above results show that protein and Q-c4 are present in different conformational states. Accordingly, the whole trajectory was clustered to explore these conformational changes. The aligned structures were clustered according to the Q-c4 structure to explore its binding to the binding site. The clustering procedure returned 8 clusters with the two most populated clusters Fig. 11A and B shows representative structures of the two most populated clusters, respectively. Interestingly, HIV-1 CA has two different conformers, one similar to the X-ray structure and one folded. This new protein conformation was as stable as the first conformer since the two conformers spend nearly identical time during the 500 ns MD simulation.

Fig. 11.

(A) Representative structure of the first cluster. (B) Binding of Q-c4 to the binding site of the first cluster. (C) Representative structure of the second cluster. (D) Binding interactions of Q-c4 to the binding site of the second cluster. Region 1 and Region 2 represent the two symmetrical regions of Q-c4.

Fig. 11B and D shows the interactions between Q-c4 and the binding site in the first and second clusters, respectively. Both structures have the binding site located at the position of the binding site represented by the X-ray structure. However, there are differences in the binding interactions due to the difference in the protein and Q-c4 conformations. It is evident from the Figure that Q-c4 has a nearly similar number of interactions with the binding site in both structures; however, new amino acids are involved in the binding to the second cluster. Q-c4 in the second cluster has nearly its half exposed to the solvent in the HIV-1 CA monomer folded conformation, Fig. 11C and D. For discussion purposes, Q-c4 will be represented by two regions because it is symmetrical nature, Region 1 and Region2 in Fig. 11B and D. Q-c4 interacts with Lys70 by charge assisted hydrogen bond with the carbonyl oxygen atom of Region 2 in the first cluster structure, and by hydrophobic interaction with the pyrazine ring. While it interacts with Lys70 by an ion-induced dipole with 1,3-difluorobenzen of Region 1 and charge assisted hydrogen bond with the carbonyl oxygen of Region 1 in the second cluster structure. According to the hydrogen bond analysis of the MD trajectory, these hydrogen bonds are rarely formed at 1% of the MD time. Q-c4 interacts with Ala64 by hydrophobic interaction with methoxybenzene of Region 1 in the first cluster structure, while it does not interact with Ala64 in the second cluster structure. Ala64 interacts only with the first cluster structure by a hydrogen bond to Region 1 by its backbone carbonyl oxygen. Asn57 interacts with the carbonyl group by hydrogen bond at Region 2 in the first cluster structure. However, it interacts by a hydrogen bond to the carbonyl of Region 1 of the second cluster structure. The total time spent by this hydrogen bond is 62%, which shows the importance of this binding. Also, Asn57 interacts by its backbone carbonyl with an amide hydrogen atom in Region 2 of the second cluster structure at a fraction percent of 41% during the MD time. Leu56 interacts with 1,3-difluorobenzene of Region 2 of the first cluster structure by hydrophobic interactions; however, it interacts with 1,3-difluorobenzene of Region 1 in the second cluster structure. Q-c4 interacts by an aliphatic hydrogen bond with Thr107 through its amide nitrogen in Region 2 in the first cluster structure and with its nitrogen atom in Region 2 of the second cluster structure. Gln50 interacts by hydrogen bonding with methoxybenzene of Region 2 of the first cluster structure. The second cluster structure has some different interactions with Q-c4. It interacts via aliphatic hydrogen bonding between methyl methoxybenzene in Region 1 and backbone carbonyl of Ser109 and Gly106 amino acids. Also, it interacts with Met66 by its 1,3-difluorobenzene of Region 1 by hydrophobic interaction. Val59 and Thr58 interact by aliphatic hydrogen bonding at region 2 through their backbone carbonyl groups.

As shown in Fig. 11, there are two conformers of HIV-1 CA with their corresponding Q-c4 bound conformers. The two bound conformers of Q-c4 show shifting in the second cluster structure due to the different conformers of HIV-1 CA.

3.6. Human plasma and liver microsome (HLM) stability as compared to PF74 and 11l

Poor metabolic stability prevents the further clinical application of PF74. Therefore, we tested the human plasma and liver microsome stability of Q-c4, Q-a2, and leads PF74, 11l. As shown in Fig. 12, after incubation for 120 min at 37 °C, 95% Q-c4 and 96.5% Q-a2 remained in the plasma. However, 85.2% PF74 and 98.9% 11l remained at 120 min. In conclusion, compared with PF74, the plasma stability of Q-a2 and Q-c4 was improved. Human liver microsomes (HLM) assay was performed with testosterone, diclofenac, and propranolol as controls. As shown in Table 6, PF74 was metabolized rapidly (CL_int(mic) = 2862.5 μL/min/mg) while Q-a2 and Q-c4 had slightly improvement on metabolic stability with CL_int(mic) value of 843.9 and 1799.6 μL/min/mg. Nevertheless, although improved compared with PF74, the stabilities of Q-c4 and Q-a2 still have room for further improvements in their stability.

Fig. 12.

Result summary of human plasma stability assay. Experiments were performed in triplicate. % remaining = 100 × (PAR at appointed incubation time/PAR at time T₀). PAR is the peak area ratio of a test compound to the internal standard. Accuracy should be within 80–120% of the indicated value.

Table 6.

Metabolic stability assay in human liver microsomes.

Sample	HLM (Final concentration of 0.5 mg protein/mL)
	R2a	T1/2b (min)	CLint(mic)c (μL/min/mg)	CLint(liver)d (mL/min/kg)	Remaining (T = 60min)	Remaining (NCFe = 60min)
Q-a2	0.8811	1.6	843.9	759.5	0.0%	104.7%
Q-c4	1.0000	0.8	1799.6	1619.6	0.0%	102.4%
11l	0.9803	3.5	395.6	356.0	0.1%	105.8%
PF74	1.0000	0.5	2862.5	2576.2	0.0%	112.6%
Testosterone	0.9982	16.7	82.8	74.5	7.9%	90.7%
Diclofenac	0.9947	3.7	372.0	334.8	0.0%	96.7%
Propafenone	0.9350	5.0	279.5	251.5	0.0%	93.6%

^aR² is the correlation coefficient of the linear regression to determine the kinetic constant (see raw data worksheet in the Supporting Information).

^bT_1/2 is half-life.

^cCL_int(mic) = 0.693/T_1/2/mg microsome protein per Ml.

^dCL_int(liver) = CL_int(mic) × mg microsomal protein/g liver weight × g liver weight/kg body weight.

^eNCF: abbreviation of no co-factor. No NADPH is added to NCF samples (replaced by buffer) during the 60-min incubation. If the NCF remaining is less than 60%, then possibly non-NADPH-dependent metabolism occurs.

3.7. In silico prediction of physicochemical properties

According to the classical definition for drug evaluation, a potential drug candidate should own favorable drug-likeness features. Therefore, to further guide subsequent structural optimization and preliminary drug-like properties evaluation, some significant physicochemical properties of compounds Q-a2, Q-c4 and PF74 were predicted comprehensively by using free online molinspiration software (http://www.molinspiration.com/), including molecular weight (MW), number of hydrogen bond acceptors (nON), number of hydrogen bond donors (nOHNH), number of rotatable bonds (nrotb), topological polar surface area (TPSA), molinspiration predicted LogP (miLogP), number of violations(nViol), number of atoms (natoms) and molecular volume (MV). As displayed in Table 7, PF74 were in full compliance with the “Lipinski rule of five” and Q-a2 and Q-c4 showed a slight violation.

Table 7.

Predicted physicochemical properties calculated by using Molinspiration software for Q-a2, Q-c4 and PF74.

Compounds	MW(Da)	nON	nOHNH	nrotb	TPSA(A2)	miLogP	nViol	natoms	MV(m3/mol)
Accepted range	<500	<10	<5	≤10	<140	<5	–	–
Q-a2	834.82	14	2	16	157.90	4.39	2	60	715.67
Q-c4	820.84	13	2	16	140.83	5.07	3	59	713.49
PF74	425.53	5	2	7	65.20	4.43	0	32	402.55

4. Conclusion

This study has designed, synthesized, and evaluated a novel series of dimerized phenylalanine derivatives as HIV-1 capsid inhibitors. Most of the synthesized compounds displayed anti-HIV-1 activity. Among all, Q-c4 is the most potent, with an EC₅₀ value of 0.57 μM. The SPR assay indicated that Q-c4 and Q-a2 might prefer interaction with the monomeric for CA over the hexameric CA. However, this needs to be further investigated. Competitive SPR experiments showed Q-c4 binds to the same site with NUP153 and CPSF6. The combined results of direct binding SPR, competitive SPR and the single-round infection assay indicate that Q-c4 shares a dual-stage inhibitor profile, typical of compounds that bind to the interprotomer pocket of the hexameric configuration of HIV-1 CA. However, this analysis also implies that its late-stage activity dominates the EC₅₀ derived against the fully infectious virus. Together, our work identifies Q-c4 as a novel CA inhibitor with a unique mechanistic preference for the late-stage that is different from other inhibitors reported. Stability analyses of the Q series demonstrated a small improvement over the poor metabolic stability of PF74 but with room for further improvement. In silico prediction of physicochemical properties shows the dimers have some violation from the Lipinski rule of five. In order to improve the antiviral activity, we plan to introduce substituents at appropriate positions in the molecule to form more effective interaction with surrounding hot residues. Meanwhile, the great rigidity of the molecule may be the reason for the low antiviral activity and poor drug likeness. Using more flexible linkers to improve the flexibility of the molecule may solve the problem. Since the selectivity index is related to CC₅₀ and IC₅₀, improving the antiviral activity and remove the toxic groups will help to increase the SI. Overall, Q-c4 is an attractive chemotype due to its increased molecular size, significant opportunities for optimization of potency and metabolic stability.

5. Experimental section

5.1. Chemistry

¹H NMR and ¹³C NMR spectra were recorded on Bruker AV-400 spectrometer or Bruker AV-600 spectrometer using solvents as indicated (DMSO-d₆). Chemical shifts were reported in δ values (ppm) with tetramethylsilane (TMS) as the internal reference, and J values were reported in hertz (Hz). Melting points (mp) were determined on a micromelting point apparatus and were uncorrected. TLC was performed on Silica Gel GF254 for TLC (Merck), and spots were visualized by iodine vapor or irradiation with UV light (λ = 254 nm). Flash column chromatography was performed on a column packed with Silica Gel60 (200–300 mesh). Thin-layer chromatography was performed on pre-coated HUAN-GHAI_HSGF254, 0.15—0.2 mm TLC-plates. Solvents were of reagent grade and were purified and dried by standard methods when necessary. The concentration of the reaction solutions involved the use of a rotary evaporator at reduced pressure. The solvents of dichloromethane, TEA and methanol etc., were obtained from Sinopharm Chemical Reagent Co., Ltd (SCRC), which were of AR grade. The key reactants, including 4-methoxy-N-methylaniline, N-(tert-butoxycarbonyl)-l-phenylalanine etc. were purchased from Bide Pharmatech Co. Ltd. The purity of final representative compounds was checked by HPLC and was >95%.

5.1.1. General procedure for the synthesis of 2a-2c

To a solution of (tert-butoxycarbonyl)-l-phenylalanine or (tert-butoxycarbonyl)-3,5-difluoro-l-phenylalanine (1.5 eq.) in 20 mL dichloromethane was added PyBop (1.5 eq.) at 0 °C, and the mixture was stirred for 0.5 h. Subsequently, DIEA (3 eq.) and N-methylaniline or 4-methoxy-N-methylaniline (1 eq.) were added to the mixture and then stirred at room temperature for another 6 h (monitored by TLC). The resulting mixture was evaporated under reduced pressure, and the residue was initially washed by 1 N HCl and extracted with ethyl acetate (3 × 20 mL). Then, the combined organic layer was washed with saturated sodium bicarbonate (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to afford a corresponding crude product, purified by flash column chromatography to afford intermediate 2a-2c.

5.1.1.1. Tert-butyl (S)-(1-(methyl(phenyl)amino)-1-oxo-3-phenylpropan-2-yl)carbamate (2a).

Yellow oil, yield: 91%. ¹H NMR (400 MHz, DMSO-d₆): δ 7.49 (t, J = 6.8 Hz, 2H, Ph-H), 7.43 (d, J = 7.1 Hz, 1H, Ph-H), 7.32 (d, J = 7.6 Hz, 2H, Ph-H), 7.12 (s, 4H, Ph-H), 6.72 (s, 2H, Ph-H), 4.17 (s, 1H, CH), 3.17 (s, 3H, NCH₃), 2.68 (dd, J = 46.1, 11.8 Hz, 2H, PhCH₂), 1.31 (s, 9H, C(CH₃)). ESI-MS: m/z 354.87 (M+1)⁺, 377.10 (M+23)⁺. C₂₁H₂₆N₂O₃ [354.45].

5.1.1.2. Tert-butyl (S)-(1-((4-methoxyphenyl)(methyl)amino)-1-oxo-3-phenylpropan-2-yl)carbamate (2b).

Yellow oil, yield: 88%. ¹H NMR (400 MHz, DMSO-d₆): δ 7.22 (d, J = 8.3 Hz, 2H, Ph-H), 7.20–7.11 (m, 3H, Ph-H), 7.09 (d, J = 8.2 Hz, 1H, Ph-H), 7.03 (d, J = 8.6 Hz, 2H, Ph-H), 6.79 (d, J = 7.3 Hz, 2H, Ph-H), 4.27–4.06 (m, 1H, CH), 3.81 (s, 3H, OCH₃), 3.13 (s, 3H, NCH₃), 2.75 (dd, J = 13.4, 3.8 Hz, 1H, PhCH), 2.61 (dd, J = 13.3, 10.3 Hz, 1H, PhCH), 1.30 (s, 9H, C(CH₃)). ESI-MS: m/z 385.4 (M+1)⁺, 407.5 (M+23)⁺. C₂₂H₂₈N₂O₄ [384.48].

5.1.1.3. Tert-butyl (S)-(3-(3,5-difluorophenyl)-1-((4-methoxyphenyl) (methyl)amino)-1-oxopropan-2-yl)carbamate (2c).

Yellow oil, yield: 89%. ¹H NMR (400 MHz, DMSO-d₆): δ 7.31 (d, J = 8.4 Hz, 2H, Ph-H), 7.13–7.04 (m, 3H, Ph-H), 7.01 (d, J = 9.5 Hz, 1H, Ph-H), 6.44 (d, J = 8.3 Hz, 2H, Ph-H), 4.20–4.10 (m, 1H, CH), 3.81 (s, 3H, OCH₃), 3.14 (s, 3H, NCH₃), 2.82–2.61 (m, 2H, PhCH₂), 1.29 (s, 9H, C(CH₃)). ESI-MS: m/z 421.07 (M+1)⁺, 443.17 (M+23)⁺. C₂₂H₂₆F₂N₂O₄ [420.46].

5.1.2. General procedure for the synthesis of 3a-3c

Trifluoroacetic acid (5.0 eq.) was added dropwise to the corresponding substituted intermediate 2 (1.0 eq.) in 30 mL dichloromethane and stirred at room temperature for 1 h (monitored by TLC). Then, the resulting mixture solution was alkalized to pH ~7 with saturated sodium bicarbonate solution and then extracted with dichloromethane (40 mL). Then, the combined organic layer was washed with saturated sodium bicarbonate (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to afford corresponding crude products 3a-3c.

5.1.2.1. (S)-2-amino-N-methyl-N,3-diphenylpropanamide (3a).

Yellow oil, yield: 95%. ¹H NMR (600 MHz, DMSO-d₆): δ 7.40—7.31 (m, 3H, Ph-H), 7.23–7.14 (m, 3H, Ph-H), 7.00 (d, J = 5.2 Hz, 2H, Ph-H), 6.86 (d, J = 4.5 Hz, 2H, Ph-H), 3.35 (t, J = 7.0 Hz, 1H, CH), 3.10 (s, 3H, NCH₃), 2.76 (dd, J = 12.7, 6.6 Hz, 1H, PhCH), 2.48—2.42 (m, 1H, PhCH), 1.80 (s, 2H, NH₂). ESI-MS: m/z 255.07 (M+1)⁺, 508.81 (2 M)⁺. C₁₆H₁₈N₂O [254.33].

5.1.2.2. (S)-2-amino-N-(4-methoxyphenyl)-N-methyl-3-phenylpropanamide (3b).

Yellow oil, yield: 80%. ¹H NMR (400 MHz, DMSO-d₆): δ 7.29—7.13 (m, 3H, Ph-H), 7.03—6.75 (m, 6H, Ph-H), 3.77 (s, 3H, OCH₃), 3.44—3.35 (m, 1H, CH), 3.06 (s, 3H, NCH₃), 2.75 (dd, J = 12.8, 6.7 Hz, 1H, PhCH), 2.45 (dd, J = 12.9, 7.1 Hz, 1H, PhCH), 1.87 (s, 2H, NH₂). ESI-MS: m/z 285.05 (M+1)⁺. C₁₇H₂₀N₂O₂ [284.36].

5.1.2.3. (S)-2-amino-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-N-methylpropanamide (3c).

Yellow oil, yield: 85%. ¹H NMR (400 MHz, DMSO-d₆): δ 7.10—6.93 (m, 5H, Ph-H), 6.57 (h, J = 4.1 Hz, 2H, Ph-H), 3.78 (s, 3H, OCH₃), 3.35 (dd, J = 7.6, 5.9 Hz, 1H, CH), 3.09 (s, 3H, NCH₃), 2.74 (dd, J = 13.1, 5.8 Hz,1H, PhCH), 2.54—2.45 (m, 1H, PhCH), 1.82 (s, 2H, NH₂). ESI-MS: m/z 321.11 (M+1)⁺, m/z 343.25 (M+23)⁺. C₁₇H₁₈F₂N₂O₂ [320.34].

5.1.3. General procedure for the synthesis of 4a-4c

Bromoacetic acid (1.2 eq.) and HATU (1.5 eq.) were mixed in 15 mL dichloromethane and stirred in an ice bath for 0.5 h. Then, the corresponding substituted intermediate 3 (1 eq.) and DIEA (2 eq.) were slowly added to the above solution at 0 °C. The reaction system was then stirred at room temperature for an additional 0.5 h (monitored by TLC). The resulting mixture was evaporated under reduced pressure, and the residue was initially washed by 1 N HCl and extracted with ethyl acetate (3 × 20 mL). Then, the combined organic layer was washed with saturated sodium bicarbonate (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to afford a corresponding crude product, purified by flash column chromatography to afford intermediates 4a-4c.

5.1.3.1. (S)-2-(2-bromoacetamido)-N-methyl-N,3-diphenylpropanamide (4a) yellow oil, yield.

80%. ¹H NMR (600 MHz, DMSO-d₆): δ 8.64 (d, J = 7.7 Hz, 1H, NH), 7.46—7.36 (m, 3H, Ph-H), 7.17 (s, 5H, Ph-H), 6.82 (d, J = 5.2 Hz, 2H, Ph-H), 4.47—4.41 (m, 1H, CH), 3.82 (d, J = 3.5 Hz, 2H, CH₂), 3.14 (s, 3H, NCH₃), 2.86 (dd, J = 13.3, 5.5 Hz, 1H, PhCH), 2.64 (dd, J = 13.6, 9.0 Hz, 1H, PhCH). ¹³C NMR (150 MHz, DMSO-d₆): δ 170.84 (C═O), 165.98 (C═O), 143.20, 137.57, 130.04, 129.30, 128.61, 128.36, 127.90, 126.96, 52.27, 38.24, 37.63, 29.48. ESI-MS: m/z 377.28 (M+1)⁺, 378.69 (M+23)⁺. C₁₈H₁₉BrN₂O₂ [375.27].

5.1.3.2. (S)-2-(2-bromoacetamido)-N-(4-methoxyphenyl)-N-methyl-3-phenylpropanamide (4b).

White oil, yield: 68%. ¹H NMR (600 MHz, DMSO-d₆): δ 8.62 (d, J = 7.9 Hz, 1H, NH), 7.22–7.16 (m, 3H, Ph-H), 7.05 (d, J = 8.3 Hz, 2H, Ph-H), 6.96 (d, J = 9.2 Hz, 2H, Ph-H), 6.88 (d, J = 6.2 Hz, 2H, Ph-H), 4.44 (td, J = 8.4, 5.6 Hz, 1H, CH), 3.82 (d, J = 2.4 Hz, 2H, CH₂), 3.79 (s, 3H, OCH₃), 3.10 (s, 3H, NCH₃), 2.87 (dd, J = 13.5, 5.5 Hz, 1H, PhCH), 2.65 (dd, J = 13.5, 8.7 Hz, 1H, PhCH). ¹³C NMR (150 MHz, DMSO-d₆): δ 171.05 (C═O), 165.90 (C═O), 159.07, 137.62, 135.89, 129.36, 129.06, 128.61, 126.95, 115.18, 55.92, 52.10, 37.94, 37.77, 29.51. ESI-MS: m/z 405.4 (M+1)⁺. C₁₉H₂₁BrN₂O₃ [405.29].

5.1.3.3. (S)-2-(2-bromoacetamido)-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-N-methylpropanamide (4c).

White solid, yield: 86%. ¹H NMR (400 MHz, DMSO-d₆): δ 8.70 (d, J = 8.0 Hz, 1H, NH), 7.22 (d, J = 8.8 Hz, 2H, Ph-H), 7.04 (d, J = 8.6 Hz, 3H, Ph-H), 6.52 (d, J = 6.3 Hz, 2H, Ph-H), 4.44 (qd, J = 8.6, 4.6 Hz, 1H, CH), 3.81 (s, 2H, BrCH₂), 3.80 (s, 3H, OCH₃), 3.13 (s, 3H, NCH₃), 2.89 (dd, J = 13.7, 4.6 Hz, 1H, PhCH), 2.69 (dd, J = 13.7, 9.3 Hz, 1H, PhCH). ¹³C NMR (151 MHz, DMSO-d₆): δ 170.57 (C═O), 166.05 (C═O), 162.55 (dd, ¹J_C-F = 245.9, ³J_C-F = 13.3 Hz), 159.24, 142.20 (t, ³J_C-F = 9.5 Hz), 135.79, 129.10, 115.31, 112.42 (dd, ²J_C-F = 19.8, 5.0 Hz), 102.50 (t, ²J_C-F = 25.6 Hz), 55.95, 51.60, 37.76, 37.31, 29.32. ESI-MS: m/z 443.15 (M+2)⁺, m/z 463.16 (M-1+23)⁺. C₁₇H₁₈F₂N₂O₂ [441.27].

5.1.4. General procedure for the synthesis of Q-a1-a3

Under ice bath, the corresponding substituted key intermediate 4 (2 eq.), piperazine-2,5-dione (1 eq.), Cs₂CO₃ (3 eq.) were dissolved in the solution of DMF (6 mL). The resulting mixture was then stirred at 40 °C (monitored by TLC). Then the reaction mixture was extracted with ethyl acetate (20 mL), and the combined organic phase was washed with saturated NaCl solution (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to give the corresponding crude product which was purified by flash column chromatography to afford products Q-a1-a3.

5.1.4.1. (2S,2′S)-2,2'-((2,2'-(2,5-dioxopiperazine-1,4-diyl)bis(acetyl))bis(azanediyl)) bis(N-(4-methoxyphenyl)-N-methyl-3-phenylpropanamide) (Q-a1).

Yellow solid, yield: 68%. mp: 110-112 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.49 (d, J = 7.8 Hz, 2H, NH × 2), 7.19 (q, J = 6.9, 6.1 Hz, 6H, Ph-H), 7.10 (d, J = 6.3 Hz, 4H, Ph-H), 6.98 (d, J = 8.3 Hz, 4H, Ph-H), 6.87 (d, J = 6.7 Hz, 4H, Ph-H), 4.46 (q, J = 7.8 Hz, 2H, CH × 2), 4.04–3.82 (m, 8H, CH₂ × 4), 3.79 (s, 6H, OCH₃ × 2), 3.11 (s, 6H, NCH₃ × 2), 2.87 (dd, J = 13.3, 4.5 Hz, 2H, PhCH), 2.65 (dd, J = 12.7, 9.8 Hz, 2H, PhCH). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.36 (C═O), 167.23 (C═O), 164.03 (C═O), 159.00, 137.88, 135.90, 129.32, 129.13, 128.63, 126.92, 115.13, 55.87, 51.84, 50.58, 47.43, 37.84, 37.78. HRMS: m/z 763.3448 (M+1)⁺. C₄₂H₄₆N₆O₈ [762.3377].

5.1.4.2. (2S,2S)-22′-((2,2'-(2,5'-dioxopiperazine-1,4-diyl)bis(acetyl))bis(azanediyl)) bis(3-(3,5-di-N-difluorophenyl)-N-(4-methoxyphenyl)-N-methylpropanamide) (Q-a2).

White solid, yield: 32%. mp: 195-197 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.52 (d, J = 7.8 Hz, 2H, NH × 2), 7.23 (d, J = 7.9 Hz, 4H, Ph-H), 7.03 (t, J = 8.0 Hz, 6H, Ph-H), 6.51 (d, J = 7.0 Hz, 4H, Ph-H), 4.46 (q, J = 8.3 Hz, 2H, CH × 2), 4.08–3.83 (m, 8H, CH₂ × 4), 3.79 (s, 6H, OCH₃ × 2), 3.13 (s, 6H, NCH₃ × 2), 2.95–2.83 (m, 2H, PhCH), 2.69 (dd, J = 13.2, 9.7 Hz, 2H, PhCH). ¹³C NMR (100 MHz, DMSO-d₆): δ 170.83 (C=O), 167.35 (C=O), 164.07 (C=O), 162.52 (dd, ¹J_CF = 245.9, ³J_CF = 13.4 Hz), 159.15, 142.50 (d, ³J_CF = 9.5 Hz), 135.79, 129.16, 115.25, 112.59–112.10 (m), 102.53 (t, ²J_CF = 25.8 Hz), 55.89, 51.43, 50.63, 47.47, 37.75, 37.20. HRMS: m/z 835.3068 (M+1)⁺. C₄₂H₄₂F₄N₆O₈ [834.3000].

5.1.4.3. (2S,2′S)-2,2'-((2,2'-(2,5-dioxopiperazine-1,4-diyl)bis(acetyl))bis(azanediyl))bis(N-methyl-N,3-diphenylpropanamide) (Q-a3).

Yellow solid, yield: 71%. mp: 126-127 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.51 (d, J = 7.6 Hz, 2H, NH × 2), 7.54–7.36 (m, 6H, Ph-H), 7.20 (dd, J = 21.7, 5.8 Hz, 10H, Ph-H), 6.81 (d, J = 5.0 Hz, 4H, Ph-H), 4.47 (q, J = 7.8 Hz, 2H, CH × 2), 4.07–3.83 (m, 8H, CH₂ × 4), 3.16 (s, 6H, NCH₃ × 2), 2.95–2.78 (m, 2H, PhCH), 2.65 (dd, J = 12.9, 9.6 Hz, 2H, PhCH). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.13 (C═O), 167.34 (C═O), 164.06 (C═O), 143.24, 137.84, 130.08, 129.24, 128.62, 128.37, 127.99, 126.93, 52.04, 50.60, 47.46, 37.77, 37.63. HRMS: m/z 703.3131 (M+1)⁺. C₄₀H₄₂N₆O₆ [702.3166].

5.1.5. General procedure for the synthesis of 5a-5c

Under ice bath, the corresponding substituted key intermediate 4 (1 eq.), 1-Boc-3-oxopiperazine (1.2 eq.), K₂CO₃ (2 eq.) were dissolved in the solution of DMF (6 mL). The resulting mixture was then stirred at 40 °C (monitored by TLC). Then the reaction mixture was extracted with ethyl acetate (20 mL), and the combined organic phase was washed with saturated NaCl solution (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to give the corresponding crude product which was purified by flash column chromatography to afford products 5a-5c.

5.1.5.1. Tert-butyl (S)-4-(2-((1-(methyl(phenyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazine-1-carboxylate (5a).

White oil, yield: 81%. ¹H NMR (600 MHz, DMSO-d₆): δ 8.36 (d, J = 8.0 Hz, 1H, NH), 7.48–7.35 (m, 3H, Ph-H), 7.22 (d, J = 7.5 Hz, 2H, Ph-H), 7.15 (s, 3H, Ph-H), 6.80 (d, J = 8.1 Hz, 2H, Ph-H), 4.48 (s, 1H, CH), 3.92 (q, J = 16.5 Hz, 4H, CH₂ × 2), 3.48 (s, 2H, CH₂), 3.15 (s, 3H, NCH₃), 3.13 (s, 2H, CH₂), 2.84 (dd, J = 13.8, 4.7 Hz, 1H, PhCH)), 2.68–2.57 (m, 1H, PhCH)), 1.41 (s, 9H C(CH₃)₃). ESI-MS: m/z 495.17 (M+1)⁺, 517.30 (M+23)⁺, 493.13 (M – 1)⁻. C₂₇H₃₄N₄O₅ [494.59].

5.1.5.2. Tert-butyl (S)-4-(2-((1-((4-methoxyphenyl)(methyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazine-1-carboxylate (5b).

White solid, yield: 44%. ESI-MS: m/z 523.09 (M – 1)⁻. C₂₈H₃₆N₄O₆ [524.62].

5.1.5.3. Tert-butyl (S)-4-(2-((3-(3,5-difluorophenyl)-1-((4-methoxyphenyl) (methyl) amino)-1-oxopropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazine-1-carboxylate (5c).

White solid, yield: 91%. ¹H NMR (400 MHz, DMSO-d₆): 5 8.42 (d, J = 8.0 Hz, 1H, NH), 7.24 (d, J = 8.3 Hz, 2H, Ph-H), 7.02 (d, J = 7.1 Hz, 3H, Ph-H), 6.49 (d, J = 6.3 Hz, 2H, Ph-H), 4.46 (td, J = 8.8, 4.2 Hz, 1H, CH), 4.00–3.85 (m, 4H, CH₂ × 2), 3.79 (s, 3H, OCH₃), 3.50 (s, 2H, CH₂), 3.16 (t, J = 5.4 Hz, 2H, CH₂), 3.13 (s, 3H, NCH₃), 2.87 (dd, J = 13.6, 4.4 Hz, 1H, PhCH), 2.69 (dd, J = 13.6, 9.5 Hz, 1H, PhCH), 1.41 (s, 9H, C(CH₃)₃). ESI-MS: m/z 561.00 (M+1)⁺, 583.19 (M+23)⁺, 599.16 (M+39)⁺, 559.22 (M – 1)⁻. C₂₈H₃₄F₂N₄O₆ [560.60].

5.1.6. General procedure for the synthesis of Q-b1-b3

Trifluoroacetic acid (5.0 eq.) was added dropwise to corresponding substituted intermediate 5 (1.0 eq.) in 30 mL dichloromethane and stirred at room temperature for 1 h (monitored by TLC). Then, the resulting mixture solution was alkalized to pH ~7 with saturated sodium bicarbonate solution and then extracted with dichloromethane (40 mL). Then, the combined organic layer was washed with saturated sodium bicarbonate (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to afford a corresponding crude product purified by flash column chromatography to afford products Q-b1-b3.

5.1.6.1. (S)–N-methyl-2-(2-(2-oxopiperazin-1-yl)acetamido)-N,3-diphenyl propanamide (Q-b1).

White oil, yield: 66%. ¹H NMR (600 MHz, DMSO-d₆): δ 8.25 (d, J = 8.0 Hz, 1H, NH), 7.44 (t, J = 7.5 Hz, 3H, Ph-H), 7.25–7.11 (m, 5H, Ph-H), 6.81 (d, J = 6.6 Hz, 2H, Ph-H), 4.50 (q, J = 8.2 Hz, 1H, CH), 3.93–3.79 (m, 2H, CH₂), 3.20 (s, 2H, CH₂), 3.15 (s, 3H, NCH₃), 3.03 (t, J = 5.6 Hz, 2H, CH₂), 2.87–2.81 (m, 2H, CH₂), 2.80 (d, J = 5.5 Hz, 1H, PhCH), 2.66 (dd, J = 13.6, 9.3 Hz, 1H). ¹³C NMR (150 MHz, DMSO-d₆): δ 171.23 (C═O), 168.12 (C═O), 167.98 (C═O), 143.35, 137.88, 130.02, 129.27, 128.55, 128.30, 127.95, 126.86, 51.90, 50.19, 48.68, 42.87, 37.84, 37.64. HRMS: m/z 395.2083 (M+1)⁺, 789.4086 (2 M + 1)⁺. C₂₂H₂₆N₄O₃ [394.2005].

5.1.6.2. (S)–N-(4-methoxyphenyl)-N-methyl-2-(2-(2-oxopiperazin-1-yl)acetamido)-3-phenylpropanamide (Q-b2).

Yellow oil, yield: 63%. ¹H NMR (600 MHz, DMSO-d₆): δ 8.23 (d, J = 8.0 Hz, 1H, NH), 7.21 —7.07 (m, 5H, Ph-H), 6.97 (d, J = 9.0 Hz, 2H, Ph-H), 6.86 (d, J = 6.3 Hz, 2H, Ph-H), 4.48 (td, J = 8.6, 5.1 Hz, 1H, CH), 3.87 (q, J = 16.1 Hz, 2H, CH₂), 3.79 (s, 3H, OCH₃), 3.24 (s, 2H, CH₂), 3.10 (s, 3H, NCH₃), 3.04 (t, J = 5.7 Hz, 2H, CH₂), 2.85 (dd, J = 13.6, 5.1 Hz, 2H, CH₂), 2.82 (d, J = 5.7 Hz, 1H, PhCH), 2.66 (dd, J = 13.6, 9.0 Hz, 1H, PhCH), 1.24 (s, 1H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 170.94 (C═O), 168.23 (C═O), 168.03 (C═O), 162.51 (dd, ¹J_C-F = 245.8, ³J_C-F = 13.3 Hz), 159.18 (C═O), 142.60 (t, ³J_C-F = 9.4 Hz), 135.89, 129.17, 115.28, 112.58–112.08 (m), 102.44 (t, ²J_C-F = 25.7 Hz), 55.93, 51.43, 50.24, 48.80, 48.66, 42.88, 37.75, 37.16. HRMS: m/z 425.2162 (M+1)⁺. C₂₃H₂₈N₄O₄ [424.2111].

5.1.6.3. (S)–N-methyl-2-(2-(2-oxopiperazin-1-yl)acetamido)-N,3-diphenyl propanamide (Q-b3).

Yellow oil, yield: 83%. ¹H NMR (400 MHz, DMSO-d₆): δ 8.31 (d, J = 8.0 Hz, 1H, NH), 7.26 (d, J = 8.5 Hz, 2H, Ph-H), 7.03 (d, J = 8.3 Hz, 3H, Ph-H), 6.50 (h, J = 4.6 Hz, 2H, Ph-H), 4.46 (td, J = 8.8, 4.2 Hz, 1H, CH), 3.87 (d, J = 1.9 Hz, 2H, CH₂), 3.80 (s, 3H, OCH₃), 3.21 (s, 2H, CH₂), 3.13 (s, 3H, NCH₃), 3.04 (t, J = 5.4 Hz, 2H, CH₂), 2.86 (dd, J = 13.6, 4.2 Hz, 1H, PhCH), 2.81 (q, J = 4.9 Hz, 2H, CH₂), 2.70 (dd, J = 13.9, 9.6 Hz, 1H, PhCH), 1.24 (s,1H, NH). ¹³C NMR (100 MHz, DMSO-d₆): δ 170.94 (C═O), 168.23 (C═O), 168.03 (C═O), 162.51 (dd, ¹J_CF = 245.8, ³J_CF = 13.3 Hz), 159.18, 142.60 (t, ³J_CF = 9.4 Hz), 135.89, 129.17, 115.28, 112.56–112.15 (m), 102.44 (t, ²J_CF = 25.7 Hz), 55.93, 51.43, 50.24, 48.80, 48.66, 42.88, 37.75, 37.16. HRMS: m/z 461.1993 (M+1)⁺, 921.3886 (2 M + 1)⁺. C₂₃H₂₆F₂N₄O₄ [460.1922].

5.1.7. General procedure for the synthesis of Q-c1-c9

Under ice bath, the corresponding substituted key intermediate 4 (1 eq.), corresponding substituted chemicals Q-b1-b3 (1.1 eq.), K₂CO₃ (2 eq.) were dissolved in the solution of DMF (6 mL). The resulting mixture was then stirred at 40 °C (monitored by TLC). Then the reaction mixture was extracted with ethyl acetate (20 mL), and the combined organic phase was washed with saturated NaCl solution (3 × 20 mL), dried over anhydrous Na₂SO₄, filtered, and concentrated under reduced pressure to give the corresponding crude products which were purified by flash column chromatography to afford compounds Q-c1-c9.

5.1.7.1. (2S,2′S)-2,2'-((2,2'-(2-oxopiperazine-1,4-diyl) bis(acetyl))bis(azanediyl)) bis(N-(4-methoxyphenyl)-N-methyl-3-phenylpropanamide) (Q-c1).

White solid, yield: 95%. mp: 125-126 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.35 (d, J = 7.9 Hz, 1H, NH), 7.92 (d, J = 8.0 Hz, 1H, NH), 7.17 (t, J = 7.1 Hz, 10H, Ph-H), 6.97 (d, J = 9.5 Hz, 4H, Ph-H), 6.86 (d, J = 7.7 Hz, 4H, Ph-H), 4.56–4.41 (m, 2H, CH × 2), 3.88 (dd, J = 25.3, 18.3 Hz, 2H, CH₂), 3.80 (s, 3H, OCH₃), 3.78 (s, 3H, OCH₃), 3.11 (s, 6H, NCH₃ × 2), 3.10–2.86 (m, 8H, CH₂ × 4), 2.86—2.57 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.48 (C═O), 171.42 (C═O), 168.75 (C═O), 167.73 (C═O), 166.46 (C═O), 159.06, 159.00, 137.95, 137.85, 135.96, 129.34, 129.32, 129.16, 128.60, 126.90, 115.20, 115.14, 59.44, 56.83, 55.91, 55.88, 51.82, 51.36, 49.37, 48.07, 47.05, 37.83, 37.78, 37.56. HRMS: m/z 749.3601 (M+1)⁺, 771.3411 (M+23)⁺. C₄₂H₄₈N₆O₇ [748.3584].

5.1.7.2. (S)-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-2-(2-(4-(2-(((S)-1-((4-methoxyphenyl)(methyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazin-1-yl)acetamido)-N-methylpropanamide (Q-c2).

White solid, yield: 92%. mp: 114-115 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.37 (d, J = 8.0 Hz, 1H, NH), 8.08 (s, 1H, NH), 7.28 (d, J = 8.3 Hz, 2H, Ph-H), 7.24–7.09 (m, 5H, Ph-H), 7.02 (dd, J = 27.0, 8.4 Hz, 5H, Ph-H), 6.86 (d, J = 7.0 Hz, 2H, Ph-H), 6.49 (d, J = 7.7 Hz, 2H, Ph-H), 4.50 (dt, J = 21.5, 7.1 Hz, 2H, CH × 2), 3.89 (s, 2H, CH₂), 3.81 (s, 3H, OCH₃), 3.79 (s, 3H, OCH₃), 3.14 (s, 3H, NCH₃), 3.11 (s, 3H, NCH₃), 3.11–2.81 (m, 8H, CH₂ × 4), 2.81–2.53 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.46 (C═O), 170.97 (C═O), 167.70(C═O), 162.49 (dd, ¹J_C-F = 246.8, ³J_C-F = 14.2 Hz), 159.20 (C═O), 159.00 (C═O), 137.94, 135.88 (d, ²J_C-F = 12.9 Hz), 129.34, 129.23, 129.15, 128.59, 115.30, 115.14, 112.33 (d, ³J_C-F = 24.9 Hz), 102.80–102.12 (m), 55.94, 55.87, 51.80, 51.14, 49.34, 48.05, 46.83, 37.79, 36.86. HRMS: m/z 785.424 (M+1)⁺, 807.3228 (M+23)⁺. C₄₂H₄₆F₂N₆O₇ [784.3396].

5.1.7.3. (S)–N-(4-methoxyphenyl)-N-methyl-2-(2-(4-(2-(((S)-1-(methyl(phenyl) amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-2-oxopiperazin-1-yl)acetamido)-3-phenylpropanamid (Q-c3).

White solid, yield: 70%. mp: 93-94 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.37 (d, J = 7.6 Hz, 1H, NH), 7.99 (s, 1H, NH), 7.46 (dq, J = 14.8, 8.2, 7.8 Hz, 3H, Ph-H), 7.29 (d, J = 6.5 Hz, 2H, Ph-H), 7.24–7.06 (m, 8H, Ph-H), 6.98 (d, J = 8.2 Hz, 2H), 6.87 (d, J = 6.5 Hz, 2H, Ph-H), 6.79 (d, J = 4.1 Hz, 2H, Ph-H), 4.64–4.38 (m, 2H, CH × 2), 4.04–3.82 (m, 2H, CH₂), 3.79 (s, 3H, OCH₃), 3.17 (s, 3H, NCH₃), 3.11 (s, 3H, NCH₃), 3.10–2.77 (m, 8H, CH₂ × 4), 2.76–2.53 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.48 (C═O), 171.20 (C═O), 167.72 (C═O), 159.00 (C═O), 143.29, 137.95, 137.83, 135.95, 130.14, 129.34, 129.23, 129.16, 128.61, 128.44, 128.05, 126.91, 115.14, 55.88, 51.83, 51.61, 49.33, 48.06, 37.79, 37.68, 37.45. HRMS: m/z 719.3531 (M+1)⁺, 741.3351 (M+23)⁺,. C₄₂H₄₆F₂N₆O₇ [718.3479].

5.1.7.4. (2S,2′S)-2,2'-((2,2'-(2-oxopiperazine-1,4-diyl)bis(acetyl))bis(azanediyl)) bis(3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-N-methylpropanamide) (Q-c4)

White solid, yield: 79%. mp: 133-134 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.41 (d, J = 7.8 Hz, 1H, NH), 8.08 (s, 1H, NH), 7.27 (t, J = 8.7 Hz, 4H, Ph-H), 7.04 (t, J = 7.2 Hz, 6H, Ph-H), 6.50 (t, J = 7.1 Hz, 4H, Ph-H), 4.62–4.38 (m, 2H, CH × 2), 4.03–3.83 (m, 2H, CH₂), 3.83–3.77 (m, 6H, OCH₃ × 2), 3.14 (s, 6H, NCH₃ × 2), 3.12–2.76 (m, 8H, CH₂ × 4), 2.76–2.52 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 170.95 (C═O), 167.81 (C═O), 162.50 (dd, ¹J_C-F = 245.7, ³J_C-F = 14.1 Hz), 159.21 (C═O), 159.16 (C═O), 142.61, 129.21, 115.30, 115.26, 112.35 (d, ²J_C-F = 23.5 Hz), 102.49 (t, ³J_C-F = 25.9 Hz), 55.94, 55.91, 51.43, 51.12, 49.33, 48.10, 37.79, 37.74, 36.91. HRMS: m/z 821.3279 (M+1)⁺, 843.3096 (M+23)⁺. C₄₂H₄₆F₂N₆O₇ [820.3208].

5.1.7.5. (S)-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-2-(2-(4-(2-(((S)-1-((4-methoxyphenyl)(methyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-2-oxopiperazin-1-yl)acetamido)-N-methyl-propanamide (Q-c5).

White solid, yield: 82%. mp: 115-116 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.43 (d, J = 7.7 Hz, 1H, NH), 7.98 (s, 1H, NH), 7.26 (d, J = 8.1 Hz, 2H, Ph-H), 7.17 (d, J = 7.3 Hz, 5H, Ph-H), 7.02 (t, J = 8.5 Hz, 5H, Ph-H), 6.85 (d, J = 6.8 Hz, 2H, Ph-H), 6.51 (d, J = 7.1 Hz, 2H, Ph-H), 4.61–4.37 (m, 2H, CH × 2), 4.04–3.85 (m, 2H, CH₂), 3.80 (s, 3H, OCH₃), 3.79 (s, 3H, OCH₃), 3.13 (d, J = 6.5 Hz, 6H, NCH₃ × 2), 3.12–2.77 (m, 8H, CH₂ × 4), 2.76–2.53 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.37 (C═O), 170.94 (C═O), 167.79 (C═O), 162.52 (dd, ¹J_C-F = 247.3, ³J_C-F = 12.7 Hz), 159.16 (C═O), 159.07 (C═O), 142.55, 137.82, 135.93, 135.85, 129.32, 129.18, 128.62, 126.92, 115.26, 115.20, 112.37 (d, ²J_C-F = 24.9 Hz), 102.50, 55.91, 51.45, 49.30, 48.12, 37.82, 37.75, 37.59, 37.18. HRMS: m/z 785.3433 (M+1)⁺, 807.3258 (M+23)⁺. C₄₂H₄₆F₂N₆O₇ [784.3396].

5.1.7.6. (S)-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-N-methyl-2-(2-(4-(2-(((S)-1-(methyl(phenyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-2-oxopiperazin-1-yl)acetamido )propenamide (Q-c6).

Yellow solid, yield: 97%. mp: 112-113 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.42 (d, J = 7.8 Hz, 1H, NH), 7.98 (s, 1H, NH), 7.46 (dq, J = 14.8, 8.2, 7.7 Hz, 3H, Ph-H), 7.34–7.22 (m, 4H, Ph-H), 7.14 (s, 3H, Ph-H), 7.04 (t, J = 8.6 Hz, 3H, Ph-H), 6.79 (s, 2H, Ph-H), 6.51 (d, J = 7.0 Hz, 2H, Ph-H), 4.61–4.39 (m, 2H, CH × 2), 3.92 (q, J = 16.8 Hz, 2H, CH₂), 3.79 (s, 3H, OCH₃), 3.17 (s, 3H, NCH₃), 3.14 (s, 3H, NCH₃), 3.12–2.77 (m, 8H, CH₂ × 4), 2.77–2.52 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.18 (C═O), 170.94 (C═O), 167.82 (C═O), 162.51 (dd, ¹J_C-F = 246.5, ³J_C-F = 14.0 Hz), 159.16 (C═O), 143.29, 126.91, 115.26, 112.37 (d, ²J_C-F = 24.3 Hz), 102.50, 60.22, 55.91, 51.59, 51.46, 49.32, 48.12, 37.75, 37.67, 37.50, 37.15, 21.23, 14.55. HRMS: m/z 755.3367 (M+1)⁺, 777.3168 (M+23)⁺. C₄₂H₄₆F₂N₆O₇ [754.3290].

5.1.7.7. (2S,2′S)-2,2'-((2,2'-(2-oxopiperazine-1,4-diyl)bis(acetyl))bis(azanediyl)) bis(N-methyl-N,3-diphenylpropanamide) (Q-c7).

Yellow solid, yield: 98%. mp: 91-93 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.41 (d, J = 7.9 Hz, 1H, NH), 7.98 (d, J = 7.9 Hz, 1H, NH), 7.52–7.40 (m, 6H, Ph-H), 7.27 (dd, J = 14.6, 6.8 Hz, 4H, Ph-H), 7.21 −7.13 (m, 6H, Ph-H), 6.81 (d, J = 7.4 Hz, 4H, Ph-H), 4.50 (dt, J = 21.8, 6.7 Hz, 2H, CH × 2), 3.88 (q, J = 17.9, 16.7 Hz, 2H, CH₂), 3.16 (s, 6H, NCH₃ × 2), 3.15–2.87 (m, 8H, CH₂ × 4), 2.87–2.78 (m, 2H, PhCH), 2.77–2.61 (m, 2H, PhCH). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.24 (C═O), 171.19 (C═O), 167.82 (C═O), 143.29, 137.91, 137.82, 130.08, 129.25, 128.60, 128.44, 128.36, 128.02, 126.92, 60.22, 52.02, 51.59, 49.32, 48.07, 37.64, 37.49. HRMS: m/z 689.347 (M+1)⁺, 711.3229 (M+23)⁺. C₄₀H₄₄N₆O₅ [688.3373].

5.1.7.8. (S)-3-(3,5-difluorophenyl)-N-(4-methoxyphenyl)-N-methyl-2-(2-(4-(2-(((S)-1-(methyl(phenyl)amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazin-1-yl)acetamido)propenamide(Q-c8).

White solid, yield: 81%. mp: 172-174 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.40 (d, J = 7.9 Hz, 1H, NH), 8.08 (s, 1H, NH), 7.44 (dt, J = 14.2, 7.1 Hz, 3H, Ph-H), 7.26 (dd, J = 15.2, 7.9 Hz, 4H, Ph-H), 7.16 (d, J = 5.6 Hz, 3H, Ph-H), 7.03 (dd, J = 18.7, 8.9 Hz, 3H, Ph-H), 6.80 (d, J = 6.4 Hz, 2H, Ph-H), 6.49 (d, J = 7.8 Hz, 2H, Ph-H), 4.51 (ddd, J = 19.3, 9.2, 4.9 Hz, 2H, CH × 2), 4.01–3.83 (m, 2H, CH₂), 3.81 (s, 3H, OCH₃), 3.16 (s, 3H, NCH₃), 3.14 (s, 3H, NCH₃), 3.12–2.82 (m, 8H, CH₂ × 4), 2.82–2.54 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, DMSO-d₆): δ 171.23 (C═O), 170.96 (C═O), 167.79 (C═O), 162.49 (dd, ¹J_C-F = 245.8, ³J_C-F = 13.5 Hz), 159.21 (C═O), 143.29, 142.62, 137.89, 135.83, 130.08, 129.24, 128.58, 128.36, 128.00, 126.89, 115.31, 112.34 (d, ²J_C-F = 24.3 Hz), 102.49, 59.31, 55.95, 51.99, 51.12, 49.34, 48.5 37.79, 37.63, 36.90. HRMS: m/z 755.3363 (M+1)⁺, 777.3158 (M+23)⁺. C₄₀H₄₄N₆O₅ [754.3290].

5.1.7.9. (S)–N-(4-methoxyphenyl)-N-methyl-2-(2-(4-(2-(((S)-1-(methyl(phenyl) amino)-1-oxo-3-phenylpropan-2-yl)amino)-2-oxoethyl)-3-oxopiperazin-1-yl)acetamido)-3-phenylpropanamide (Q-c9).

White solid, yield: 96%. mp: 108-110 °C. ¹H NMR (400 MHz, DMSO-d₆): δ 8.40 (d, J = 7.5 Hz, 1H, NH), 7.94 (d, J = 7.6 Hz, 1H, NH), 7.43 (dq, J = 13.6, 6.8 Hz, 3H, Ph-H), 7.25 (d, J = 6.5 Hz, 2H, Ph-H), 7.17 (d, J = 6.3 Hz, 8H, Ph-H), 7.01 (d, J = 8.1 Hz, 2H, Ph-H), 6.85 (d, J = 6.8 Hz, 2H, Ph-H), 6.81 (d, J = 4.9 Hz, 2H, Ph-H), 4.50 (dq, J = 14.7, 7.8 Hz, 2H, CH × 2), 3.92 (q, J = 18.3, 17.4 Hz, 2H, CH₂), 3.80 (s, 3H, OCH₃), 3.16 (s, 3H, NCH₃), 3.12 (s, 3H, NCH₃), 3.10–2.80 (m, 8H, CH₂ × 4), 2.76–2.51 (m, 4H, PhCH₂ × 2). ¹³C NMR (100 MHz, Methanol-d₄): δ 172.21 (C═O), 172.06 (C═O), 168.64 (C═O), 159.87 (C═O), 144.07, 138.69, 138.61, 136.71, 130.90, 130.12, 130.05, 129.99, 129.44, 129.41, 129.19, 128.81, 127.74, 116.00, 56.71, 52.86, 52.21, 50.10, 38.64, 38.45, 38.35.HRMS: m/z 719.3537 (M+1)⁺, 741.3334 (M+23)⁺. C₄₀H₄₄N₆O₅ [718.3479].

5.2. In vitro Anti-HIV assay in MT-4 cells

Evaluation of the antiviral activity of the compounds against HIV in MT-4 cells was performed using the MTT assay as described below. Stock solutions (10 × final concentration) of test compounds were added in 25 μL volumes to two series of triplicate wells to allow simultaneous evaluation of their effects on mock- and HIV-infected cells at the beginning of each experiment. Serial 5-fold dilutions of test compounds were made directly in flat-bottomed 96-well microtiter trays using a Biomek 3000 robot (Beckman Instruments, Fullerton, CA). Untreated HIV- and mock-infected cell samples were included as controls. HIV stock (50 μL) at 100–300 CCID₅₀ (50% cell culture infectious doses) or culture medium was added to either the infected or mock-infected wells of the microtiter tray. Mock-infected cells were used to evaluate the effects of the test compound on uninfected cells to assess the test compounds’ cytotoxicity. Exponentially growing MT-4 cells were centrifuged for 5 min at 220 g, and the supernatant was discarded. The MT-4 cells were resuspended at 6 × 10⁵ cells/mL, and 50 μL volumes were transferred to the microtiter tray wells. Five days after infection, the viability of mock-and HIV-infected cells was examined spectrophotometrically using the MTT assay. The MTT assay is based on the reduction of yellow colored 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Acros Organics) by mitochondrial dehydrogenase activity in metabolically active cells to a blue-purple formazan that can be measured spectrophotometrically. The absorbances were read in an eight-channel computer-controlled photometer (Infinite M1000, Tecan), at two wavelengths (540 and 690 nm). All data were calculated using the median absorbance value of three wells. The 50% cytotoxic concentration (CC₅₀) was defined as the concentration of the test compound that reduced the absorbance (OD540) of the mock-infected control sample by 50%. The concentration achieving 50% protection against the cytopathic effect of the virus in infected cells was defined as the 50% effective concentration (EC₅₀).

5.3. Determination of the mechanism of representative compounds

5.3.1. Cells

Human embryonic kidney 293T (a gift from Dr. Irwin Chaiken, Drexel University, Philadelphia, PA) were cultured in Dulbecco's Modified Eagle's Medium (DMEM), 10% FBS, 100 U/mL penicillin, 100 μg/mL streptomycin and 2 mM L-glutamine. Human astro-glioma U87 cells stably expressing CD4/CXCR4 (obtained from Prof. Hongkui Deng, Peking University, and Prof. Dan Littman, New York University, USA, through the AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH) [51,52] were cultured in DMEM supplemented with 10% FBS, 100 U/ml penicillin, 100 μg/mL streptomycin and 2 mM L-glutamine, 300 μg/mL G418 (Thermo Scientific, Waltham, MA) and 1 μg/mL Puromycin (Thermo Scientific). Cells were incubated continuously, unless otherwise stated, at 37 °C in a humidified 5% CO₂/95% air environment.

5.3.2. Proteins

IgG b12 anti-HIV-1 gp120; was obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH: Anti-HIV-1 gp120 Monoclonal (IgG1 b12) from Dr. Dennis Burton and Carlos Barbas); p24 was produced in-house as previously described [53]. Briefly, a vector containing C-terminally His-tagged H1V-1_NL4-3CA (a gift from Dr. Eric Barklis, Oregon Health and Science University, Portland, OR) was transformed into BL21-Codon Plus (DE3)-RIL Competent Cells (Agilent Technologies, Wilmington, DE) and grown up in autoinduction ZYP-5052 medium overnight with shaking (225 rpm) at 30 °C [54]. Bacterial cultures were spun down at 7000 rpm, and the supernatant was discarded. Cell pellets were resuspended in PBS and lysed via sonication. The resultant supernatant was clarified and immediately applied to a Talon cobalt resin affinity column (Clonetech Laboratories, Mountain View, CA). Protein was eluted using 1X PBS with 250 mM imidazole. Purified CA-H6 monomers were dialyzed overnight into 20 mM Tris-HCl pH 8.0 at 4 °C, concentrated to 120 μM, flash-frozen in liquid nitrogen, aliquoted, and stored at −80 °C. The CA hexamer was generated by introducing mutations at the following sites: A14C, E45C, W184A, and M185A through site-directed mutagenesis (Stratagene). The CA hexamer construct was expressed and purified following the same protocol as described above. After purification, the CA-H6 hexamers were dialyzed into 200 mM ß-ME followed by sequential dialyzes to remove the ß-ME to allow for hexamer assembly slowly.

5.3.3. Peptides

For SPR-based competition experiments, peptides derived from host factors CPSF6 and NUP153 were used. The PCSF6 peptide (313-PVLFPGQPFGQPPLG-327) and NUP153 peptide (1407-TNNSPSGVFTFGANSST-1423) were synthesized by GenScript Corp. (Piscataway, NJ).

5.3.4. Production of pseudotyped viruses

Single-round infectious envelope-pseudotyped luciferase-reporter viruses were produced by dual transfection of two vectors (3 μg of vector 1 and 4 μg of vector 2) in 6-well plated 293T cells (1 × 10⁶ cells/well) [52]. Vector 1 is an envelope-deficient HIV-1 pNL4-3-Luc + R-E plasmid that carries the luciferase-reporter gene [55]. Vector 2 is a plasmid expressing the HIV-1_BG505 gp160 Env [56]. Transfections of these vectors were carried out via calcium phosphate (ProFection Mammalian Transfection System, Promega, Madison, WI) for 5 h. Following the 5 h transfection incubation, DNA-containing medium was removed, cells were washed with DMEM and replenished with fresh culture media. Supernatants containing pseudovirus were collected 72 h post-transfection, clarified, filtered, aliquoted and stored at −80 °C.

5.3.5. SPR direct interaction analysis

All binding assays were performed on a ProteOn XPR36 SPR Protein Interaction Array System (Bio-Rad Laboratories, Hercules, CA). The instrument temperature was set at 25 °C for all kinetic analyses. ProteOn GLH sensor chips were preconditioned with two short pulses each (10 s) of 50 mM NaOH, 100 mM HCl, and 0.5% sodium dodecyl sulfide. Then the system was equilibrated with PBS-T buffer (20 mM sodium phosphate, 150 mM NaCl, and 0.005% polysorbate 20, pH 7.4). The surface of a GLH sensor chip was activated with a 1:100 dilution of a 1:1 mixture of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (0.2 M) and sulfo-N-hydroxysuccinimide (0.05 M). Immediately after chip activation, the HIV-1 NL4-3 capsid protein constructs, purified as described above, were prepared at a concentration of 100 μg/mL in 10 mM sodium acetate, pH 5.0 and injected across ligand flow channels for 5 min at a flow rate of 30 μL/min. Then, after unreacted protein had been washed out, excess active ester groups on the sensor surface were capped by a 5 min injection of 1 M ethanol-amine HCl (pH 8.0) at a flow rate of 5 μL/min. A reference surface was similarly created by immobilizing a non-specific protein (IgG b12 anti-HIV-1 gp120; was obtained through the NIH AIDS Reagent Program, Division of AIDS, NIAID, NIH: Anti-HIV-1 gp120 Monoclonal (IgG1 b12) from Dr. Dennis Burton and Carlos Barbas) and was used as a background to correct non-specific binding. To prepare a compound for direct binding analysis, compound stock solutions, along with 100% DMSO, and totalling 30 μL was made to a final volume of 1 mL by addition of sample preparation buffer (PBS, pH 7.4). Preparation of analyte in this manner ensured that the concentration of DMSO was matched with that of running buffer with 3% DMSO. Serial dilutions were then prepared in the running buffer (PBS, 3% DMSO, 0.005% polysorbate 20, pH 7.4) and injected at a flow rate of 100 μl/min, for a 1 min association phase, followed by up to a 5 min dissociation phase using the “one-shot kinetics” capability of the Proteon instrument. Data were analyzed using the ProteOn Manager Software version 3.0 (Bio-Rad). The responses from the reference flow cell were subtracted to account for the nonspecific binding and injection artefacts. Experimental data were fitted to a simple 1:1 binding model (where applied). The average kinetic (dissociation [k_d] rates) and equilibrium parameters generated from 3 replicates were used to define the off-rates and equilibrium dissociation constant (K_D).

Competition experiments with CPSF6 and NUP153 peptides were performed using the co-analyte function on the ProteOn XPR36 system. Compounds in various concentrations were injected over the CA surface to pre-occupy/saturate the immobilized CA protein. Immediately, and without dissociation phase of the compound, a second injection followed, including identical compound concentrations but in the presence of 100 μM CPSF6 or NUP153 peptides at a flow rate of 100 μL/min, for a 1 min association phase. The response at equilibrium was recorded and compared to the control, which was the peptide without any compound added. The responses from at least 3 replicate injections over each CA surface were analyzed.

5.3.6. Single-round infection (SRI) assay

The single-round HIV-1 infection assay details have been published previously [55,57,58]. Briefly, U87.CD4.CXCR4 (1.2 × 10⁴ cells/well) target cells were seeded in 96-well luminometer-compatible tissue culture plates (Greiner Bio-one). After 24 h compound, DMSO (vehicle control for compounds, Sigma) was mixed with pseudotyped viruses (normalized to p24 content). The mixture was added to the target cells and incubated for 48 h at 37 °C. Following this, the media was removed from each well, and the cells were lysed by the addition of 50 μL/well of luciferase lysis buffer (Promega) and one freeze-thaw cycle. A GloMax 96 microplate luminometer (Promega) was used to measure the luciferase activity of each well after the addition of 50 μL/well of luciferase assay substrate (Promega).

5.3.7. Viral late-stage infection assay

Single-round infectious specific envelope-pseudotyped luciferase-reporter viruses were produced from 293T cells [52] in the presence of 100 μM compound or DMSO (a vehicle control for compounds, Sigma). After 48 h of incubation at 37 °C, the resulting culture supernatants containing pseudotype virus stocks were diluted twenty-fold and then used to infect U87.CD4.CXCR4 target cells. Target cells with or without pseudotyped viruses were incubated for 48 h at 37 °C. Subsequently, the media was removed from each well, and the cells were lysed by the addition of 50 μl of luciferase lysis buffer (Promega) and one freeze-thaw cycle. A GloMax 96 microplate luminometer (Promega) was used to measure the luciferase activity of each well after the addition of 50 μl of luciferase assay substrate (Promega). Compound induced effects are manifested as a decrease in infectivity in the target cells (measured as luciferase activity), normalized against the infectivity of virus produced from DMSO (vehicle control) treated cells.

5.3.8. Docking simulations

The PDB files for Qa-4 were prepared and then energy-minimized using Flare version 4 (Cresset®, Litlington, Cambridgeshire, UK, http://www.cresset-group.com/flare/) with a root mean squared (RMS) gradient cutoff of 0.2 kcal/mol/A and 10000 iterations. The crystal structure of PF-74-bound HIV-1 capsid (PDB code: 4XFZ) was prepared using Flare, version 4 (Cresset®, Litlington, Cambridgeshire, UK, http://www.cresset-group.com/flare/) to allow protonation at pH 7.0 and removal of residue gaps. A dimeric unit of the pre-prepared HIV-1 capsid protein structure was further prepared using Autodock tools [59], where essential hydrogen atoms, Kollman united atom type charges, and solvation parameters were added. The grid box for the docking search was centered around the PF-74 binding site with a spacing grid of 0.375 Å using the Autogrid program. Docking calculations were performed using AutoDock via DockingServer. Q-c4 was further energy minimized using the MMFF94 Force Field method and Gasteiger partial charges added. Non-polar hydrogen atoms were merged, and rotatable bonds were defined. AutoDock parameter set- and distance-dependent dielectric functions were used in the calculation of the van der Waals and the electrostatic terms, respectively. Docking simulations were performed using the Lamarckian genetic algorithm (LGA) and the Solis & Wets local search method [60]. Initial position, orientation, and torsions of the ligand molecules were set randomly. Rotatable torsion angles were released during docking. Each docking experiment was derived from 100 different runs that were set to terminate after a maximum of 2500000 energy evaluations. The population size was set to 150. During the search, a translational step of 0.2 Å, and quaternion and torsion steps of 5 were applied.

5.4. Molecular dynamics simulation of Q-c4

5.4.1. Initial structure preparation

A monomer of the hexameric HIV-1 CA was used in the MD simulation (PDB code 5HGL with a resolution of 3.1 Å) [61]. The monomer structure misses amino acids Ala88 to Gln95 and Lys182 to Ala185. HIV-1 CA 3GV2 structure was used to complete the missing amino acids with their corresponding tertiary structure. The missed amino acids were added to 5HGL structure after their alignment using discovery studio software, and then the whole system was minimized to remove any restraints due to the additional step [62].

Marvin was used to sketch Q-c4 compound in S-configuration and to check its ionization states, Marvin 20.21, ChemAxon (https://www.chemaxon.com). No ionizable groups were detected at physiological pH. Q-c4 conformers were generated by OMEGA module of OPENEYE Scientific Software Inc, and then they were docked to the binding site of 5HGL by OEDOCKING 3.0.1 with chemgauss4 scoring function [9,63-67]. The docked conformer with the highest chemgauss4 was used in the subsequent molecular dynamics simulation.

5.4.2. Molecular dynamics simulation production

ANTECHAMBER module of AMBER14 was used to derive atomic point charges for Q-c4 from the AM1-BCC charge model [68]. Force field ff14sb was used to generate the protein residues and GAFF force field parameters for the Q-c4 molecule. TIP3PBOX octahedral solvent box model with 8 Å cut was used to solvate the complex. Initially, the solvent was minimized for 10000 cycles using the steepest descent and then conjugate gradient algorithms. Then, the whole system was minimized for 5000 cycles using the steepest descent followed by conjugate gradient algorithms. After the minimization procedure, water molecules were equilibrated for 20 ps at constant volume and periodic boundaries. Then, the whole system (water and the complex system) was equilibrated for 40 ps using constant pressure periodic boundaries. Finally, the equilibrated structure was used in 500 ns NPT MD simulation. Non-bonded forces were calculated at a Cutoff distance of 10 Å, and the SHAKE algorithm for hydrogen atoms was turned on.

5.4.3. Clustering

After autoimage all frames of the MD trajectory, water molecules and ions were stripped off. Then, all frames were aligned against the first frame of the MD simulation. All frames were clustered by the DBSCAN algorithm [9,69] on Q-c4 using minimum points of 2.3 and epsilon of 2.1 with no frame orientation, which will cluster all frames according to the Q-c4.

5.5. Metabolic stability in human liver microsomes and human plasma

Details of the analytical method and raw data are given in the Supporting Information.