Discovery and molecular docking studies of nitrogen containing naphthyridine derivatives as potent HIV1 NNRTIs with anticancer potential

Acquired Immunodeficiency Syndrome (AIDS) is caused by the Human Immunodeficiency Virus (HIV), a lentivirus belonging to the Retroviridae family, resulting in immunological failure¹. Human immunodeficiency virus infection is a major cause of adult mortality and morbidity, which in turn affects healthcare and the disintegration of society². Initially, HIV treatment relied on monotherapy, employing only one medication³. In subsequent years, this method had severe negative consequences, including the emergence of drug-resistant virus strains. These resilient strains developed resistance by undergoing mutations and changing their genetic makeup. The diminishing efficacy of monotherapy prompted an investigation of more potent therapies. Consequently, combination therapy emerged, employing a range of antiretroviral medications. This strategy not only minimizes adverse effects but also lowers the risk of treatment resistance, as the virus struggles to adapt to multiple medications simultaneously. Compared with monotherapy, combination therapy effectively slows disease progression and curbs the development of drug resistance^4,5. Consequently, combination therapy is increasingly regarded as superior to monotherapy, providing greater viral suppression and sustainable outcomes⁶. Highly Active Antiretroviral Therapy (HAART) involves the use of multiple drugs for HIV treatment. HAART components consist of HIV protease inhibitors (PIs), reverse transcriptase (RT) inhibitors, fusion inhibitors (FIs), CCR5 antagonists, and integrase strand transfer inhibitors (INSTIs)⁷. HAART has been shown to dramatically reduce both morbidity and mortality in individuals infected with HIV by effectively suppressing the virus and preventing the progression of disease⁸. The reverse transcriptase plays a crucial role in the process of retroviral replication. Nucleoside reverse transcriptase inhibitors (NRTIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) are the two classes of reverse transcriptase inhibitors that are employed to inhibit this process⁸. Non-nucleoside reverse transcriptase inhibitors (NNRTIs) act as allosteric inhibitors of reverse transcriptase, offering higher specificity and lower toxicity compared to nucleoside reverse transcriptase inhibitors (NRTIs)^9,10. NNRTIs suppress HIV-1 reverse transcription by binding to a non-catalytic site of the enzyme. Notably, medications such as efavirenz, dapivirine, and etravirine have been shown to inhibit viral replication by disrupting HIV-1 Gag-Pol polyprotein processing. This process is essential for the subsequent reverse transcription of viral RNA into DNA. Furthermore, pyrimidinedione derivatives, exemplified by IQP-0410, demonstrated potent inhibition of RT-mediated reverse transcription, the critical step responsible for converting the viral RNA genome into double-stranded DNA^11,12. Clinically, efavirenz (EFV), nevirapine (NVP), and delavirdine (DLV) were the first-generation NNRTIs approved by the U.S. Food and Drug Administration (FDA), establishing the foundation for subsequent therapeutic advancements. However, these drugs have a low genetic barrier and are prone to the development of drug-resistant strains^7,9,13,14. To address these limitations, second-generation NNRTIs such as etravirine (ETR) and rilpivirine (RPV) were developed. These compounds exhibit enhanced binding flexibility, allowing them to retain activity against common NNRTI-resistant HIV-1 variants. Moreover, they possess improved pharmacokinetic properties and reduced side effect profiles, making them valuable components of modern combination antiretroviral therapy¹³. Interestingly, emerging studies have shown that certain NNRTIs, such as EFV and RPV, exert off-target effects on host cellular pathways involved in cancer progression. These include modulation of oncogenic signaling (e.g., Akt, STAT3), regulation of apoptotic proteins (e.g., BCL-2 family), and effects on immune responses^15,16,17. In addition, benzene-sulfonamide derivatives were found to effectively inhibit HL-60 cell proliferation, primarily through selective inhibition of the Src/Abl tyrosine kinases¹⁸. The derivatives of pyrazole could inhibit drug-resistant non-small cell lung cancer (EGFRL858R/T790M) compared to osimertinib and induce G0/G1 cell cycle arrest¹⁹. Although unrelated to RT inhibition, these findings suggest potential intrinsic anticancer activity of NNRTIs. These findings highlight the potential for drug repurposing and underscore the importance of developing NNRTIs with better pharmacological properties. While NNRTIs remain central to HIV-1 treatment, the rise of drug-resistant strains continues to limit their effectiveness. In response, ongoing efforts focus on designing novel NNRTIs with enhanced potency, broader resistance coverage, and reduced toxicity. Several structurally diverse scaffolds have shown promise as NNRTIs. For example, indolylarylsulfone derivatives exhibited potent inhibitory activity against both single and double HIV-1 RT mutants, with EC₅₀ values of lower than 0.7 nM for K103N and Y181C, 21.3 nM for Y188L, and 6.2 nM for the K103N/Y181C double mutant²⁰. Similarly, piperazine sulfonyl-substituted diarylpyrimidines demonstrated strong activity against HIV-1 IIIB and RES056 strains, with EC₅₀ values of 0.0014 µM and 0.093 µM, respectively²¹. Likewise, 1,3,5-triazine-based compounds had emerged as promising candidates against the HIV-1 wild-type and the K103N/Y181C-resistant strains, with EC₅₀ values of 8.5 nM and 1.3 µM, respectively²². Building on this structural diversity, quinoline-based compounds have also attracted attention due to their broad-spectrum biological activities^{23,24,25,26,27}. Quinoline, a heterocyclic aromatic scaffold, is associated with diverse biological activities, including anticonvulsant, antimicrobial, and notably, HIV inhibitory effects. Quinoline-based dihydrazone derivatives had demonstrated potent and selective antiproliferative activity across various human cancer cell lines, particularly MCF-7, while exhibiting minimal toxicity toward normal liver cells²⁸. Similarly, quinazoline derivatives represented a prominent class of anticancer agents, exerting their therapeutic effects through inhibition of multiple molecular targets, including breast cancer resistance protein (BCRP)²⁹. In addition, iridium(III) complexes containing 2-(1 H-benzimidazol-2-yl)quinoline had shown significant anticancer activity against breast cancer cells (MDA-MB-231 and MCF-7) by inducing apoptosis via upregulation of Bax and caspase-3 and downregulation of BCL-2, while also effectively suppressing the colony-forming ability of these cells³⁰. Expanding on the antiviral potential of quinoline derivatives, Deo et al. recently reported the synthesis of quinoline-based compounds containing diketo acids, which exhibited notable inhibition of HIV-1 integrase, further highlighting the versatility of the quinoline scaffold in both anticancer and antiviral drug development²³. Moreover, the Forezi group demonstrated the efficacy of reverse transcriptase inhibitors derived from 4-oxoquinoline ribonucleoside derivatives²⁴. Our previous study, Makarasen et al. (2019) reported the invention of novel quinoline derivatives via molecular hybridization across commercialized medications, namely NVP, EFV, and RPV, which were 6-amino-4-oxydiarylquinoline and 2-amino-4-oxydiarylquinoline derivatives, respectively^25,26. These analogues were synthesized, and their biological activities were assessed, including molecular docking studies. The results indicated that 2,4-disubstituted quinolines exhibited a higher percentage of inhibitory activity against HIV-1 RT than 4,6-disubstituted quinolines. In addition, the compound 4-(4’-cyanophenoxy)−2-(4’’-cyanophenyl)-aminoquinoline exhibited a percentage inhibition value of 39.71 against HIV-1 RT, which was lower than the value of 53.39 for NVP. On the other hand, EFV and RPV revealed greater inhibition rates of 96.69% and 97.06%, respectively. Nevertheless, this compound has shown much less cytotoxicity against normal embryonic lung (MRC-5) cells in comparison to EFV and RPV. The molecular docking study revealed that the 2,4-disubstituted quinoline derivatives were able to fit into the HIV-1 RT pocket, but the 4,6-disubstituted quinoline derivatives did not arrange in the same pattern. A hydrogen bond formed between LYS101, a crucial amino acid residue in HIV-1 viral replication, and 2,4-disubstituted quinolines. Therefore, Makarasen et al. conducted a study in 2022 to investigate the impact of substituent groups at 2,4-positions on quinoline derivatives²⁷. The impact of substituent groups on the biological activities of 2,4-disubstituted quinolines was investigated by varying the substituent groups, evaluating the biological activities, and conducting a molecular docking study. The study revealed that 4-(2’,6’-dimethyl-4’-formylphenoxy)−2-(5’’-cyanopyridin-2’’ylamino)quinoline and 4-(2’,6’-dimethyl-4’-cyanophenoxy)−2-(5’’-cyanopyridin-2’’ylamino)quinoline exhibited percentage inhibition values of 44.5% and 45.1%, with IC₅₀ values of 1.93 µM and 1.22 µM against HIV-1 RT, respectively. 4-(2’,6’-dimethyl-4’-formylphenoxy)−2-(5’’-cyanopyridin-2’’ylamino)quinoline displayed significant cytotoxicity against acute lymphoblastic leukemia cells (MOLT-3), cervical carcinoma cells (HeLa), and promyeloblast cells (HL-60), with demonstrated IC₅₀ values of 12.7, 25.7, and 20.5 µM, respectively. However, both compounds showed minimal biological activity against MRC-5. Molecular docking analysis also revealed that both compounds formed hydrogen bonds with LYS101, HIS235, and PRO236, and interacted via π-π stacking with TYR188, TRP229, and TYR318 in HIV-1 RT. A previous investigation on molecular docking and biological activity against HIV-1 RT indicated that increasing the nitrogen atoms on the substituent group of quinoline at the 2-position, specifically the pyridinyl group, could improve the inhibition of HIV-1 RT for the derivatives of 2,4-disubstituted quinolines. The current research involved the invention of novel core structures, namely 1,6-naphthyridine and 1,7-naphthyridine, as alternatives to quinolines (Fig.��1). The objective was to enhance the polarity, solubility, and interaction by introducing nitrogen atoms onto the quinoline core structure. The biological activities of the 2,4-disubstituted-1,6-naphthyridine and 2,4-disubstituted-1,7-naphthyridine derivatives were investigated. Molecular docking and molecular dynamics analysis were employed to determine and validate the binding energy, binding site, stability, and interaction between the designed compounds and the amino acid residues of HIV-1 RT. The objective was to ascertain these compounds as prospective inhibitors for HIV-1 RT.

The core structure of 2-amino-4-phenoxy quinoline derivatives (a), 2-amino-4-phenoxy-1,6-naphthyridine derivatives (b) and 2-amino-4-phenoxy-1,7-naphthyridine derivatives (c).

Full size image

Synthesis

In this study, the derivatives of 1,6-naphthyridine and 1,7-naphthyridine with the substituent groups at 2,4-positions were synthesized and evaluated for their biological activities, including the molecular docking study, in order to increase the potency of HIV-1 RT inhibition. For our synthetic route, the 2-phenylamino-4-phenyloxy-1,6- and 1,7-naphthyridines and the 2,4-diphenyloxy-1,6- and 1,7-naphthyridines would be achieved from nucleophilic substitution reactions of 2,4-dichloronaphthyridine intermediates 7 and 8. In addition, these intermediates can be synthesized from analogues of nicotinate via a similar synthetic route according to the literature review³¹. Firstly, the synthesis of 2,4-dichloro-1,6-naphthyridine (7) was started from commercially available methyl 4-aminonicotinate (1) in 3 steps, as shown in Fig. 2. At the beginning, acetylation of amine 1 with acetic anhydride (Ac₂O) provided acetamide 3 in quantitative yield, which was used in the next step without further purification. In addition, the cyclization to form the naphthyridine ring of acetamide 3 using lithium bis(trimethylsilyl)amide (LiHMDS) gave the 2,4-dione-1,6-naphthyridine intermediate, which was tautomerized to provide 2,4-dihydroxy-1,6-naphthyridine (5) in quantitative yield. This compound was used in the next step without further purification. Finally, the chlorination of dihydroxynaphthyridine 5 using phosphorus oxychloride (POCl₃) established dichloronaphthyridine 7 in 50% yield over 2 steps from acetamide 3. Lastly, the 2,4-dichloro-1,7-naphthyridine (8) was synthesized under similar conditions to 2,4-dichloro-1,6-naphthyridine (7) using ethyl 3-aminoisonicotinate (2) as a starting material. This synthesis yielded dichloronaphthyridine 8 in 41% yield over 3 steps from amino 2.

Syntheses of 2,4-dichloro-1,6-naphthyridine (7) and 2,4-dichloro-1,7-naphthyridine (8). Reagents and conditions: (a) Ac₂O, 1,4-dioxane, rt; (b) LiHMDS, dry THF, 0 °C; (c) POCl₃, 120 °C, sealed tube.

Full size image

With 2,4-dichloro-naphthyridines 7 and 8 in hand, the synthesis of 2,4-diphenyloxy- and 2-phenylamino-4-phenoxy-1,6- and 1,7-naphthyridine derivatives was described in Fig. 3. From our report in 2022, 4-hydroxy-3,5-dimethylbenzaldehyde (a) and 4-hydroxy-3,5-dimethylbenzonitrile (b) as phenoxy substituent groups at the 4-position on quinolines can enhance the inhibition potency against HIV-1 RT²⁷. Thus, we selected these substituents for our syntheses of the 1,6- and 1,7-naphthyridine core structures. The nucleophilic substitution reaction of dichloronaphthyridine 7 with phenoxy groups a and b under basic conditions at room temperature provided 2-chloro-4-phenoxy-1,6-naphthyridines 9a and 9b in 49% and 48%, respectively, as shown in Table 1. For 2-chloro-4-phenoxy-1,7-naphthyridines, 10a and 10b were synthesized from dichloronaphthyridine 8 via the similar conditions of 9a and 9b in 45% and 86%, respectively. Furthermore, in the syntheses of 9a and 9b, 4-chloro-2-phenoxy-1,6-naphthyridines 11a and 11b were formed as minor products in 23% and 21%, respectively. On the other hand, the 2-phenoxy side products 12a and 12b were observed as trace amounts from the syntheses of 1,7-naphthyridine intermediates 10a and 10b under the same reaction conditions. Interestingly, 2,4-diphenoxy-naphthyridines 13(a-b) and 14(a-b) can be synthesized via nucleophilic substitution under basic conditions at high temperature (120 °C). 2,4-Diphenoxy-1,6-naphthyridines 13a and 13b were obtained in 89% and 78% from 1,6-naphthyridine 7, respectively, while 2,4-diphenoxy-1,7-naphthyridines 14a and 14b were established from 1,7-naphthyridine 8 in 86% and 90%, respectively. As previously mentioned, the nitrogen-containing scaffolds assisted on the quinoline core structure were utilized from our work in 2022. Therefore, 4-aminobenzonitrile and 2-amino-5-cyanopyridine were selected as the 2-phenylamino groups of the desired naphthyridine derivatives. For nucleophilic substitution of 2-phenylamino, the Buchwald-Hartwig reaction of 1,6-naphthyridines 9(a-b) and 1,7-naphthyridines 10(a-b) with 4-aminobenzonitrile established 2-phenylamino-4-phenoxy-naphthyridines 15(a-b) and 17(a-b) in 32–70%, as demonstrated in Fig. 3. In addition, naphthyridine derivatives 16(a-b) and 18(a-b) were achieved from 2-chloro-naphthyridines 9(a-b) and 10(a-b) with 2-amino-5-cyanopyridine via the same reaction in 54–89%. Finally, with 4-formylphenoxy-naphthyridines 16a and 18a in hand, 2-phenylamino-4-cyanovinylphenoxy-naphthyridines 19a and 20a were accomplished, as shown in Fig. 4, via the Horner-Wadsworth-Emmons reaction with diethyl cyanomethyl phosphonate under basic conditions²¹. The cyanovinyl 19a and 20a were synthesized in 29% and 33% from 16a to 18a, respectively, which consisted of E:Z-isomers in a ratio of 1:1.55 and 1:0.87, respectively. In summary, the designed 1,6- and 1,7-naphthyridine derivatives were established from 4-aminonicotinate (1) and 3-aminoisonicotinate (2), respectively, with 4–6 steps in 7–45% overall yields. The substituent groups at the 2- and 4-positions of 1,6- and 1,7-naphthyridine derivatives are summarized in Table 1. The biological activities of these 14 naphthyridine analogues against HIV-1 RT were evaluated, accompanied by molecular docking studies to investigate their binding modes.

Syntheses of naphthyridine derivatives 9(a-b)-17(a-b)²⁷ Reagents and conditions: (a) Ar¹OH, Cs₂CO₃, DMF, rt; (b) Ar¹OH, Cs₂CO₃, DMF, 120 °C, sealed tube; (c) Ar²NH₂, Pd(OAc)₂, Ligands (PhDavePhos, SPhos and Xantphos), Cs₂CO₃, DMF, 120 °C, sealed tube.

Full size image

Syntheses of 2-cyanopyridinylamino-4-cyanovinylphenoxy-naphthyridine derivatives 19a and 20a. Reagents and conditions: (a) (EtO)₂P(O)CH₂CN, t-BuOK, THF, 0 °C.

Full size image

HIV-1 RT inhibition assay

To evaluate the biological activities of the targeted naphthyridines, the percentage inhibitory activities against HIV-1 RT were utilized at a 1 µM concentration using a reverse transcriptase assay with a colorimetric method. The results are illustrated in Fig. 5, with percentage inhibitory activities indicated in parentheses. 2-Phenylamino-4-phenoxy-naphthyridines 15–18(a-b) and 19–20(a) revealed greater rates ranging from 23% to 84%, whereas 2,4-diphenoxy-naphthyridines 13(a-b) and 14(a-b) exhibited low to moderate inhibition rates of 17% to 32%. This result corresponds with the previous study of Makarasen et al. (2022), wherein 2-phenylamino-4-phenoxy-quinolines exhibited a greater percentage of inhibition compared to 2,4-diphenoxy-quinolines²⁷. The cyanopyridinyl group and 2,4-dimethyphenyl group are crucial substituents of 2,4-disubstituted quinolines, located at the second and fourth positions, respectively, as demonstrated through the inhibition rates of 44% and 45% against HIV-1 RT for 4-(2’,6’-dimethyl-4’-formylphenoxy)−2-(5’’-cyanopyridin-2’’ylamino)quinoline and 4-(2’,6’-dimethyl-4’-cyanophenoxy)−2-(5’’-cyanopyridin-2’’ylamino)quinoline, respectively²⁷. The current study indicated that the novel 2,4-disubstituted-1,6- and 1,7-naphthyridines 16(a-b) and 18(a-b), featuring analogous substituent groups as reported previously along with a cyanopyridinyl group at the 2-position of naphthyridine, exhibited high percentage inhibition values ranging from 58% to 84%. In contrast, the 2-cyanophenylamino-4-phenyloxy-1,6- and 1,7-naphthyridines 15(a-b) and 17(a-b) indicated lower percentage inhibitory values of 23% to 35%, respectively, which corresponded with prior studies. Additionally, the effect of the substituent group at the fourth position of naphthyridine, specifically dimethylformylphenoxy and dimethylcyanophenoxy, was examined. It was determined that these substituent groups exhibited comparable efficacy in inhibiting HIV-1 reverse transcriptase when compared to the same substituents at the second position. This evidence indicates that the 1,6- and 1,7-naphthyridine core structures could enhance the efficacy of HIV-1 reverse transcriptase inhibition more than quinoline core structures compared with the previous report. Interestingly, the 1,6- and 1,7-naphthyridine derivatives demonstrated differing inhibitory effects on HIV-1 reverse transcriptase (RT). Cyanopyridinylamino-substituted 1,6-naphthyridines (16a, 16b, and 19a) exhibited markedly greater inhibitory potency than their 1,7-naphthyridine counterparts (18a, 18b, and 20a). Consequently, compounds 16(a-b), 18(a-b), 19a, and 20a were selected for further examination of their IC₅₀ values. As summarized in Table 2, all tested naphthyridine analogues showed IC₅₀ values ranging from 0.175 to 0.634 µM, which were lower than those of two compounds in the previous publication and the reference compound nevirapine (NVP, IC₅₀ = 1.053 µM). Notably, the cyanovinyl-substituted compound 19a exhibited the most potent inhibitory activity among the tested derivatives. It is noteworthy that the IC₅₀ values of the synthesized naphthyridines exceeded those of efavirenz (EFV, IC₅₀ = 0.058 µM) and rilpivirine (RPV, IC₅₀ = 0.063 µM). These findings suggest that the novel 2,4-disubstituted 1,6- and 1,7-naphthyridine cores may enhance HIV-1 RT inhibition compared to the corresponding 2,4-disubstituted quinoline analogues. In particular, the 2-cyanopyridinylamino-1,6-naphthyridine scaffold represents a promising new structural framework for the development of potent HIV-1 RT inhibitors. To gain a deeper insight into the observed variations in biological activity, molecular docking was performed to examine the binding interactions and binding energies of each molecular structure with the active site of HIV-1 reverse transcriptase, as well as the strength of their binding.

Structures and Inhibitory activity (%) of 13–18(a-b), 19a, 20a, and commercial medicines at 1 µM against HIV-1 RT^a.

Full size image

Molecular docking studies

The molecular docking investigation examined the binding interactions of commercial medicines (NVP, EFV, ETR, and RPV) with the synthesized naphthyridine derivatives (13–18(a-b), 19a, and 20a) within the NNRTIs pocket of HIV-1 RT, as illustrated in Fig. 6; Table 3. These compounds interacted with amino acid residues through conventional hydrogen bonding and hydrophobic interactions within a 3.0 Å radius. The synthesized naphthyridines exhibited binding conformations that closely matched those of the commercial medicines within the HIV-1 RT pocket. A comparison among structurally related compounds provided insights into their potential as HIV‑1 RT inhibitors.

In the comparison between compounds 13(a-b) and 14(a-b), compounds 13a and 13b exhibited significant binding affinities for HIV‑1 RT, with binding energies of −13.66 and − 13.98 kcal/mol, respectively. These two compounds formed hydrogen bonds with VAL106 and TYR318, in addition to π-π stacking interactions with TYR188 and TRP229. Significantly, 13b demonstrated the lowest binding energy (−13.98 kcal/mol), indicating a particularly strong interaction. Conversely, 14a and 14b showed slightly decreased binding affinities, with binding energies of −13.17 and − 13.64 kcal/mol, respectively. 14a established conventional hydrogen bonds with VAL106 and TYR318 and participated in π-π interactions with TYR181, TYR188, PHE227, and TRP229. Compound 14b formed a hydrogen bond with PHE227 and exhibited π-π stacking interactions with TYR188 and TYR318. The results suggested that 13b and 14b demonstrated the most consistent and sustained interaction with HIV‑1 RT among the compared derivatives.

In the comparison between compounds 15(a-b) and 17(a-b), compounds 15a and 15b showed moderate binding affinities, with binding energies of −13.05 and − 13.39 kcal/mol, respectively. 15a formed conventional hydrogen bonds with LYS101 and PHE227 and engaged in π-π stacking interactions with TYR181, TYR188, and TRP229. In comparison, 15b established conventional hydrogen bonds with LYS101 and PHE227, although it exhibited π-π stacking only with PHE227. Conversely, 17a and 17b displayed enhanced binding affinities of −13.51 and − 13.63 kcal/mol, respectively. Both compounds formed conventional hydrogen bonds with LYS101 and PHE227, as well as π-π stacking interactions with TYR181, PHE227, and TYR318. The results indicate that compounds 17a and 17b revealed greater binding stability to HIV‑1 RT compared to compounds 15a and 15b.

In the comparison between compounds 16(a-b) and 18(a-b), compounds 16a and 16b displayed binding energies of −13.45 and − 13.61 kcal/mol, respectively. Both compounds formed conventional hydrogen bonds with LYS101 and PHE227. Additionally, 16a engaged in π–π stacking interactions with TYR181 and TRP229, whereas 16b established π-π interactions with TYR181, PHE227, and TRP229. In contrast, 18a and 18b showed slightly weaker binding affinities, with binding energies of −13.14 and − 13.28 kcal/mol, respectively. Both compounds formed conventional hydrogen bonds with LYS101 and PHE227, as well as π-π stacking interactions with TYR181, PHE227, and TRP229. The results demonstrate that 16a and 16b exhibited greater binding affinities and more stable interactions with HIV‑1 RT compared to 18a and 18b.

Among all tested compounds, 19a has the highest binding affinity, with a binding energy of −14.54 kcal/mol. It formed conventional hydrogen bonds with LYS101, TYR188, and PHE227 and participated in π-π stacking interactions with TYR181 and TRP229. In comparison, 20a displayed a significant binding affinity (−14.20 kcal/mol), establishing conventional hydrogen bonds with the identical residues as 19a. Furthermore, 20a exhibited expanded π–π stacking with TYR188. The lower binding energy of 19a suggests a more favorable binding orientation within the HIV-1 RT pocket.

Compared to the reference medications (NVP, EFV, ETR, and RPV), compounds 16a, 16b, 19a, and 20a had superior binding affinities for HIV-1 RT. Among the reference medicines, RPV exhibited the strongest binding (−13.25 kcal/mol), followed by ETR (−11.75 kcal/mol), EFV (−9.25 kcal/mol), and NVP (−8.39 kcal/mol). All reference medicines formed conventional hydrogen bonds with LYS101. RPV displayed π-π stacking interactions with TYR181 and TRP229, whereas ETR interacted with TYR181, TYR188, and TYR318. Conversely, NVP and EFV exhibited an absence of π-π stacking and formed only limited conventional hydrogen bonds. Notably, 19a and 20a demonstrated the strongest binding affinities among all tested compounds, while 16a and 16b also exceeded RPV, suggesting their potential as promising HIV-1 RT inhibitors. These findings are further supported by 2D binding interaction diagrams and quantitative data on hydrogen bond and π–π stacking distances, as presented in the Supplementary Information. The experimental results confirmed that all tested compounds bound to the active site of HIV-1 RT, involving key amino acid residues such as LYS101, TYR181, TYR188, PRO225, PHE227, TRP229, and TYR318, which play crucial roles in enzyme inhibition³².

The molecular docking results aligned with biological activity data. 19a had the highest inhibition rate and the lowest IC₅₀ value among all synthesized derivatives. Compounds 16a, 16b, 19a, and RPV demonstrated significant inhibition of HIV-1 RT, with inhibition rates of 83.50%, 77.25%, 84.86%, and 97.06% and IC₅₀ values of 0.222, 0.218, 0.175, and 0.063 µM, respectively. Following these findings, compounds 16a, 16b, 19a, and RPV were selected for subsequent molecular dynamics (MD) simulations to evaluate the stability of their complexes with HIV-1 RT.

The overlay of the conformations of the naphthyridine derivatives, namely 13–18(a-b), 19a and 20a and commercially available drugs (NVP, EFV, ETR and RPV) within the binding pocket of HIV-1 RT.

Full size image

Molecular dynamics simulations

Molecular dynamics (MD) simulations were conducted to assess the binding stability of selected ligands and their impact on the conformational dynamics of HIV-1 RT in an explicit solvent environment^32,33,34,35. Four ligands (RPV, 16a, 16b, and 19a) were chosen based on their molecular docking results and biological efficacy, and were analyzed using their interaction with the HIV-1 RT active site. Figure 7 illustrates the evaluation of key MD parameters, such as root mean square deviation (RMSD), root mean square fluctuation (RMSF), and radius of gyration (Rg), over a 300 ns simulation. RMSD was employed to assess the structural stability and convergence of the protein-ligand complexes throughout the simulation. Figure 7A demonstrates that all ligand-bound complexes (HIV-1 RT-RPV, HIV-1 RT-16a, HIV-1 RT-16b, and HIV-1 RT-19a) exhibited reduced RMSD values compared to the HIV-1 RT-free, suggesting that ligand binding enhanced structural stability. The HIV-1 RT-16b complex had the lowest RMSD across all frames, indicating stronger and more persistent binding interactions compared to the other complexes. In addition, the protein structure’s compactness was assessed by determining the radius of gyration (Rg) (Fig. 7B). All ligand-bound systems exhibited reduced Rg values relative to the HIV-1 RT-free, suggesting that ligand binding promoted a more compact and stable protein conformation. The Rg values of the HIV-1 RT-16b and HIV-1 RT-19a complexes were notably lower than those of the HIV-1 RT-16a complex. The results suggest that the 16b and 19a ligands facilitate the most compact protein structures, validating their enhanced binding affinity and stabilizing effects. RMSF analysis was performed to investigate the flexibility of individual amino acid residues during the simulation. The RMSF values (Fig. 7C) revealed that the HIV-1 RT-RPV, HIV-1 RT-16b, and HIV-1 RT-19a complexes exhibited reduced fluctuations at key residues, including PRO95, LYS101, LYS103, VAL106, VAL179, TYR181, TYR183, TYR188, PRO225, PRO226, PHE227, and TYR318, in comparison with both the HIV-1 RT-free and the HIV-1 RT-16a complex. The RMSF profile of HIV-1 RT-16a closely resembled that of the unbound form, suggesting weakened stability following ligand binding.

The analyses of MD simulations. (A) RMSD of protein backbone of HIV-1 RT-free and the HIV-1 RT complexes (RPV, 16a, 16b, and 19a). (B) Rg graphs of HIV-1 RT-free and the HIV-1 RT complexes (RPV, 16a, 16b, and 19a). (C) RMSF of protein backbone of HIV-1 RT-free and the HIV-1 RT complexes (RPV, 16a, 16b, and 19a).

Full size image

Pharmacokinetics studies

Pharmacokinetics defines the processes of drug absorption, distribution, metabolism, and excretion (ADME) in the body, providing essential information for the optimization of therapeutic regimens. The partition coefficient (logP) is a crucial metric for evaluating drug-likeness and pharmacological potential, offering a rapid and effective prediction of membrane permeability and systemic distribution³⁶. Lipinski’s Rule of Five is commonly employed to assess the oral bioavailability of substances. This rule stipulates that drug-like molecules generally meet the following criteria: a maximum of 5 hydrogen bond donors, a maximum of 10 hydrogen bond acceptors, a molecular weight below 500 Da, and a logP value not greater than 5. A logP value beyond 5 suggests inadequate water solubility and reduced oral bioavailability. Parameters influencing oral bioavailability include no more than 10 rotatable bonds, as an excessive amount may reduce membrane permeability and medication absorption, alongside a topological polar surface area (TPSA) of less than 140 Ų. Furthermore, a TPSA below 90 Ų typically suggests a probability of blood–brain barrier permeability^37,38. The molecular characteristics of the synthesized naphthyridine derivatives (13–18(a-b), 19a, and 20a) and the reference medications (NVP, EFV, and RPV) were computed utilizing SwissADME software³⁹ and are presented in Table 4. All synthesized compounds adhered to Lipinski’s Rule of Five, and all compounds exhibited TPSA values below 140 Ų, suggesting good favorable membrane permeability potential.

Molecular dynamics simulations indicated that RPV had the lowest RMSF values among all residues, implying the most stability at the local residue level. The RMSD and Rg values of the HIV-1 RT-16b and HIV-1 RT-19a complexes were lower than those of the HIV-1 RT-RPV complex, signifying enhanced structural stability and compactness. Moreover, RPV possesses a reduced number of hydrogen bond acceptors and an increased number of hydrogen bond donors relative to the synthesized naphthyridine derivatives, possibly enhancing its binding affinity⁴⁰. Nonetheless, the logP values of RPV and the naphthyridine derivatives exhibit similarity. According to their inhibitory efficacy against HIV-1 RT, 16a, 16b, and 19a exhibit drug-likeness profiles comparable to RPV, suggesting that they likely possess favorable pharmacokinetic characteristics.

Cytotoxic activity

In recent years, NNRTI drugs have demonstrated inhibitory activity not only against HIV-1 but also, to a lesser extent, against certain types of cancer. Based on this dual potential, the present study evaluated the cytotoxic effects of newly synthesized compounds (13–18(a-b), 19a, 20a) in comparison with reference drugs including nevirapine, efavirenz, rilpivirine, doxorubicin, and etoposide against selected cancer cell lines. The panel of cancer cell lines comprised four human cancer cell lines, for example, HepG2 (hepatocellular carcinoma), MOLT-3 (acute lymphoblastic leukemia), HuCCA-1 (cholangiocarcinoma), and A549 (lung carcinoma), as well as six resistant cancer cell lines, including MDA-MB-231 (hormone-independent breast cancer), S102 (Thai liver cancer), HeLa (cervical cancer), T47-D (hormone-dependent breast cancer), H69AR (multidrug-resistant lung cancer), and HL-60 (promyelocytic leukemia). A normal embryonic lung fibroblast cell line, MRC-5, was also included to assess selective cytotoxicity. The cytotoxicity results of the synthetic compounds (13–18(a-b), 19a, 20a) and the reference drugs are demonstrated in Tables 5 and 6.

Cytotoxic evaluation of compounds 13–14(a-b)

Compounds 13a and 13b demonstrated a broad range of cytotoxic effects across various human cancer cell lines, revealing significant variations in potency and activity spectrum. Compound 13a exhibited very potent cytotoxicity (IC₅₀ < 20 µM) against MOLT-3 (11.3 µM) and HL-60 (17.1 µM), together with moderate activity (IC₅₀ = 20–50 µM) toward HeLa (24.1 µM), T47-D (35.9 µM), and HuCCA-1 (42.2 µM). Compound 13b also showed significant activity against MOLT-3 (11.3 µM) and moderate activity against HeLa (24.4 µM), T47-D (34.1 µM), HuCCA-1 (32.4 µM), and A549 (36.1 µM), indicating a slightly broader cytotoxic profile compared to 13a. Compound 14a displayed strong cytotoxic effects across multiple cell lines, including MOLT-3 (9.6 µM), HL-60 (10.4 µM), and HeLa (12.2 µM). Moderate activity was observed against T47-D (26.1 µM) and HuCCA-1 (21.7 µM). In contrast, compound 14b showed no measurable cytotoxicity across all tested cell lines. Results are expressed as percentage cytotoxicity at 50 µg/mL (values in parentheses), indicating the effect observed at the highest tested concentration. IC₅₀ values were expressed to µM, based on the highest tested concentration of 50 µg/mL. The investigation of toxicity on the normal embryonic lung fibroblast cell line (MRC-5) revealed that only 13a and 14a produced quantifiable IC₅₀ values of 68.1 µM and 65.3 µM, respectively, which are considered as exhibiting very low toxicity to normal cells. In contrast, both 13b and 14b showed only percentage cytotoxicity values, indicating that these compounds were non-toxic to normal cells at the tested concentration.

Cytotoxic evaluation of compounds 15 and 17(a-b)

Compound 15a showed considerable activity against MOLT-3 (17.6 µM) and HL-60 (18.2 µM), as well as moderate cytotoxicity against HeLa (31.6 µM), T47-D (42.1 µM), and A549 (45.9 µM). Compound 15b did not show significant activity, as all values were provided as percentage cytotoxicity. Compound 17a showed high activity against MOLT-3 (9.1 µM), HL-60 (8.9 µM), and HeLa (13.2 µM), whereas 17b was essentially inactive. For MRC-5 toxicity, 15 and 17(a-b) exhibited no significant cytotoxicity across all tested cell lines, with the data reported as percentage cytotoxicity at the substance concentration of 50 µg/mL.

Cytotoxic evaluation of compounds 16 and 18(a-b), 19a, and 20a

Compound 16a exhibited significant cytotoxicity against MOLT-3 (17.4 µM) and moderate cytotoxicity against HL-60 (41.1 µM), whereas 16b displayed minimal activity, with most of the cell lines recorded as percentage cytotoxicity. Compounds 18(a-b) demonstrated varying effects. 18a showed moderate activity against MOLT-3 with an IC₅₀ of 14.9 µM and against HepG2 with an IC₅₀ of 23.8 µM, while demonstrating inactivity in other tested cell lines. 18b did not exhibit IC₅₀ values in any cell line and was considered inactive overall. Compound 19a exhibited consistent inactivity across all cancer cell lines, whereas compound 20a demonstrated notable efficacy, showing strong activity against HuCCA-1 (5.6 µM), HepG2 (9.6 µM), and MDA-MB-231 (16.4 µM), along with moderate activity in A549 (26.0 µM). The data for MRC-5 for all compounds, with the exception of 18a and 20a, indicated percentage cytotoxicity, suggesting no toxicity to normal cells.

Comparative analysis of compounds 16(a-b) and 19a with reference drugs (NVP, EFV, RPV)

The reference medicines, namely NVP, EFV, and RPV, which are well known for their anti-HIV-1 reverse transcriptase (RT) activity, were tested for cytotoxicity. RPV revealed high cytotoxicity against MOLT-3 (4.3 µM), HeLa (11.3 µM), HL-60 (11.5 µM), and T47-D (15.1 µM), while EFV exhibited moderate activity against HL-60 (33.7 µM) and HeLa (46.2 µM). NVP was largely inactive, with all data reported as percentage cytotoxicity. Compounds 16a, 16b, and 19a are particularly interesting due to their strong inhibitory activity against HIV-1 RT. In the cytotoxic examination, it was found that 16a showed moderate activity against MOLT-3 (17.4 µM) and HL-60 (41.1 µM), but had no toxicity toward MRC-5. 16b displayed minimal activity and no toxicity. 19a was completely inactive across all cancer cell lines and was non-toxic to MRC-5. When compared to NVP, EFV, and RPV, which shared the same molecular background as RT inhibitors, 16a demonstrated a more selective and favorable cytotoxic profile in hematological cancers, with minimal off-target toxicity. While RPV remains the most potent among the reference compounds, the dual activity of 16a as an RT inhibitor and as a selective cytotoxic agent in certain cancer types suggests promising therapeutic potential.

Selectivity index (SI) of compounds (13–18(a-b), 19a, 20a) and the reference drugs (NVP, EFV, RPV)

The selectivity index (SI) is widely used to evaluate the safety margin of bioactive compounds. SI values above 3 are considered indicative of selective anticancer potential in preliminary screening. From the cytotoxicity evaluation of compounds 13–18(a-b), 19a, and 20a in comparison with the reference drugs (NVP, EFV, and RPV), most of our compounds exhibited significant cytotoxic effects against cancer cell lines while being non-toxic toward normal human lung fibroblasts (MRC-5). Tables 5 and 6 show that a considerable fraction of the naphthyridine derivatives exhibited a good selectivity index (SI > 3) across a broad range of tested cancer cell lines. Against MOLT-3 cells, compounds 13(a-b), 14–18(a), and EFV exhibited SI values of 6.03–13.93, while RPV displayed very high selectivity (SI = 31.81). Against HL-60 cells, compounds 13–17(a) and RPV showed SI values of 3.08–14.25. For HeLa cells, compounds 13(b), 14(a), 15(a), 17(a), and RPV exhibited SI values of 4.01–12.11. Notably, compound 13b demonstrated consistent selectivity across five cancer cell lines (A549, T47-D, HuCCA-1, HeLa, and MOLT-3) with SI values of 3.29–10.52, whereas compound 17a displayed outstanding selectivity against MOLT-3 and HL-60, with SI values of 13.93 and 14.25, respectively. Moreover, RPV exhibited the broadest selectivity profile, with SI values of 5.41–31.81 across MOLT-3, MDA-MB-231, S102, HeLa, T47-D, and HL-60 cells. Collectively, these findings highlight the promise of the naphthyridine derivatives as potential anticancer agents.

Materials

The starting materials and reagents were purchased from Aldrich Company and Tokyo Chemical Industry. These materials and reagents were used without purification. The melting point was received from the SMP3 Stuart™ digital melting point apparatus from Bibby Sterlin, Ltd. (Stone, United Kingdom). All of the synthesized compounds were characterized by nuclear magnetic resonance (NMR) and mass spectrometry (MS). NMR spectra were recorded on Bruker Avance III HD 300 and 400 spectrometers. The 300 MHz instrument was used to obtain ¹H (300 MHz) and ¹³C (75 MHz) spectra, whereas the 400 MHz instrument was used to acquire ¹H (400 MHz) and ¹³C (100 MHz) spectra. High-resolution mass spectra data were received from a micrOTOF electrospray ionization time-of-flight mass spectrometer (Bruker Daltonics, Germany). HepG2 (hepatocarcinoma), HuCCA-1 (cholangiocarcinoma), MOLT-3 (acute lymphoblastic leukemia), H69AR (lung cancer, multidrug resistant), MRC-5 (normal embryonic lung cell), A549 (lung carcinoma), MDA-MB-231 (hormone-independent breast cancer), S102 (Thai liver cancer), HeLa (cervical carcinoma), T47-D (hormone-dependent breast cancer), and HL-60 cell lines (promyeloblast) were purchased from Hyclone Laboratories. Etoposide (purity 98%) and Doxorubicin (purity 98%) were purchased from Aldrich Company.

Synthesis

Procedure for the preparation of Methyl 4-acetamidonicotinate (3)

To a solution of methyl 4-aminonicotinate (5.37 g, 35.3 mmol) in 1,4-dioxane (72.0 mL, 0.50 M) was added acetic anhydride (17.0 mL, 5.00 equiv), and it was stirred at room temperature for 41 h. The reaction mixture was concentrated in vacuo before the water was added. The crude product was neutralized by saturated sodium hydrogen carbonate and was extracted thrice with EtOAc. The combined organic layers were washed with brine and dried over with anhydrous sodium sulfate. The organic layers were concentrated in vacuo. The crude residue was purified by column chromatography with 40% of EtOAc in Hexanes to obtain 3 with quantitative yield (6.84 g) as a white solid; m.p. 130.3–131.6 °C; ¹H NMR (400 MHz, CDCl₃) δ 11.12 (s, 1H), 9.15 (s, 1H), 8.61 (s, 2 H), 3.98 (s, 3 H), 2.27 (s, 3 H); ¹³C NMR (100 MHz, CDCl₃) δ 169.7, 168.1, 154.7, 152.5, 147.6, 113.6, 110.3, 52.6, 25.5 ppm; HRMS (ESI) m/z calcd for C₉H₁₁N₂O₃ (M + H)⁺ 195.0764, found 195.0765.

Procedure for the preparation of Ethyl 3-acetamidoisonicotinate (4)

To a solution of ethyl 3-aminonicotinate (1.52 g, 9.16 mmol) in 1,4-dioxane (19.0 mL, 0.50 M) was added acetic anhydride (9.00 mL, 10.0 equiv), and it was stirred at room temperature for 12 h. The 50 mL of cool water was added to the reaction mixture and neutralized by saturated sodium hydrogen carbonate. The reaction mixture was extracted thrice with EtOAc. The combined organic layers were washed with water and brine and dried over with anhydrous sodium sulfate. The organic layers were concentrated in vacuo. The crude residue was purified by column chromatography with 60% of EtOAc in Hexanes to obtain 4 (1.88 g, 99% yield) as a brown solid; m.p. 62.3–63.9 °C; ¹H NMR (300 MHz, CDCl₃) δ 10.65 (s, 1H), 10.00 (s, 1H), 8.42 (d, J = 5.1 Hz, 1H), 7.79 (d, J = 5.1 Hz, 1H), 4.43 (q, J = 7.2 Hz, 2 H), 2.27 (s, 3 H), 1.44 (t, J = 7.2 Hz, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 168.7, 166.9, 143.8, 143.6, 136.2, 122.5, 120.8, 62.2, 25.0, 14.0 ppm; HRMS (ESI) m/z calcd for C₁₀H₁₃N₂O₃ (M + H)⁺ 209.0921, found 209.0927.

Procedure for the preparation of 2,4-dichloro-1,6-naphthyridine (7) and 2,4-dichloro-1,7-naphthyridine (8)

To a solution of N-acetyl nicotinate (3 or 4) in THF (0.02 M) was cooled at −15 °C, and LiHMDS (1.00 M in THF, 3.00 equiv) was slowly added. The reaction mixture was stirred at −15 °C for 3 h before being stirred at 0 °C to room temperature overnight. The suspension was concentrated in vacuo at 30 °C before the water was added. The crude product was neutralized by 10% HCl and concentrated in vacuo. The dihydroxy quinoline intermediate 5 or 6 was obtained with quantitative yield and was used without purification. POCl₃ was added to a solution of 5 or 6 (15.0 mL), and the solution was refluxed in a sealed tube at 120 °C overnight. The reaction mixture was cooled to room temperature and was basified with saturated sodium hydrogen carbonate and 10 M sodium hydroxide to pH 10−11. The aqueous phase was extracted thrice with DCM. The combined organic layers were washed with brine, dried over with anhydrous sodium sulfate, and concentrated in vacuo. The crude residue was purified by column chromatography with 40% of EtOAc in Hexanes to obtain the 2,4-dichloro-naphthyridine derivative as 7 with 8 in 50% and 41% yield, respectively.

2,4-dichloro-1,6-naphthyridine (7) with 50% yield (1.24 g) as a white solid, started from 2.39 g (12.3 mmol) of 3, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.63 (s, 1H), 8.88 (d, J = 6.0 Hz, 1H), 7.86 (d, J = 5.6 Hz, 1H), 7.60 (s, 1H); ¹³C NMR (75 MHz, CDCl₃) δ 154.9, 150.9, 149.4, 149.2, 144.9, 123.7, 121.0, 120.4 ppm; m.p. 183.4–184.0 °C; HRMS (ESI) m/z calcd for C₈H₅Cl₂N₂ (M + H)⁺ 198.9825 found 198.9819.

2,4-dichloro-1,7-naphthyridine (8) with 41% yield (120 mg) as a white solid, started from 309 mg (1.49 mmol) of 4, column chromatography with 40% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.44 (s, 1H), 8.75 (d, J = 5.7 Hz, 1H), 7.97 (d, J = 5.7 Hz, 1H), 7.70 (s, 1H); ¹³C NMR (75 MHz, CDCl₃) δ 153.1, 151.6, 145.4, 143.7, 142.7, 128.8, 126.0, 116.1 ppm; m.p. 154.9–156.7 °C; HRMS (ESI) m/z calcd for C₈H₅Cl₂N₂ (M + H)⁺ 198.9824 found 198.9820.

General procedure for the synthesis of 4-oxy-aryl Naphthyridine derivatives (9–10(a-b))

To a solution of dichloronaphthyridines (7 or 8), 4-hydroxy-3,5-dimethylbenzaldehyde (for synthesis of 9a and 10a) or 4-hydroxy-3,5-dimethyl benzonitrile (for synthesis of 9b and 10b) (1.10−1.30 equiv) and Cs₂CO₃ (1.00−2.00 equiv) in DMF (0.05−0.20 M) was stirred at room temperature for 18–21 h. The water was added to the reaction mixture and then extracted thrice with EtOAc. The combined organic layers were washed with water and brine and then dried over with anhydrous sodium sulfate. The crude mixture was concentrated in vacuo and purified by column chromatography to obtain 4-oxy-aryl naphthyridine derivatives (45−86%).

4-(2´,6´,-dimethyl-4´-formylphenoxy)−2-chloro-1,6-naphthyridine (9a)

49% yield (791 mg) as yellow solid, started from 1.02 g (5.15 mmol) of 7, oxy-aryl (840 mg, 5.59 mmol, 1.10 equiv), Cs₂CO₃ (2.47 g, 7.58 mmol, 1.50 equiv), DMF 0.05 M, stirred at room temperature for 18 h, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 10.01 (s, 1H), 9.80 (d, J = 0.8 Hz, 1H), 8.87 (d, J = 6.0 Hz, 1H), 7.84 (dd, J = 6.0, 0.8 Hz, 1H), 7.75 (s, 2 H), 6.28 (s, 1H), 2.25 (s, 6 H); ¹³C NMR (100 MHz, CDCl₃) δ 191.1, 161.9, 156.5, 153.8, 151.5, 149.3, 147.4, 134.7, 131.9, 131.1, 121.0, 115.2, 104.5, 16.1 ppm; m.p. 183.4–184.0 °C; HRMS (ESI) m/z calcd for C₁₇H₁₄ClN₂O₂ (M + H)⁺ 313.0744 found 313.0740.

2-(2´,6´,-dimethyl-4´-formylphenoxy)−4-chloro-1,6-naphthyridine (11a)

2-oxy-aryl naphthyridine derivative as a side product of 9a, 23% yield (363 mg) and as a yellow solid, ¹H NMR (400 MHz, CDCl₃) δ 10.00 (s, 1H), 9.51 (s, 1H), 8.69 (d, J = 6.0 Hz, 1H), 7.69 (s, 2 H), 7.49 (dd, J = 6.0, 0.4 Hz, 1H), 7.37 (s, 1H), 2.20 (s, 6 H); ¹³C NMR of (100 MHz, CDCl₃) δ 191.6, 162.7, 154.7, 150.5, 148.8, 148.7, 145.7, 134.0, 132.1, 130.4, 120.7, 119.7, 113.3, 16.7 ppm; m.p. 160–161.6 °C; HRMS (ESI) m/z calcd for C₁₇H₁₄ClN₂O₂ (M + H)⁺ 313.0738 found 313.0743.

4-(2´,6´,-dimethyl-4´-formylphenoxy)−2-chloro-1,7-naphthyridine (10a): 45% yield (670 mg) as a yellow solid, started from 950 mg (4.77 mmol) of 8, oxy-aryl (788.4 mg, 5.25 mmol, 1.10 equiv), Cs₂CO₃ (2.33 g, 7.15 mmol, 1.50 equiv), DMF 0.05 M, stirred at room temperature for 18 h, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 10.01 (s, 1H), 9.43 (d, J = 0.6 Hz, 1H), 8.74 (d, J = 5.7 Hz, 1H), 8.16 (dd, J = 5.7, 0.9 Hz, 1H), 7.75 (s, 2 H), 6.35 (s, 1H), 2.22 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 191.0, 160.3, 154.0, 153.0, 152.6, 144.5, 143.5, 134.7, 131.8, 131.1, 123.3, 114.1, 106.3, 16.1 ppm; m.p. 196.4–197.7 °C; HRMS (ESI) m/z calcd for C₁₇H₁₄ClN₂O₂ (M + H)⁺ 313.0738 found 313.0730. Nevertheless, 2-(2´,6´,-dimethyl-4´-formylphenoxy)−4-chloro-1,7-naphthyridine (12a) was found as byproduct with trace amount during synthesis of compound 10a.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-chloro-1,6-naphthyridine (9b)

48% yield (151 mg) as a yellow solid, started from 200 mg (1.00 mmol) of 7, oxy-aryl (177 mg, 1.21 mmol, 1.20 equiv), Cs₂CO₃ (327 mg, 1.00 mmol, 1.00 equiv), DMF 0.20 M, stirred at room temperature for 21 h. column chromatography with 30% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.78 (s, 1H), 8.89 (d, J = 5.6 Hz, 1H), 7.85 (d, J = 5.6 Hz, 1H), 7.55 (s, 2 H), 6.27 (s, 1H), 2.22 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 161.6, 156.4, 152.6, 151.6, 149.4, 147.3, 133.5, 132.5, 121.0, 117.9, 115.0, 111.0, 104.4, 16.0 ppm; m.p. 226.5–226.8 °C; HRMS (ESI) m/z calcd for C₁₇H₁₃ClN₃O (M + H)⁺ 310.0742 found 310.0745.

2-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−4-chloro-1,6-naphthyridine (11b)

2-oxy-aryl naphthyridine derivative as a side product of 9b, 21% yield (64.8 mg) and as a white solid, ¹H NMR (300 MHz, CDCl₃) δ 9.51 (d, J = 0.3 Hz, 1H), 8.70 (d, J = 6.0 Hz, 1H), 7.49 (dd, J = 6.0, 0.6 Hz, 1H), 7.46 (s, 2 H), 7.37 (s, 1H), 2.15 (s, 6 H);¹³C NMR (75 MHz, CDCl₃) δ 162.5, 153.4, 150.4, 148.9, 148.8, 145.8, 132.7, 132.6, 120.7, 119.7, 118.7, 113.3, 109.7, 16.5 ppm; m.p. 198–200 °C; HRMS (ESI) m/z calcd for C₁₇H₁₃ClN₃O (M + H)⁺ 310.0742 found 310.0747.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-chloro-1,7-naphthyridine (10b)

86% yield (42.0 mg) as a white solid, started from 50.0 mg (0.250 mmol) of 8, oxy-aryl (48.0 mg, 0.303 mmol, 1.30 equiv), Cs₂CO₃ (163 mg, 0.50. mmol, 2.00 eq), DMF 0.20 M, stirred at room temperature for 21 h. column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.45 (d, J = 0.8 Hz, 1H), 8.76 (d, J = 5.6 Hz, 1H), 8.14 (dd, J = 5.6, 0.8 Hz, 1H), 7.55 (s, 2 H), 6.34 (s, 1H), 2.20 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.0, 152.9, 152.7, 152.5, 144.5, 143.4, 133.4, 132.4, 123.1, 117.9, 113.9, 110.9, 106.2, 15.9 ppm; m.p. 214.6–216.4 °C; HRMS (ESI) m/z calcd for C₁₇H₁₃ClN₃O (M + H)⁺ 310.0742 found 310.0736. However, 2-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−4-chloro-1,6-naphthyridine (12b) was detected in trace amounts as a byproduct during the synthesis of compound 10b.

General procedure for the synthesis of 2,4-dioxy-aryl Naphthyridine derivatives (13–14(a-b))

To a solution of dichloronaphthyridine (7 or 8), 4-hydroxy-3,5-dimethylbenzaldehyde (for synthesis of 13a and 14a) or 4-hydroxy-3,5-dimethyl benzonitrile (for synthesis of 13b and 14b) (2.50 equiv) and Cs₂CO₃ (2.00 equiv) in DMF (0.05 M) was Heated in a sealed tube at 120 °C for 12 h. The reaction mixture was cooled to room temperature before the water was added and was extracted thrice with EtOAc. The combined organic layers were washed with water and brine and then dried over with anhydrous sodium sulfate. The crude mixture was concentrated in vacuo and purified by column chromatography to obtain 2,4-dioxy-aryl naphthyridine derivatives (78−90%).

2,4-di-(2ˊ,6ˊ- dimethyl-4ˊ-formylphenoxy)−1,6-naphthyridine (13a)

89% yield (204 mg) as a white solid, started from 107 mg (0.54 mmol) of 7, oxy-aryl (194.9 mg, 1.30 mmol, 2.50 equiv), Cs₂CO₃ (329 mg, 1.01 mmol, 2.00 equiv), DMF 0.05 M, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 10.03 (s, 1H), 9.97 (s, 1H), 9.68 (s, 1H), 8.69 (d, J = 6.0 Hz, 1H), 7.78 (s, 2 H), 7.65 (s, 2 H), 7.47 (d, J = 5.6 Hz, 1H), 6.05 (s, 1H), 2.31 (s, 6 H), 2.14 (s, 6 H); ¹³C NMR (100 MHz, CDCl₃) δ 191.6, 191.2, 164.7, 163.0, 155.0, 154.3, 151.3, 148.5, 146.5, 134.6, 133.8, 132.1, 132.0, 131.1, 130.3, 120.8, 114.8, 93.2, 16.6, 16.2 ppm; m.p. 221.9–223.1 °C; HRMS (ESI) m/z calcd for C₂₆H₂₃N₂O₄ (M + H)⁺ 427.1658 found 427.1651.

2,4-di-(2ˊ,6ˊ- dimethyl-4ˊ-formylphenoxy)−1,7-naphthyridine (14a)

86% yield (190 mg) as a white solid, started from 104 mg (0.52 mmol) of 8, oxy-aryl (195 mg, 1.30 mmol, 2.50 equiv), Cs₂CO₃ (336 mg, 1.03 mmol, 2.00 equiv), DMF 0.05 M, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 10.04 (s, 1H), 9.98 (s, 1H), 9.05 (s, 1H), 8.62 (d, J = 5.6 Hz, 1H), 8.12 (d, J = 5.6 Hz, 1H), 7.78 (s, 2 H), 7.66 (s, 2 H), 6.17 (s, 1H) 2.29 (s, 6 H), 2.16 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 191.6, 191.2, 162.4, 161.4, 155.1, 154.4, 151.5, 142.7, 142.4, 134.6, 133.8, 132.2, 132.0, 131.1, 130.3, 122.5, 114.2, 95.5, 16.6, 16.1 ppm; m.p. 104–107 °C; HRMS (ESI) m/z calcd for C₂₆H₂₃N₂O₄ (M + H)⁺ 427.1658 found 427.1648.

2,4-di-(2ˊ,6ˊ- dimethyl-4ˊ-cyanophenoxy)−1,6-naphthyridine (13b)

78% yield (168 mg) as a white solid, started from 102 mg (0.51 mmol) of 7, oxy-aryl (187 mg, 1.27 mmol, 2.50 equiv), Cs₂CO₃ (333 mg, 1.02 mmol, 2.00 equiv), DMF 0.05 M, column chromatography with 60% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.66 (s, 1H), 8.70 (d, J = 5.6 Hz, 1H), 7.56 (s, 2 H), 7.47 (d, J = 5.6 Hz, 1H), 7.42 (s, 2 H), 6.03 (s, 1H), 2.27 (s, 6 H), 2.09 (s, 6 H); ¹³C NMR (100 MHz, CDCl₃) δ 164.3, 162.8, 153.6, 153.1, 151.1, 149.1, 146.6, 133.5, 132.7, 132.7, 132.5, 120.7, 118.8, 118.1, 114.7, 110.8, 109.4, 93.0, 16.5, 16.1 ppm; m.p. 233.2–234.1 °C; HRMS (ESI) m/z calcd for C₂₆H₂₀NaN₄O₂ (M + Na)⁺ 443.1478 found 443.1478.

2,4-di-(2ˊ,6ˊ- dimethyl-4ˊ-cyanophenoxy)−1,7-naphthyridine (14b)

90% yield (28.0 mg) as a white solid, started from 20.0 mg (0.10 mmol) of 8, oxy-aryl (37.0 mg, 0.25 mmol, 2.50 equiv), Cs₂CO₃ (65.0 mg, 0.20 mmol, 2.00 equiv), DMF 0.05 M, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.05 (s, 1H), 8.62 (d, J = 5.6 Hz, 1H), 8.09 (d, J = 5.6 Hz, 1H), 7.57 (s, 2 H), 7.44 (s, 2 H), 6.15 (s, 1H), 2.25 (s, 6 H), 2.11 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 162.1, 161.2, 153.6, 153.1, 151.6, 143.0, 142.2, 133.4, 132.7, 132.6, 132.5, 122.4, 118.7, 118.1, 114.0, 110.7, 109.3, 95.3, 16.5, 16.0 ppm; m.p. 206.1–206.4 °C; HRMS (ESI) m/z calcd for C₂₆H₂₁N₄O₂ (M + H)⁺ 421.1659 found 421.1652.

General procedure for the synthesis of 2-amino-aryl-4-oxy-aryl Naphthyridine derivatives (15–18(a-b))

To a solution of each compound 9–10(a-b), 4-aminobenzonitrile (for synthesis of 15, 17(a-b)) or 2-amino-5-cyanopyridine (for synthesis of 16, 18(a-b)) (0.90−1.30 equiv), ligand (0.10 equiv), Pd(OAc)₂ (0.10 equiv), and Cs₂CO₃ (1.50 equiv) in DMF (0.02−0.05 M) was Heated in a sealed tube at 120 °C for 10−16 h. The reaction mixture was cooled to room temperature before being filtered via Celite. The water was added to the reaction mixture and then extracted thrice with EtOAc. The combined organic layers were washed with water and brine and then dried over with anhydrous sodium sulfate. The crude mixture was concentrated in vacuo and purified by column chromatography to obtain 2-amino-aryl-4-oxy-aryl-naphthyridine (32−70%).

4-(2ˊ,6ˊ-dimethyl-4ˊ-formylphenoxy)−2-(4ˊˊ-cyanophenyl)-amino-1,6-naphthyridine (15a)

51% yield (42.6 mg) as a yellow solid, started from 65.7 mg (0.21 mmol) of 9a, amino-aryl (27.3 mg, 0.23 mmol, 1.10 equiv), PhDavePhos (8.01 mg, 0.02 mmol, 0.10 equiv), Pd(OAc)₂ (4.71 mg, 0.02 mmol, 0.10 equiv) and Cs₂CO₃ (103 mg, 0.32 mmol, 1.5 0equiv), DMF (0.02 M), Heated at 120 °C for 12 h, column chromatography with 50% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.95 (s, 1H), 9.58 (d, J = 0.6 Hz, 1H), 8.71 (d, J = 5.7 Hz, 1H), 7.91 (d, J = 9.0 Hz, 2 H), 7.70 (s, 2 H), 7.68 (dd, J = 5.7, 0.6 Hz, 1H), 7.61 (d, J = 9.0 Hz, 2 H), 5.75 (s, 1H), 2.28 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 191.3, 161.2, 157.0, 154.5, 152.3, 148.9, 146.2, 143.7, 134.4, 133.3, 132.3, 130.9, 120.4, 119.1, 113.9, 105.4, 94.0, 16.2 ppm; m.p. 296–298 °C; HRMS (ESI) m/z calcd for C₂₄H₁₉N₄O₂ (M + H)⁺ 395.1503 found 395.1513.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(4ˊˊ-cyanophenyl)-amino-1,6-naphthyridine (15b)

32% yield (15.0 mg) as a white solid, started from 37.0 mg (0.12 mmol) of 9b, amino-aryl (18.0 mg, 0.15 mmol, 1.30 equiv), SPhos (5.00 mg, 0.01 mmol, 0.10 equiv), Pd(OAc)₂ (2.00 mg, 0.01 mmol, 0.10 equiv) and Cs₂CO₃ (58.4 mg, 0.18 mmol, 1.50 equiv), DMF (0.02 M), Heated at 120 °C for 10 h. column chromatography with 50% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.47 (s, 1H), 8.62 (d, J = 6 Hz, 1H), 8.00 (d, J = 8.7 Hz, 2 H), 7.69 (d, J = 6 Hz, 1H), 7.60 (d, J = 8.7 Hz, 2 H), 7.54 (s, 2 H), 5.87 (s, 1H), 2.25 (s, 7 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.2, 157.7, 153.2, 152.4, 147.8, 145.4, 144.2, 133.0, 132.9, 120.4, 119.3, 119.0, 118.0, 113.7, 109.9, 104.3, 94.6, 15.7 ppm; m.p. decomposed at 300 °C; HRMS (ESI) m/z calcd for C₂₄H₁₈N₅O (M + H)⁺ 392.1506 found 392.1513.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(4ˊˊ-cyanophenyl)-amino-1,7-naphthyridine (17a)

40% yield (28.5 mg) as a yellow solid, started from 57.1 mg (0.18 mmol) of 10a, amino-aryl (28.0 mg, 0.24 mmol, 0.10 equiv), PhDavePhos (6.96 mg, 0.02 mmol, 0.10 equiv), Pd(OAc)₂ (4.10 mg, 0.02 mmol, 0.10 equiv) and Cs₂CO₃ (89.2 mg, 0.27 mmol, 1.50 equiv), DMF (0.03 M), Heated at 120 °C for 16 h, column chromatography with 40% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.98 (s, 1H), 9.28 (s, 1H), 8.57 (d, J = 5.4 Hz, 1H), 8.06 (d, J = 5.4 Hz, 1H), 7.90 (d, J = 8.7 Hz, 2 H), 7.72 (s, 2 H), 7.61 (d, J = 8.7 Hz, 2 H), 6.99 (s, 1H), 5.85 (s, 1H), 2.26 (s, 6 H); ¹³C NMR (100 MHz, CDCl₃) δ 191.9, 159.0, 155.8, 154.8, 150.3, 144.9, 143.9, 140.7, 134.3, 133.1, 132.3, 130.9, 121.9, 119.6, 118.5, 114.7, 103.6, 97.5, 15.9 ppm; m.p. decomposed at 300 °C; HRMS (ESI) m/z calcd for C₂₄H₁₉N₄O₂ (M + H)⁺ 395.1503 found 395.1493.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(4ˊˊ-cyanophenyl)-amino-1,7-naphthyridine (17b)

70% yield (71.3 mg) as a yellow solid, started from 80.0 mg (0.26 mmol) of 10b, amino-aryl (3.90 mg, 0.03 mmol, 1.03 equiv), SPhos (10.6 mg, 0.03 mmol, 0.10 equiv), Pd(OAc)₂ (5.80 mg, 0.03 mmol, 0.10 equiv) and Cs₂CO₃ (126 mg, 0.39 mmol, 1.50 equiv), DMF (0.03 M), Heated at 120 °C for 10 h, column chromatography with 20% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.11 (s, 1H), 8.37 (dd, J = 5.7, 3.3 Hz, 1H), 7.97 (d, J = 5.7 Hz, 1H), 7.92 (d, J = 8.7 Hz, 2 H), 7.51 (d, J = 8.7 Hz, 2 H), 7.45 (s, 2 H), 5.94 (s, 1H), 2.14 (s, 6 H);¹³C NMR (75 MHz, CDCl₃) δ 158.8, 155.4, 153.4, 150.4, 144.6, 143.7, 140.8, 133.0, 133.0, 132.9, 121.5, 119.4, 118.4, 118.0, 114.3, 109.8, 103.7, 97.2, 15.7 ppm; m.p. decomposed at 300 °C; HRMS (ESI) m/z calcd for C₂₄H₁₈N₅O (M + H)⁺ 392.1506 found 392.1495.

4-(2ˊ,6ˊ-dimethyl-4ˊ-formylphenoxy)−2-(5ˊˊ-cyanopyridin-2ˊˊ-ylamino)−1,6-naphthyridine (16a)

89% yield (458 mg) as a yellow solid, started from 200 mg (0.64 mmol) of 9a, amino-aryl (68.6 mg, 0.58 mmol, 0.90 equiv), Xantphos (37.0 mg, 0.06 mmol, 0.10 equiv), Pd(OAc)₂ (14.4 mg, 0.06 mmol, 0.10 equiv) and Cs₂CO₃ (313 mg, 0.96 mmol, 1.50 equiv) in DMF (0.05 M), Heated at 120 °C for 12 h, column chromatography with 60% of EtOAc in Hexanes, ¹H NMR (400 MHz, DMSO) δ 10.59 (s, 1H), 10.03 (s, 1H), 9.54 (s, 1H), 8.70–8.68 (m, 2 H), 8.62 (d, J = 2.0 Hz, 1H), 8.19 (dd, J = 8.88, 2.4 Hz, 1H), 7.86 (s, 2 H), 7.70 (d, J = 5.6 Hz, 1H), 6.63 (s, 1H), 2.22 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 191.5, 161.2, 157.3, 155.1, 154.1, 151.4, 146.1, 145.0, 140.9, 134.4, 132.0, 130.9, 120.6, 117.0, 114.0, 113.4, 102.6, 95.8, 29.5, 15.9 ppm; m.p. 243.0–243.6 °C; HRMS (ESI) m/z calcd for C₂₃H₁₈N₅O₂ (M + H)⁺ 396.1455 found 396.1446.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(5ˊˊ-cyanopyridin-2ˊˊ-ylamino)−1,6-naphthyridine (16b)

72% yield (349 mg) as a yellow solid, started from 381 mg (1.23 mmol) of 9b, amino-aryl (137 mg, 1.15 mmol, 0.90 equiv), Xantphos (73.6 mg, 0.13 mmol, 0.10 equiv), Pd(OAc)₂ (35.2 mg, 0.16 mmol, 0.10 equiv) and Cs₂CO₃ (611 mg, 1.87 mmol, 1.50 equiv) in DMF (0.05 M), Heated at 120 °C for 12 h, column chromatography with 40% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.63 (s, 1H), 8.76 (d, J = 6.0 Hz, 1H), 8.59 (d, J = 8.8 Hz, 1H), 8.45 (d, J = 2 Hz, 1H), 7.94 (dd, J = 8.8, 2.4 Hz, 1H), 7.71 (s, 1H), 7.70 (d, J = 5.6 Hz, 1H), 7.55 (s, 2 H), 6.11 (s, 1H), 2.24 (s, 6 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.9, 156.6, 155.2, 153.1, 151.5, 148.1, 146.0, 140.9, 133.1, 132.8, 120.4, 118.0, 117.1, 113.9, 113.1, 110.2, 102.4, 95.2, 60.4, 15.8 ppm; m.p. decomposed at 300 °C; HRMS (ESI) m/z calcd for C₂₃H₁₇N₆O (M + H)⁺ 393.1458 found 393.1446.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(5ˊˊ-cyanopyridin-2ˊˊ-ylamino)−1,7-naphthyridine (18a)

76% yield (192 mg) as a yellow solid, started from 200 mg (0.64 mmol) of 10a, amino-aryl (70.3 mg, 0.35 mmol, 0.90 equiv), Xantphos (40.4 mg, 0.07 mmol, 0.10 equiv), Pd(OAc)₂ (16.0 mg, 0.07 mmol, 0.10 equiv) and Cs₂CO₃ (350 mg, 1.07 mmol, 1.50 equiv) in DMF (0.05 M), Heated at 120 °C for 12 h, column chromatography with 40% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 10.03 (s, 1H), 9.31 (s, 1H), 8.62 (d, J = 5.6 Hz, 1H), 8.59 (d, J = 8.8 Hz, 1H), 8.42 (s, 1H), 8.10 (d, J = 5.2 Hz, 1H), 7.94 (d, J = 8.8 Hz, 1H), 7.76 (s, 2 H), 7.75 (s, 1H), 6.26 (s, 1H), 2.26 (s, 6 H); ¹³C NMR (75 MHz, MeOD) δ 191.6, 159.3, 155.4, 154.5, 154.3, 151.4, 150.4, 143.2, 141.1, 140.7, 134.2, 132.0, 130.8, 121.9, 117.2, 114.5, 112.5, 101.6, 97.9, 15.7 ppm; m.p. 250–250.1 °C; HRMS (ESI) m/z calcd for C₂₃H₁₈N₅O₂ (M + H)⁺ 396.1455 found 396.1452.

4-(2ˊ,6ˊ-dimethyl-4ˊ-cyanophenoxy)−2-(5ˊˊ-cyanopyridin-2ˊˊ-ylamino)−1,7-naphthyridine (18b)

54% yield (72.3 mg) as a yellow solid, started from 106 mg (0.34 mmol) of 10b, amino-aryl (45.2 mg, 0.38 mmol, 1.10 equiv), SPhos (14.1 mg, 0.03 mmol, 0.10 equiv), Pd(OAc)₂ (7.70 mg, 0.03 mmol, 0.10 equiv) and Cs₂CO₃ (168 mg, 0.52 mmol, 1.50 equiv) in DMF (0.05 M), Heated at 120 °C for 10 h, column chromatography with 5% of EtOAc in DCM, ¹H NMR (300 MHz, DMSO) δ 10.53 (s, 1H), 9.22 (s, 1H), 8.70 (d, J = 9.0 Hz, 1H), 8.64 (s, 1H), 8.57 (d, J = 5.4 Hz, 1H), 8.20 (dd, J = 9.0, 2.4 Hz, 1H), 8.12 (d, J = 5.4 Hz, 1H), 7.87 (s, 2 H), 6.75 (s, 1H), 2.16 (s, 6 H); ¹³C NMR (100 MHz, DMSO) δ 158.9, 156.1, 155.1, 153.6, 152.5, 150.9, 143.3, 142.5, 141.9, 133.9, 133.1, 121.4, 118.9, 118.3, 114.5, 112.5, 109.7, 101.4, 98.8, 15.8 ppm; m.p. more than 315 °C; HRMS (ESI) m/z calcd for C₂₃H₁₇N₆O (M + H)⁺ 393.1459 found 393.1454.

General procedure for the synthesis of 2-amino-pyridinyl-4-phenoxy-vinyl-naphthyridine derivatives (19–20)a

To a solution of 2-amino-pyridinyl-4-oxy-aryl-naphthyridine (16a or 18a), tert-BuOK (1.50 eq) in THF (0.03 M) was stirred at 0 °C for 10−15 min before adding (EtO)₂P(O)CH₂CN (1.50 eq). The reaction mixture was stirred at 0 °C for 48 h and concentrated in vacuo. The crude product was purified by column chromatography (eluent: hexane/ethyl acetate) to obtain 2-amino-pyridinyl-4-phenoxy-vinyl-naphthyridine derivatives as 19a with 20a in 29% and 33% yield, respectively.

4-(4´-(2´´-(E, Z)-cyanovinyl)−2´,6´-dimethyl-phenoxy)−2-(5´´-cyanopyridin-2´´ylamino)-amino-1,6-naphthyridine (19a)

29% yield (10.8 mg) as a white solid, started from 35.4 mg (0.09 mmol) of 16a, T-BuOK (15.1 mg, 0.13 mmol, 1.50 eq), (EtO)₂P(O)CH₂CN (21.8 µL, 0.13 mmol, 1.50 eq), THF 0.03 M, column chromatography with 60% of EtOAc in Hexanes, ¹H NMR (400 MHz, CDCl₃) δ 9.67 (d, J = 2.4 Hz, 2 H), 8.75 (d, J = 5.7 Hz, 2 H), 8.66 (d, J = 8.7, 1H), 8.58 (d, J = 8.7 Hz, 1H), 8.43 (s, 2 H), 8.23 (s, 2 H),7.94 (dt, J = 8.7, 2.4 Hz, 2 H), 7.70 (d, J = 9.9 Hz, 4 H), 7.40 (d, J = 16.5 Hz, 1H), 7.32 (s, 2 H), 7.16 (d, J = 12 Hz, 1H), 6.27 (s, 1H), 6.19 (s, 1H), 5.92 (d, J = 16.8 Hz, 1H), 5.53 (d, J = 12 Hz, 1H), 2.24 (d, J = 4.5 Hz, 12 H); ¹³C NMR (75 MHz, CDCl₃) δ 162.0, 161.9, 156.1, 156.1, 155.0, 151.8, 151.8, 151.7, 151.7, 151.6, 149.4, 149.0, 148.9, 147.5, 146.7, 146.6, 141.1, 141.0, 132.0, 132.0, 131.9, 131.7, 130.2, 128.5, 120.2, 120.1, 117.9, 117.3, 117.2, 117.2, 114.2, 114.1, 113.0, 112.9, 102.9, 102.8, 96.9, 95.7, 94.9, 94.8, 16.1, 16.1 ppm; m.p. 243.0–243.6 °C; HRMS (ESI) m/z calcd for C₂₅H₁₉N₆O (M + H)⁺ 419.1615 found 419.1614.

4-(4´-(2´´-(E, Z)-cyanovinyl)−2´,6´-dimethyl-phenoxy)−2-(5´´-cyanopyridin-2´´ylamino)-amino-1,7-naphthyridine (20a) : 33% yield (15.4 mg) as a white solid, started from 44.5 mg (0.11 mmol) of 18a, T-BuOK (18.9 mg, 0.17 mmol, 1.50 eq), (EtO)₂P(O)CH₂CN (27.0 µL, 0.17 mmol, 1.50 eq), THF 0.03 M, column chromatography with 60% of EtOAc in Hexanes, ¹H NMR (300 MHz, CDCl₃) δ 9.25 (s, 2 H), 8.65 (d, J = 8.7 Hz, 1H), 8.64 (d, J = 9.0 Hz, 1H), 8.54 (d, J = 5.7 Hz, 2 H), 8.43–8.41 (m, 2 H), 8.14 (d, J = 5.4 Hz, 1H), 8.13 (d, J = 4.8 Hz, 1H), 7.94 (dt, J = 9.0, 1.8 Hz, 2 H), 7.66 (s, 2 H), 7.42 (d, J = 16.5 Hz, 1H), 7.34 (s, 2 H), 7.19 (d, J = 12 Hz, 1H), 6.51 (s, 1H), 6.46 (s, 1H), 5.94 (d, J = 16.5 Hz, 1H), 5.54 (d, J = 12 Hz, 1H) 2.22 (s, 12 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.0, 159.4, 155.5, 155.4, 154.3, 154.3, 154.2, 151.6, 151.5, 150.7, 150.7, 149.6, 147.9, 143.3, 143.3, 141.6, 141.5, 141.1, 140.9, 131.9, 131.8, 131.6, 130.1, 128.5, 122.2, 122.1, 117.9, 117.4, 117.3, 117.2, 117.2, 114.6, 114.6, 112.7, 112.6, 101.9, 101.9, 97.8, 97.7, 96.5, 95.5, 16.0 ppm; m.p. 250.0–250.1 °C; HRMS (ESI) m/z calcd for C₂₅H₁₉N₆O (M + H)⁺ 419.1615 found 419.1615.

HIV-1 RT inhibition assay

The inhibition assay of HIV-1 RT involved the utilization of the template/primer hybrid poly(A) × oligo(dT)15, along with digoxigenin (DIG)- and biotin-labeled nucleotides. An antibody specific to DIG, conjugated to peroxidase (anti-DIG-POD), was employed, along with the peroxidase substrate ABTS. The quantities of digoxigenin (DIG)- and biotin-labeled dUTP incorporated into DNA serve as a measure of HIV-1 RT activity. The HIV-1 RT inhibition assay was conducted utilizing an RT assay kit (Roche Applied Science, Mannheim, Germany). The procedures for assessing RT inhibition were carried out according to the protocol provided with the kit^41,42. The tested compounds, along with three control drugs (NVP, EFV, and RPV), were utilized at a concentration of 1 µM to determine the percentage inhibitory values. The reaction mixture comprised a template/primer complex, 2’-deoxy-nucleotide-5’-triphosphates (dNTPs), and reverse transcriptase (RT) enzyme, all suspended in lysis buffer. Optionally, the lysis buffer may contain an inhibitor. The reaction mixture was incubated for 1 h at 37 °C and subsequently transferred to a streptavidin-coated microtiter plate (MTP). The biotin-labeled dNTPs, which were incorporated into the template, bound to streptavidin due to the presence of RT. Unbound dNTPs were washed away using a wash buffer, following which anti-DIG-POD was added to the MTP. The DIG-labeled dNTPs, which were incorporated into the template, bound to the anti-DIG-POD antibody. The unbound anti-DIG-POD was thoroughly washed five times using a wash buffer. Finally, the peroxide substrate, ABTS, was added to the microtiter plate (MTP). A colored reaction product appeared during the substrate cleavage catalyzed by the peroxidase enzyme. The absorbance of the sample was measured at an optical density (OD) of 490 nm using the microtiter plate (MTP) reader, which employs enzyme-linked immunosorbent assay (ELISA) principles. The percentage inhibitory activities of the RT inhibitors were determined by comparing the sample with the one lacking an inhibitor. The resulting color intensity was directly proportional to the reverse transcriptase (RT) activity. Percentage inhibitory values were calculated using the following formula: Percentage inhibition = [1–(OD value with RT and inhibitor − OD value without RT and inhibitors)/(OD value without inhibitors with RT − OD value without RT and inhibitors)] × 100. The half maximal inhibitory concentration (IC₅₀) was determined as the concentration of the compounds of interest and the three control drugs at which 50% cell growth inhibition occurred. This was achieved by plotting a sigmoid curve between the logarithm of the concentration on the x-axis and the inhibition rate on the y-axis.

Molecular docking studies

Ligand and protein structure preparation involved the utilization of four commercially available drugs (nevirapine, efavirenz, etravirine, and rilpivirine) and fourteen synthetic compounds of phenylamino-phenoxy-naphthyridine derivatives as ligands. The ligands were generated using Gaussian 16 and fully optimized with density functional theory (DFT) at the B3LYP/6-31G(d, p) level⁴³. The crystal structures of the HIV-1 reverse transcriptase (HIV-1 RT) enzyme in a complex with rilpivirine were obtained from the Protein Data Bank (7VH8.pdb). Molecular docking was performed after removing ligands and water molecules from the protein structure and adding hydrogen atoms. The binding interaction between the ligands and HIV-1 RT was conducted using the AutoDock 4.2 program, treating the protein as a rigid structure. The Lamarckian Genetic Algorithm (LGA) was employed with a population size of 150 individuals, and the number of genetic algorithm runs was set at 200. The grid box size was set to 80 × 80 × 80 Å with a spacing value of 0.375 Å. The grid center for HIV-1 RT was applied at the values 49.082, −28.29, and 37.541 Å⁴⁴. The best-scoring compounds were selected and further visually analyzed using Accelrys Discovery Studio Client 4.0⁴⁵.

Molecular dynamic simulations

The molecular dynamics (MD) simulations were conducted using the GROMACS (v2022.3) software package (University of Groningen, Groningen, Netherlands)⁴⁶. The protein and ligand topologies were generated using the AmberTools 22 program⁴⁷. Ligands were assigned the RESP charges using the Antechamber module, and force field parameters were determined using the general AMBER force field 2 (GAFF2)⁴⁸. MD simulations were performed to evaluate the stability and conformational changes of the complexes involving HIV-1 RT-RPV, HIV-1 RT-16a, HIV-1 RT-16b, and HIV-1 RT-19a. The AMBER19SB force field was applied in the MD simulations⁴⁹. The complexes were simulated within an octahedral box solvated with the TIP3P water model, with the complex centrally placed at a 10 Å distance from the box edges⁵⁰. Energy minimization was carried out using the steepest descent algorithm with 50,000 steps to eliminate any steric clashes or unfavorable contact. Followed by a 1000 ps equilibration of the entire system. Sodium and chloride ions were added to the solvent to neutralize the system. The MD simulations were then run for 300 ns, maintaining a temperature of 310 K and a pressure of 1 bar using the NPT ensemble. The analysis and visualization of the MD trajectories were performed using the GROMACS tools and Visual Molecular Dynamics (VMD) software⁵¹.

Pharmacokinetic parameter calculation

In this study, SwissADME was employed to analyze the pharmacokinetic parameters, including absorption, distribution, metabolism, and excretion (ADME), and to screen drug-likeness and molecular properties based on Lipinski’s Rule of Five for the ligands.

Cytotoxic activity

The cell lines were seeded into a 96-well microplate (Costar No. 3599) at a density ranging from 5 × 10³ to 2 × 10⁴ cells per well, with each well receiving 100 µL of cell suspension. Background control wells contained the same volume as the complete culture medium, serving as a reference for background levels. The microplate was incubated for 24 h at 37 °C with 5% CO₂ and 95% humidity using a Shellab incubator. Samples at various concentrations were added to the microplate, followed by further incubation for another 48 h. Cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyl tetrazolium bromide (MTT) assay obtained from Sigma-Aldrich^52,53. The reagent was dissolved in phosphate-buffered saline at a concentration of 5 mg/mL and subsequently filtered to sterilize and remove any small amounts of insoluble residue present in some batches of MTT. MTT solution (10 µL/100 µL medium) was added to all wells of each assay, and the plates were then incubated at 37 °C with 5% CO₂ and 95% humidity for 2−4 hours. Afterward, dimethyl sulfoxide (100 µL; Merck, Germany) was added to dissolve the resulting formazan by sonication. The plates were analyzed using a microplate reader (Molecular Devices, CA, USA), with measurements taken at a test wavelength of 550 nm and a reference wavelength of 650 nm. The XTT assay for suspension cells was employed for MOLT-3 cells⁵². After the addition of a 50 µL mixture consisting of 1 mg/mL (5 mL) and 0.383 mg/mL (100 µL) phenazine methosulfate, the plates were incubated for 4 hours. The absorbance of the orange formazan compounds was measured at wavelengths of 492 and 690 nm. IC₅₀ values were determined as the concentration of the drug and samples at which 50% cell growth inhibition occurred. IC₅₀ values for each cell line, expressed in µM, were obtained from serial dilutions starting at 50 µg/mL and reported as greater than 50 µg/mL when 50% inhibition was not observed at this concentration. Percentage cytotoxicity at 50 µg/mL (values shown in parentheses) denotes the cytotoxicity observed at the highest tested concentration. The selectivity index (SI) denotes the ability of a bioactive compound to act with greater specificity toward target cells which evaluates the biological potential and clinical applicability. The selectivity index (SI) for cancer cell lines is calculated as the ratio of the IC₅₀ in normal cell to the IC₅₀ in cancer cells⁵⁴. The SI value is greater than 3, indicating high selectivity toward targeted anticancer activity^54,55.

This study successfully synthesized a series of fourteen novel 2,4-disubstituted-1,6- and 1,7-naphthyridine derivatives through a molecular hybridization strategy utilizing commercially available non-nucleoside reverse transcriptase inhibitors (NNRTIs), specifically nevirapine (NVP), efavirenz (EFV), and rilpivirine (RPV). The synthetic route, comprising 4–6 steps from nicotinate analogues, utilized nucleophilic substitution reaction and the Buchwald-Hartwig reaction as the essential reactions. Among the synthesized compounds, 2-cyanopyridinyl-1,6-naphthyridines 16a, 16b, and 19a exhibited significant inhibitory activity against HIV-1 RT, with IC₅₀ values of 0.222, 0.218, and 0.175 µM, respectively, similar to those of EFV and RPV (0.058 and 0.063 µM). Structural-activity analysis relative to the previously reported compounds indicated that the incorporation of an additional nitrogen atom into the quinoline scaffold markedly improves inhibitory potency. The synthesized naphthyridine derivatives demonstrated significant cytotoxic activity across a broad spectrum of cancer cell lines, with excellent selectivity index (SI) against specific cancers, particularly compound 17a, which exhibited significant cytotoxicity and high selective index against lymphoblastic leukemia cells (MOLT-3), cervical carcinoma cells (HeLa), and promyeloblast cells (HL-60) with IC₅₀ values of 9.1, 13.2, and 8.9 µM, respectively.

Compounds 16a, 16b, and 19a showed no cytotoxicity toward normal embryonic lung cells (MRC-5). Molecular docking investigations revealed that naphthyridines 16a, 16b, and 19a overlapped and aligned within the binding pocket of HIV-1 RT, similar to the commercial medicines (NVP, EFV, ETR, and RPV). In the HIV-1 RT binding pocket, these naphthyridines formed hydrogen bonds with LYS101, PRO225, and PHE227, as well as π-π stacking with TYR181 and TYP229. Additionally, LYS101 and TYR181 were crucial amino acid residues for reverse transcription and mutation in HIV-1 RT. The MD simulation results showed that the complex formed by HIV-1 RT-RPV, HIV-1 RT-16a, HIV-1 RT-16b, and HIV-1 RT-19a had lower RMSD, RMSF, and Rg values. These results indicated that the conformations of HIV-1 RT complexes were more compact and stable during the binding interaction. Pharmacokinetic predictions confirmed that compounds 16a, 16b, and 19a complied with Lipinski’s Rule of Five, suggesting good drug-likeness and potential as anti-HIV-1 agents. Therefore, 2,4-disubstituted-1,6- and 1,7-naphthyridines demonstrated anticancer potential and could enhance the potency of HIV-1 RT inhibition; especially, 2-cyanopyridinyl-1,6-naphthyridine was the essential core structure for novel pharmaceutical compounds in anti-HIV-1 RT agents.