Chemistry

In this study, we synthesized two new heterocycle-modified triterpene derivatives (compounds 7 and 8) with a 1,2,3-triazole ring introduced by click chemistry in order to mimic the 1,2,4-triazole-3-carboxamide structure of RBV. This strategy was based on the bioisosteric relationship between both rings established in several papers [32-34]. Studies have made modifications at the C-3 and C-28 positions of triterpenes to synthesize 1,2,3-triazole derivatives via the Huisgen 1,3-cycloaddition reaction, but, as far as we know, this is the first report of the application of click chemistry to triterpenes with this objective [35-37].

Click chemistry is one of the most important tools used for the synthesis of biological compounds, including RBV derivatives [38,39]. Owing to the high yields, accessibility, and low cost, the click chemistry synthetic strategy is a promising option for use in medicinal chemistry studies [40].

The desired compounds 7 and 8 were obtained with yields of 68 and 59%, respectively. Initially, 1-azido-3-nitrobenzene (c) was obtained from m-nitroaniline (a) (Scheme 1) with excellent yields (98%), as previously described [41].

Scheme 1: Synthesis of 1-azido-3-nitrobenzene (c).

The protection of the 3β-OH group of the triterpene skeleton was carried out by acetylation using acetic anhydride to prevent cleavage in acidic conditions, which resulted in the synthesis of derivatives 3 and 4, with 90 and 83% yield, respectively (Scheme 2). The C-28-propargylated triterpene esters (5 and 6, 70% yield for both) of the acetate derivatives (3 and 4) were obtained. Finally, the reaction of derivatives 5 and 6 with meta-nitro-substituted azide (c) using click chemistry resulted in the synthesis of compounds 7 and 8 (Scheme 2).

Scheme 2: Synthesis of the triazole-substituted triterpene derivatives 7 and 8.

Biological assay

The anti-RSV activities of triterpene derivatives, intermediates, and scaffolds were evaluated using MTT and SRB assays in A549 cells treated 4 hours after viral infection (Table 1 and Figure 2). Both assays were compared and resulted in a statistically similar outcome, which showed a robust activity of the compounds tested. Ninety-six hours after infection, the antiviral activity was determined based on the RSV-induced death of the A549 cells and the viability of infected A549 cells after treatment, showing the anti-RSV or protective effect of the compounds. Also, we evaluated the cytotoxic effects of these compounds in VERO, HEP2, A549, and B16F10 cells. Control cells were treated with 1% dimethyl sulfoxide (DMSO), which was used to dilute the test compounds.

Table 1: Antiviral activity and cytotoxic effect of derivatives 1–8.

  VERO HEP2 B16F10 A549 A549+RSV         compound IC50 ± SDa IC50 ± SDa IC50 ± SDa IC50 ± SDa EC50 ± SDb TIc TId TIe TIf 1 14.2 ± 0.2 28.0 ± 0.3 22.7 ± 0.8 17.8 ± 0.6 5.3 ± 0.7 3.4 2.7 5.3 4.3 2 12.9 ± 0.8 26.9 ± 0.4 16.2 ± 0.8 26.7 ± 0.9 17.3 ± 0.9 1.5 0.8 1.6 0.9 3 13.4 ± 2.1 11.5 ± 1.5 10.2 ± 1.1 53.0 ± 0.9 44.4 ± 0.5 1.2 0.3 0.3 0.2 4 12.6 ± 1.2 17.2 ± 0.9 12.4 ± 0.8 133.0 ± 1.1 14.3 ± 0.6 9.3 0.9 1.2 0.9 5 14.5 ± 0.2 18.4 ± 0.6 25.6 ± 0.5 88.8 ± 1 0.6 ± 0.8 148.0 24.2 30.7 42.7 6 26.1 ± 0.6 20.3 ± 0.4 29.0 ± 1.1 67.2 ± 0.5 36.2 ± 0.9 1.9 0.7 0.6 0.8 7 18.9 ± 1.4 23.8 ± 2 18.8 ± 0.7 42.7 ± 0.6 0.3 ± 0.1 142.3 63.0 79.3 62.7 8 21.6 ± 0.9 19.9 ± 1.2 9.8 ± 0.6 59.2 ± 0.9 0.05 ± 0.3 1184.0 432.0 398.0 196.0 RBV nd nd nd nd 4.9 ± 1.4 nd nd nd nd

aConcentration (µM) that is toxic to 50% of non-infected VERO, HEP2, B16F10, and A549 cells by MTT test. bConcentration (µM) that inhibits RSV replication by 50%. cTherapeutic index (TI) = IC50(A549)/EC50(A549 + RSV). dTherapeutic index (TI) = IC50(VERO)/EC50(A549 + RSV). eTherapeutic index (TI) = IC50(HEP2)/EC50(A549 + RSV). fTherapeutic index (TI) = IC50(B16F10)/EC50(A549 + RSV). nd = not determined.

Figure 2: (A) Activity of compound 8 in A549 cells infected with RSV. MTT assay 96 h after treatment. DMSO (0.1%) was used as a negative control. A549 cells were infected with RSV and used as positive controls for infection (PC). Closed boxes represent non-infected A549 cells treated with compound 8. Open boxes represent RSV-infected cells treated with compound 8. (B) Viral load quantification by real-time PCR after treatment with compound 8.

Our results showed that the introduction of a 1-(3-nitrophenyl)-1,2,3-triazol-4-yl substituent in the C-28 position of both compounds 7 and 8 (EC50 = 0.314 and 0.053 µM, respectively) increased their antiviral activity, compared to that of the scaffolds 1 and 2 (EC50 = 5.3 and 17.3 µM, respectively) and acetylated compounds 3 and 4 (EC50= 44.4 and 14.29 µM, respectively). Although betulinic acid (1) exhibited greater antiviral activity than ursolic acid (2), the introduction of a nitroaryl-1,2,3-triazole substituent in 2 was more efficient than its introduction in scaffold 1. Moreover, all tested compounds showed low cytotoxicity in VERO, HEP2, B16F10, and A549 non-infected cells. The therapeutic index (TI) is a comparison between the amount of a compound that causes a therapeutic effect and the amount that causes toxicity. Although all derivatives showed a reasonable cytotoxic effect in non-infected cells, compound 8 was the most efficient of them, with the highest TI of more than 1:1184 (results are summarized in Table 1). Our results are consistent with those of [42] for compounds 1 and 2 and show that our rational design was successful.

As result, the incorporation of a nitroaryl-1,2,3-triazole group into triterpenes resulted in more active and RSV-selective derivatives. The most active derivative (8) had a lower EC50 value against RSV than RBV (0.053 and 4.9 µM, respectively), which suggests that it might be a promising anti-RSV drug candidate. Compound 8 could control viral infection by preventing the proliferation of RSV in A549 cells, compared to the positive control (A549 cells infected with RSV without treatment). Furthermore, derivative 8 had low cytotoxicity in all non-infected cells tested, which is different to that observed for other derivatives where TI was expressively lower (Table 1 and Figure 2A).

The effect of compound 8 on RSV protein F gene expression was investigated using an RT-PCR assay. Total RNA was extracted from RSV-infected cells, both treated and untreated with compound 8 (12.5 and 50 µM). A real-time PCR was performed for the amplification of the RSV protein F gene using specific primers and probes: forward, 5'-AACAGATGTAAGCAGCTCCGTTATC-3'; reverse, 5'-GATTTTTATTGGATGCTGTACATTT-3'; and probe, 5'-FAM/TGCCATAGCATGACACAATGGCTCCT-TAMRA/-3', using human β-actin as an endogenous control gene using the TaqMan assay [43]. The delta cycle-threshold (ΔCt) was obtained by subtracting the endogenous control Ct value from the RSV protein F Ct value. Compound 8 reduced RSV protein F gene expression by approximately 65% at a concentration of 12.5 µM, compared to that of the control (Figure 2B). These results are in accordance with the EC50 value of compound 8, verifying that compound 8 exhibited antiviral activity against RSV.

Molecular docking

Therefore, owing to its excellent level of activity and lack of toxicity, evidenced by a high TI, we selected compound 8 for further studies, starting with the elucidation of the mechanism of action.

Our hypothesis on the study of the mechanism of action relied on a comparison of compound 8 with crystallographic ligands of IMPDH, on the basis that it would represent a secure interpretation of the site interactions similarity with inhibitors, thus, suggesting that this compound acts by the same mechanism. Therein, flexible docking for compound 8 to the IMPDH protein from Mycobacterium tuberculosis (PDB code 4ZQP) was performed. After 10 docking runs, a top-ranked solution was identified and definitively located at the same region occupied by IMP and inhibitor MAD1, ligands of the crystallographic structure in complex with IMPDH. As can be seen in Figure 3, compound 8 (in yellow sticks) interacted through hydrogen bonds with Arg108 and Tyr421, represented by magenta dots, similar to the interactions observed in the crystallographic complex with IMP and the inhibitor MAD1, in cyan sticks. Moreover, the acetate group (at C-3 position) and nitroaryl-1,2,3-triazole (at C-28 position) of compound 8 mimicked the phosphate group and triazole ring of crystallographic ligands, respectively. In addition, the triterpene skeleton of compound 8 was located at the same region occupied by the aromatic rings of both IMP and inhibitor MAD1 ligands (Figure 3). One of the main known mechanisms of action of RBV is the depletion of intracellular GTP pools via the inhibition of cellular IMPDH induced by the 5-monophosphate metabolite of RBV [44].

Figure 3: Superposition of the top-ranked docking solution of compound 8 (carbon atoms in yellow, in stick representation) with the crystallographic poses of IMP and inhibitor MAD1 (color by atoms in stick representation) inside the IMPDH active site from Mycobacterium tuberculosis (cartoon representation – PDB code 4ZQP), with selected residues and interactions here also represented.

Thus, our finding based on these results is that compound 8 may act similarly to IMPDH inhibitors and the active metabolite of RBV, leading to GTP depletion, since best poses achieved for the tested compound interact in an equivalent way as IMP and inhibitor MAD1 and in the same site of action, suggesting that compound 8 can be a bioisostere for IMPDH inhibitors that could be used to combat RSV infections. However, molecular modeling studies of the interaction of derivative 8 with human IMPDH can provide detailed information regarding the likely mechanism of antiviral action of this molecule.