Pyrazolo[3,4-d]pyrimidine-based neplanocin analogues identified as potential de novo pharmacophores for dual-target HBV inhibition

Abstract

The discovery of selective and potent inhibitors through de novo pathways is essential to combat drug resistance in chronic hepatitis B (CHB) infections. Recent studies have highlighted that neplanocin A (NepA) derivatives are biologically selective inhibitors of the hepatitis B virus (HBV). In this study, we designed, synthesized, and evaluated various pyrazolo[3,4-d]pyrimidine-based NepA analogues (4a–h) for their anti-HBV activity. Notably, analogue 4g demonstrated significant activity against HBV replication, with EC₅₀ (HBV DNA) = 0.96 μM, CC₅₀ > 100 μM and EC₅₀ (HBsAg) = 0.82 μM, showing selective inhibition of HBsAg secretion. The SAR analysis concluded that replacing the polar 4-NH₂ group with –CH₃ also acted as a weak H-bonding donor, and the presence of 3-iodo was found to be desirable for the activity/toxicity profile. The nucleoside analogues exhibited a distinct mechanism of action compared to existing nucleoside analogues for the selective inhibition of HBsAg secretion. Based on these findings, compound 4g represents a promising lead molecule for the development of new anti-HBV agents with unique mechanisms of action.

Pyrazolo[3,4-d]pyrimidine-based nucleoside as de-novo pharmacophore as dual-target HBV inhibitor.

1. Introduction

Hepatitis B virus (HBV) is a major concern in the healthcare sector, with no curative therapies currently available.¹ Chronic HBV (CHB) infection further exacerbates the burden on the healthcare sector, particularly in the post-COVID era.² Overall, 350 million patients with chronic liver disease are currently in the high-risk zone during this health emergency. Moreover, patients with chronic liver disease who are on long-term medications are increasingly entering the drug resistance zone.³ There is a pressing need to develop potent and selective HBV inhibitors with novel mechanisms of action. Carbocyclic nucleoside neplanocin A (NepA) has shown broad-spectrum antiviral potential, but its usefulness is limited by associated toxicity.^4,5 Nevertheless, its interesting antiviral endowment has spurred new exploration possibilities for developing improved molecules.⁶ In this pursuit, several modifications to the cyclopentyl sugar (carbocyclic sugar) and base moiety of NepA have been reported in the literature.^7–9 NepA and its derivatives have selectively inhibited several DNA and RNA viruses. Notably, the 5′-homoneplanocin¹⁰ and 7-deazaneplanocin analogues¹¹ have shown potential in inhibiting the replication of both wild-type HBV and lamivudine-resistant mutants. For several years, our research group has focused on rational structural modifications of NepA to generate biologically relevant molecules,^12–14 including HBV analogues.^15–17 Removing or repositioning nitrogen on the nucleobase has proven to be a valid methodology in nucleoside medicinal chemistry, introducing alternative hydrogen bond donors or acceptors and opening new avenues for enzyme-binding sites.¹⁸ Our recent study on base-modified scaffolds AR-II-04-26 (ref. 12) and MK-III-02-03 (ref. 13) concluded that novel NepA analogues are selective inhibitors of HBV, potentially using a de novo mechanism of action that inhibits HBV replication by mimicking the pgRNA activity.¹⁷ In this study, we designed and synthesized a series of targeted nucleosides (4a–h) to explore this possible new target mechanism. This represents a pivotal step in our rational approach to lead optimization, achieved by making various modifications to the base moiety, pyrazolo[3,4-d]pyrimidines, and analyzing the structure–activity relationship. Precisely, the 4-amino group on the base moiety was replaced with weak or less probable H-bond donor groups, such as methyl, cyclopropylamine, and cyclohexylamine, to evaluate antiviral activity and cytotoxicity. The synthesis of these targeted nucleosides and their antiviral evaluation of the carbocyclic nucleoside compounds 4a–h are discussed hereafter.

2. Results and discussion

2.1. Synthesis

Modification of the purine nucleobase was systematically pursued (Scheme 1) using 4-chloropyrazolo[3,4-d]pyrimidine as the starting material, which was synthesized from commercially available allopurinol following a reported literature procedure.^19,20 The electrophilic substitution at the 3-position of pyrazolo[3,4-d]pyrimidine was achieved using N-halosuccinimides (Cl/Br) in DMF to furnish 3-halo derivatives 1a and 1b in quantitative yields. The modified nucleobases (2a–g) (Table 1) were synthesized via nucleophilic substitution reactions. Treatment of 4-chloropyrazolo[3,4-d]pyrimidine with cyclopropylamine and cyclohexylamine afforded compounds 2a and 2d, respectively. Similarly, 3,4-dihalopyrazolo[3,4-d]pyrimidines (1a–b) were subjected to cyclopropylamine and cyclohexylamine, yielding compounds 2b, 2c, and 2d, respectively. The cross-coupling strategy was applied on 4-chloro pyrazolo[3,4-d]pyrimidine using trimethylaluminum Al(CH₃)₃ and tetrakis(triphenylphosphine)palladium (0) Pd(PPh₃)₄ catalyst to generate 4-methyl-1H-pyrazolo[3,4-d]pyrimidines (2f). Compound 2g was achieved from 2f by iodine/KOH in 1,4-dioxane. The cyclopentene carbocyclic sugar precursor (1) was synthesized in eight convergent steps from d-ribose as per an earlier reported protocol.²¹ The typical Mitsunobu coupling²² strategy was applied on the respective nucleobase 2a–g to couple with the carbocyclic ring 1, yielding nucleoside analogues 3a–g (Table 1) as an N-1 regio-isomer¹⁴ with β-conformation in a good yield (79–88%). The deprotection of the trityl and acetonide groups of 3a–g was performed by treatment with 5% methanolic HCl at room temperature, affording carbocyclic nucleosides 4a–g (Table 1) in excellent yields (84–90%). Subsequently, the introduction of the vinyl group onto 4g was accomplished via a Pd-coupling (Stille) reaction^12,23 to afford 4h, as shown in Scheme 2.

Scheme 1. Synthesis of 4-substituted pyrazolo[3,4-d]pyrimidine nucleobase analogues 2a–g. Reaction conditions: i) NXS (X = Cl, Br), DMF, rt, 5 h; ii) cyclopropyl/cyclohexyl amine, 1,4-dioxane, 80 °C, 4 h; iii) Pd(PPh₃)₄, Al(CH₃)₃, THF (dry), 100 °C, 3 h (for details, see the ESI†); iv) I₂, KOH, 1,4-dioxane, 75 °C, 4 h.

Table 1. List of compounds with details of R¹ and R².

Compound	R 1	R 2
2a/3a/4a	–NHC3H5	H
2b/3b/4b	–NHC3H5	Cl
2c/3c/4c	–NHC3H5	Br
2d/3d/4d	–NHC6H11	H
2e/3e/4e	–NHC6H11	Br
2f/3f/4f	–CH3	H
2g/3g/4g	–CH3	I
4h	–CH3	–CH CH2

Scheme 2. Synthesis of neplanocin carbocyclic nucleoside analogues 4a–h. Reaction conditions: i) PPh₃, DIAD, THF, 0–10 °C, 2 h; ii) 5% HCl in MeOH, rt, 1–2 h; iii) Bu₃Sn(–CH CH₂), Pd(PPh₃)₄, DMF, 100 °C, 4 h.

2.2. Anti-HBV activity

The compounds 4a–h were evaluated for anti-HBV assays in HBV replication in HepG2.2.15.7 cells, and their activities were compared to those of the approved anti-HBV drug, lamivudine (3TC). EC₅₀ and CC₅₀ values were calculated from the dose-dependent study using results from the three independent experiments, as summarized in Table 2. Among the tested compounds, 4g emerged as an active lead molecule for further exploration due to its selective inhibition of HBV replication (Fig. 2 and Table 2). Notably, unlike the approved nucleoside analogue entecavir (ETV), compound 4g demonstrated the ability to inhibit HBsAg secretion in HepG2.2.15.7 cells (Fig. 3), suggesting that the mechanism of action of 4g differs from that of the existing anti-HBV nucleoside analogues. In our previous study, compound II had shown to inhibit HBV replication (in a dose-dependent manner) by accelerating the degradation of HBV pregenomeRNA.¹⁷ These findings highlight the unique biological activity of compound 4g and suggest that its mechanism may similarly involve the pgRNA mimic, as the selective inhibition by compound II was not attributed to the inhibition of SAHase.¹⁷ Therefore, further structural modifications of NepA analogues are required to optimize their activity through HBV RNA degradation. However, the exact mechanism of action of 4g remains unknown and will require further elucidation in future studies.

Table 2. Antiviral activity against HBV with CC₅₀ and EC₅₀ in μM.

Compound	EC50 (HBV DNA) in μM	CC50 in μM	EC50 (HBsAg) in μM
4a	>100	>100	ND
4b	>100	>100	ND
4c	>100	>100	ND
4d	>100	>100	ND
4e	>100	>100	ND
4f	>100	>100	ND
4g	0.96	>100	0.82
Compound IIa	0.83	67.8	0.043
4h	>100	>100	ND
3TC	0.031	>10	ND

^aData is taken from Toyama et al.¹⁷ ND = Not determined.

Fig. 2. Anti-HBV activity of compound 4g and lamivudine (3TC).

Fig. 3. Anti-HBsAg activity of compound 4g.

2.3. SAR

It was concluded that weak H-bonding donors at the 4th position, i.e., lipophilic –CH₃ in compound 4g are desirable for the activity compared with the polar NH₂ group (i.e., on analogue II) of the pyrimidine ring where N atom acts as an H-bond acceptor. H-bond hindering groups at the 4-position, i.e., NH-cyclopropyl (compounds 4a–c) and NH-cyclohexyl (compounds 4d–e), resulted in inactivity. However, the bulky halo group I at the 3rd position of the pyrazole ring (R² in prototype-I, Fig. 1), in combination with 4-CH₃ on the pyrimidine ring, was found to be suitable for HBV activity.

Fig. 1. Chemical structures of prototype NepA-based analogues (4a–h) evaluated for anti-HBV activity and cytotoxicity in this study.

3. General procedure for the synthesis of 4a–h

To a mixture of appropriate 2a–g (1.50 mmol), cyclopentene sugar 1 (1.50 mmol) and triphenylphosphine (Ph₃P) (4.50 mmol) in THF was added diisopropyl azodicarboxylate (DIAD) (4.50 mmol) dropwise at 0 °C. The reaction mixture was thereafter brought to room temperature, and stirring was continued. The completion of the reaction was analyzed by TLC, the solvent was evaporated under reduced pressure, and the crude was purified by column chromatography on silica gel by eluting up to 30% EtOAc in hexane to give the couple product (3a–g) in 65–78% yield.

The deprotection of trityl and acetonide groups of 3a–g, was carried out by stirring in methanolic HCl at room temperature. After completion (monitored by TLC), the reaction mixture was concentrated under reduced pressure, and the crude solid was dissolved in methanol, neutralized with NaHCO₃ and the crude was purified by silica gel column chromatography using 5–10% MeOH in CH₂Cl₂ to afford pure carbocyclic nucleoside 4a–g in 85–94% yield. The compound 4h was synthesized from 4g by dissolving it in dry DMF, and to it, tributylvinyltin and Pd(PPh₃)₄ were added subsequently and stirred at 100 °C for 4 h to obtain 4h with an 84% yield (for details, see the ESI†).

(5R)-5-(4-(Cyclopropylamino)-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-3-(hydroxymethyl)cyclopent-3-ene-1,2-diol (4a)

Yield: 87%; white solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 270 nm; ¹H NMR (400 MHz, CD₃OD) δ: 0.69–0.91 (m, 4H), 2.95 (m, 1H), 4.20–4.33 (m, 2H), 4.48 (d, J = 5.4 Hz, 1H), 4.63 (d, J = 5.6 Hz, 1H), 5.78 (s, 2H), 8.10 (s, 1H), 8.31 (s, 1H); ¹³C NMR (100 MHz, CD₃OD) δ: 7.3, 23.1, 59.9, 67.7, 73.5, 76.8, 102.1, 119.2, 125.4, 150.1, 152.5, 154.0, 156.5; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₄H₁₈N₅O₃ [M⁺ + 1]: 304.1410, found: 304.0733.

(5R)-5-(3-Chloro-4-(cyclopropylamino)-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-3-hydroxylmethyl)cyclopent-3-ene-1,2-diol (4b)

Yield: 88%; off white solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 269 nm; ¹H NMR (400 MHz, DMSO-d₆) δ: 0.68–0.82 (m, 4H), 2.96–2.97 (m, 1H), 4.09 (s, 2H), 4.28 (d, J = 5.6 Hz, 1H), 4.37 (t, J = 5.6 Hz, 1H), 4.92–5.07 (m, 3H, D₂O exchangeable, –OH), 5.55–5.64 (m, 2H), 7.29 (bs, 1H, D₂O exchangeable, –NH), 8.32 (s, 1H); ¹³C NMR (100 MHz, DMSO-d₆) δ: 7.1, 8.9, 45.7, 58.9, 66.6, 72.4, 76.7, 98.2, 124.2, 130.3, 150.2, 154.4, 157.0; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₄H₁₇ClN₅O₃ [M + H]⁺: 338.1020, found: 338.0409.

(5R)-5-(3-Bromo-4-(cyclopropylamino)-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-3-(hydroxylmethyl)cyclopent-3-ene-1,2-diol (4c)

Yield: 92%; off white solid. (TLC: R_f. 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 266 nm; ¹H NMR (400 MHz, CD₃OD) δ: 0.68–0.72 (m, 2H), 0.90–0.95 (m, 2H), 2.90–2.95 (m, 1H), 4.28 (d, J = 1.6 Hz, 2H), 4.44 (t, J = 5.6 Hz, 1H), 4.62 (d, J = 5.6 Hz, 1H), 5.75 (s, 2H), 8.32 (s, 1H); ¹³C NMR (100 MHz, CD₃OD) δ: 7.3, 24.3, 59.8, 67.7, 73.5, 77.5, 101.3, 118.4, 125.8, 149.8, 154.6, 156.9, 158.4; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₄H₁₇BrN₅O₃ [M + H]⁺: 382.0515, found: 381.9590.

(5R)-5-(4-(Cyclohexylamino)-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-3-(hydroxylmethyl)cyclopent-3-ene-1,2-diol (4d)

Yield: 86%; off white solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 270 nm; ¹H NMR (400 MHz, CD₃OD) δ: 1.24–2.06 (m, 10H), 4.08–4.10 (m, 1H), 4.28 (s, 2H), 4.46 (t, J = 5.2 Hz, 1H), 4.62 (d, J = 5.2 Hz, 1H), 5.75–5.78 (m, 2H), 8.12 (s, 1H), 8.22 (s, 1H); ¹³C NMR (100 MHz, DMSO-d₆) δ: 25.2, 25.7, 32.8, 49.2, 66.7, 72.6, 76.6, 100.9, 124.8, 132.2, 149.5, 153.4, 155.9, 156.0; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₇H₂₄N₅O₃ [M + H]⁺: 346.1879, found: 346.0681.

(5R)-5-(3-Bromo-4-(cyclohexylamino)-1H-pyrazolo[3,4-d]pyrimidin-1-yl)-3-(hydroxylmethyl)cyclopent-3-ene-1,2-diol (4e)

Yield: 85%; pale yellow solid, (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 269 nm; ¹H NMR (400 MHz, CD₃OD) δ: 1.24–2.06 (m, 10H), 4.16–4.20 (m, 1H), 4.24 (s, 2H), 4.42 (t, J = 5.2 Hz, 1H), 4.62 (d, J = 5.2 Hz, 1H), 5.73–5.76 (m, 2H), 8.24 (s, 1H); ¹³C NMR (100 MHz, CD₃OD) δ: 25.3, 26.0, 32.9, 50.3, 59.8, 67.7, 73.5, 77.5, 100.8, 118.2, 125.8, 149.8, 154.7, 156.5, 157.2; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₇H₂₃BrN₅O₃ [M + H]⁺: 424.0984, found: 424.0199.

(5R)-3-(Hydroxymethyl)-5-(4-methyl-1H-pyrazolo[3,4-d]pyrimidin-1-yl)cyclopent-3-ene-1,2-diol (4f)

Yield: 91%; off white solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 288 nm; ¹H NMR (400 MHz, CD₃OD) δ: 2.82 (s, 3H), 4.30 (s, 2H), 4.53 (t, J = 5.6 Hz, 1H), 4.66 (d, J = 6.0 Hz, 1H), 5.80 (d, J = 1.6 Hz, 1H), 5.94 (d, J = 2.4 Hz, 1H), 8.37 (s, 1H), 8.80 (s, 1H); ¹³C NMR (100 MHz, DMSO-d₆) δ: 22.1, 58.9, 66.7, 72.5, 76.8, 114.1, 124.2, 133.4, 150.1, 152.5, 155.0, 166.5; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₂H₁₅N₄O₃ [M + H]⁺: 263.1144, found: 263.1209.

(5R)-3-(Hydroxymethyl)-5-(3-iodo-4-methyl-1H-pyrazolo[3,4-d]pyrimidin-1-yl)cyclopent-3-ene-1,2-diol (4g)

Yield: 90%, off white solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂); mp: >250 °C; UV (MeOH) λ_max: 290 nm; ¹H NMR (400 MHz, CD₃OD) δ: 3.02 (s, 3H), 4.30 (s, 2H), 4.51 (t, J = 5.6 Hz, 1H), 4.65 (d, J = 6.0 Hz, 1H), 5.79 (d, J = 2.0 Hz, 1H), 5.94 (d, J = 2.8 Hz, 1H), 8.84 (s, 1H); ¹³C NMR (100 MHz, DMSO-d₆) δ: 20.7, 58.9, 67.1, 72.3, 77.0, 93.0, 116.8, 124.2, 150.4, 152.8, 154.6, 163.6; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₂H₁₄IN₄O₃ [M + H]⁺: 389.0111, found: 388.9233.

(5R)-3-(Hydroxymethyl)-5-(4-methyl-3-vinyl-1H-pyrazolo[3,4-d]pyrimidin-1-yl)cyclopent-3-ene-1,2-diol (4h)

Yield: 84%, pale yellow solid. (TLC: R_f 0.1, 10% MeOH in CH₂Cl₂). mp: >250 °C; UV (MeOH) λ_max: 286 nm; ¹H NMR (400 MHz, CD₃OD) δ: 2.88 (s, 3H), 4.30 (s, 2H), 4.57 (t, J = 5.2 Hz, 1H), 4.66 (d, J = 5.6 Hz, 1H), 5.57 (dd, J = 11.2 Hz, J = 2 Hz, 1H), 5.81 (d, J = 1.6 Hz, 1H), 5.95 (d, J = 2.4 Hz, 1H) 6.17 (dd, J = 17.6 Hz, J = 2 Hz, 1H), 7.19 (dd, J = 17.6 Hz, J = 11.2 Hz, 1H), 8.74 (s, 1H); ¹³C NMR (100 MHz, DMSO-d₆) δ: 22.4, 59.8, 67.3, 73.6, 77.5, 112.8, 119.6, 126.2, 127.3, 144.0, 149.5, 154.2, 154.9, 164.4; HRMS (ESI-Orbitrap) m/z: exact mass calculated for C₁₄H₁₇N₄O₃ [M + H]⁺: 289.1301, found: 289.1124.

4. Conclusions

The search for new antiviral molecules structurally close to neplanocin A represents a valid strategy for identifying lead compounds. Rational modification of bioisosteric groups on the pyrazolo[3,4-d]pyrimidine nucleobase has proven to be a productive approach, leading to finding an anti-HBV lead molecule, 4g (HBV DNA EC₅₀ = 0.91 μM, HBsAg EC₅₀ = 0.82 μM, SI > 100). The study concludes that replacing the polar –NH₂ group with the –CH₃ group, which also acts as a weak hydrogen-bonding donor, is desirable for the activity and significantly reduces cytotoxicity. Notably, compound 4g exhibited a better antiviral profile (in terms of toxicity) compared with lamivudine (3TC), an FDA-approved drug. Unlike other FDA-approved nucleoside analogues, 4g inhibited HBsAg secretion in HepG2.2.15.7 cells. Therefore, further structural modification of neplanocin analogues may be required to affect the HBV RNA degradation. Thus, the novel NepA derivatives could help find the specific target enzyme(s) responsible for their unique biological response and mechanism of action in future studies.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

All authors contributed equally.

Conflicts of interest

The authors declare no conflict of interest.