Synthesis of 1-benzyl-3-(3,5-dimethylbenzyl)Uracil Derivatives with Potential Anti-HIV Activity

Abstract

Background:

Nine novel uracil analogues were synthesized and evaluated as inhibitors of HIV-1.

Methods:

Key structural modifications included replacement of the 6-chloro group of 1-benzyl-6-chloro-3-(3,5-dimethylbenzyl)uracil by other functional groups or N¹-alkylation of 3-(3,5-dimethylbenzyl)-5-fluorouracil.

Results:

These compounds showed only micromolar potency against HIV-1 in MT-4, though two of them; 6-azido-1-benzyl-3-(3,5-dimethylbenzyl) uracil and 6-amino-1-benzyl-3-(3,5-dimethylbenzyl) uracil were highly potent (half maximal effective concentration =0.067 and 0.069 μM) and selective (selectivity index =685 and 661), respectively. Structure–activity relationships among the newly synthesized uracil analogues suggest the importance of the H-bond formed between 6-amino group of 6-amino-1-benzyl-3-(3,5-dimethylbenzyl) uracil and amide group of HIV-1 reverse transcriptase.

Conclusions:

We discovered two 6-substituted 1-benzyl-3-(3,5-dimethylbenzyl) uracils, (6-azido-1-benzyl-3-(3,5-dimethylbenzyl) uracil and 6-amino-1-benzyl-3-(3,5-dimethylbenzyl) uracil) as novel anti-HIV agents. These compounds should be further pursued for their toxicity and pharmacokinetics in vivo as well as antiviral activity against non-nucleoside reverse transcriptase inhibitor-resistant strains.

Introduction

Non-nucleoside reverse transcriptase inhibitor (NNRTI), which is not the basic structure of nucleoside, binds at an allosteric binding site that is present in the reverse transcriptase (RT) [1–3]. In 1989, Baba et al. [4] first discovered that 1-[(2-hydroxyethoxy) methyl]-6-phenylthiothymine (HEPT) had a strong anti-HIV-1 activity, though its 5′-triphosphate derivative (HEPT-TP) showed no inhibitory effect on HIV-1 transcriptase. At present, there are a variety of pharmaceutical agents that can be classified as NNRTIs, such as nevirapine (NVP), efavirenz (EFV), delavirdine (DLV) and etravirine (ETR). NNRTIs have a high level of drug safety and good adherence, however, they have a disadvantage which is a tendency to evolve drug-resistant viruses caused by mutations of their binding site in HIV-1 RT. NVP, EFV and DLV bind to similar sites of HIV RT, thus HIV-1 resistant to one of these compounds shows cross-resistance to others; by contrast, ETR appears to bind the multiple sites of RT. As a result, ETR retains the activity against NVP, EFV or DLV-resistant mutants.

Since the discovery of HEPT, related uracil derivatives have been subsequently synthesized. Interestingly, one of them, SJ-3366, has potent antiviral activity against HIV-2 as well as HIV-1 [5]. All of these compounds are 6-substituted uracil derivatives, which are comparatively difficult to synthesize in large quantities; therefore, 1,3-disubstituted uracils have been synthesized and examined for their anti-HIV-1 activity [6]. The introduction of 3-methylbenzyl group at N3-position of uracil was highly effective in inhibiting HIV-1 replication in vitro. Moreover, 3-(3,5-dimethybenzyl) uracil derivatives exhibited an excellent anti-HIV-1 activity. As for the substituent of N¹ position, cyanomethyl group and benzyl group have brought about good results. Based on the estimation that the hydrogen bond accepting nitrogen of the cyanomethyl group might be related to the strong anti-HIV-1 activity, the introduction of 2-picolyl group or 4-picolyl group at N¹-position of the uracil analogue has been investigated and resulted in further elevation of anti-HIV-1 activity. These results were confirmed by a computing science method, in which the docking energy of a substrate on HIV-1 RT has been calculated [7].

An attempt at introducing a methyl or iodo group at C5-position of 1-benzyl-3-(3,5-dimethybenzyl)uracil resulted in significant decrease of anti-HIV-1 activity, suggesting that the bulky group at C5-position of uracil disturbs the binding to HIV-1 RT. However, a fluorine atom is about the same van der Waals radius as a hydrogen atom, and it makes the compound metabolically and chemically stable, thus many pharmaceutical agents possess fluorine atom(s) [8]. These backgrounds prompted us to prepare 1-substituted 3-(3,5-dimethybenzyl)-5-fluorouracils and examine their anti-HIV-1 activity. In addition, we attempt, in this report, to introduce a small group at 6-position of 1-benzyl-3-(3,5-dimethybenzyl)uracil to explore the effect of 6-substitution on the uracil ring.

Methods

Preparation of 6-chloro-1-benzyl-3-(3,5-dimethylbenzyl)uracil (3a)

Synthesis of 6-substituted analogues of 1-benzyl-3-(3,5-dimethylbenzyl)uracil were accomplished from the key compound 6-chloro-1-benzyl-3-(3,5-dimethylbenzyl) uracil (3a) that could be synthesized from commercially available 2,4,6-trichloropyrimidine (2,4,6-TCP; Figure 1). Thus, 2,4,6-TCP was treated with 2.5 M NaOH to give 6-chlorouracil (1) in 74% yield [9]. Then, compound 1 was treated with benzyl bromide to give the corresponding 1-benzylated congeners (2) in 55% yield as previously reported by Ichikawa et al. [10]. This result indicates that the electron-withdrawing chlorine atom facilitates the dissociation of N¹-H to produce an intermediary uracil-1-anion [7]. Compound 2 was condensed with 3,5-dimethylbenzyl alcohol using the Mitsunobu reaction [11] to afford 3-(3,5-dimethylbenzyl) congener (3a).

Figure 1.

Synthesis of 6-chloro-1-benzyl-3-(3,5-dimethylbenzyl)uracil (3a)

Synthesis of 6-substituted-1-benzyl-3-(3,5-dimethylbenzyl)uracil (3b–3f)

Briefly, displacement of the 6-chloro group of 3a was carried out by nucleophilic substitution in suitable solvents to give the 6-substituent derivatives (Figure 2). First of all, Gogoi et al. [12] and Ohkura et al. [13] have independently reported that 6-chloro-1,3-dimethyluracil underwent nucleophilic substitution to give the 6-methoxy- and 6-thiomethoxy-1,3-dimethyluracil under special reactive condition such as phase transfer or photoreaction condition. We examined displacement of 3a with NaOMe in MeOH of 6-chloro derivative (3a) under normal conditions to give 1-benzyl-6-methoxy-3-(3,5-dimethylbenzyl)uracil (3b) in 63% yield. A similar reaction of 3a with sodium thiomethoxide produced the corresponding 6-thiomethoxyl derivative (3c) in 60% yield. These reactions proceed rapidly even at room temperature and the 6-chloro group of 3a has been proved to be facile for nucleophilic substitution. Synthesis of 6-cyano derivative (3d) was carried out in a similar way reported by Nagata and Yoshioka [14]. Thus, compound 3a was treated with potassium cyanide at 50°C overnight to give 3d in 59% yield. As for the azido substitution of 6-chloro function, no report except for the reaction of 6-chloro-1,3-dimethyluracil with sodium azide in MeOH has been published [15]. In a similar way, compound 3a was treated with NaN₃ in DMF to give 6-azido derivative (3e) in moderate yield. The structures of 3d and 3e were confirmed by IR spectra in which characteristic absorptions due to 6-cyano or 6-azido group was observed at 2,231 or 2,133 cm⁻¹, respectively. Then, selective reduction of 6-azido analogue (3e) to 6-amino congener was explored. Hirota et al. [16] reported that reduction of 6-azido-1,3-dimethyluracil to 6-amino derivative using tetralin under thermal condition (150°C, 5 min) in 90% yield. To avoid the decomposition at high-temperature, reduction of 3e was carried out with lithium aluminum hydride to give 6-amino compound 3f in 79% yield. As described above, synthesis of 6-substituted analogues of 1-benzyl-3-(3,5-dimethylbenzyl)uracil (3b–3f) was accomplished from 6-chloro derivative (3a) without any special device or reagents.

Figure 2.

Synthesis of 6-substituted-1-benzyl-3-(3,5-dimethylbenzyl)uracil (3b–3f)

Preparation of 3-(3,5-dimethylbenzyl)-5-fluorouracil (9)

Next, synthesis of 1-substituted analogues of 3-(3,5-dimethylbenzyl)-5-fluorouracil (10a–10d) from 5-fluorouracil was attempted. It is reasonable to prepare these compounds from a key compound, 3-(3,5-dimethylbenzyl)-5-fluorouracil (9). However, it is not easy to introduce an alkyl group at N3 of 5-fluorouracil. Baker et al. [17] reported that alkylation with allylic-type halides, such as allyl bromide and benzyl chloride, occurred at the N¹-position. Our group reported the method to obtain N3-alkylated 5-fluorouracil, in which N3-alkylation of the 5,6-adduct was involved as a key step [18]. We employed this method for the synthesis of 3-(3,5-dimethylbenzyl)-5-fluorouracil (9; Figure 3). Thus, 5-fluorouracil was reacted with N-chlorosuccinimide to give the 5,6-adduct (6) in 58% yield. Compound 6 is supposed to cause steric hindrance at N¹ due to sp³ hybrid orbital of C6 and the bulky methoxy group so that alkylation is preferable at N3-position. Condensation of 6 with 3,5-dimethylbenzyl alcohol using triphenylphosphine and diisopropylazodicarboxylate (DIAD) resulted in the formation of 3-(3,5-dimethylbenzyl) derivative (7; Figure 3). In this reaction, it was difficult to remove impurities by column chromatography or recrystallization process from the product. Therefore, without purification, 3-(3,5-dimethylbenzyl)uracil derivative (7) was reacted with 5% palladium on activated carbon in MeOH under the hydrogen atmosphere. However, only 5-fluorouracil was obtained instead of 8, suggesting reduction of both chloro and 3,5-dimethylbenzyl group and subsequent elimination of the resulting product. Next, reduction of 7 was accomplished by zinc-catalyzed hydrogenation to give the dehalogenated product (8) successfully. However, 8 was so unstable that it gradually transformed into 9 even at room temperature. Accordingly, the crude 8 was subjected to elimination reaction using K₂CO₃ in MeOH to give 3-(3,5-dimethylbenzyl)-5-fluorouracil (9) as white crystals in 74% overall yield from 6 in three steps. The structure of 9 was confirmed as N-3 substituted product by bathochromic shift of UV spectra in which absorption maximum was observed in MeOH at 268 nm, while in 0.1 M NaOH aq, shifts to 299 nm.

Figure 3.

Synthesis of 3-(3,5-dimethylbenzyl)-5-fluorouracil (9)

N¹-alkylation of key compound 9

N¹-Alkylation of the key compound 9 (Figure 3) was achieved by the conventional manner. Thus, compound 9 was alkylated with bromoacetonitrile in the presence of K₂CO₃ in DMF to give the 1-cyanomethyl congener (10a) in 89% yield. In a similar way, 9 was treated with benzyl bromide in DMF to give 10b in 92% yield. Also, alkylation of 9 with 2-(chloromethyl)pyridene or 4-(chloromethyl)pyridene furnished 10c or 10d in 78% or 88% yield, respectively.

Anti-HIV assay

MT-4 cells were maintained in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 U/ml of penicillin G, and 0.10 g/ml of streptomycin. The III_B strain of HIV-1 was used throughout the experiment. The virus was propagated and titrated in MT-4 cells. Virus stocks were stored at −80°C until use. The anti-HIV-1 activity of the test compounds was determined by the inhibition of virus-induced cytopathogenicity in MT-4 cells [19]. Briefly, MT-4 cells (1×10⁵ cells/ml) were infected with HIV-1 at a multiplicity of infection (MOI) of 0.1 and were cultured in the presence of various concentrations of the test compounds. After a 4-day incubation at 37°C in 5% CO₂, the number of viable cells was monitored by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) method [20]. The cytotoxicity of the compounds was evaluated in parallel with their antiviral activity, based on the viability of mock-infected cells, as determined by the MTT method.

Materials

Instrumentation

¹H NMR and ¹³C NMR spectra were taken with a Ultrashield^TM 400 Plus FT NMR System (BRUKER, Yokohama, Japan). Chemical shifts and coupling constants (J) were given in δ and Hz, respectively. Melting points were determined on a Yanaco MP-500D (Yanaco, Tokyo, Japan). Elementary analyses were determined by a Perkin Elmer Series II CHNS/O Analyzer 2400 (Perkin Elmer, Yokohama, Japan). High-resolution mass spectrometry was performed on a APEX IV mass spectrometer (BRUKER) with electrospray ionization mass spectroscopy (ESI-MS).

Compounds

6-Chlorouracil ( 1 )

A mixture of 2,4,6-Trichloropyrimidine (TCI, 13 ml, 113 mmol) and NaOH (18.1 g,0.45 mol, 4 eq) in water (185 ml) was refluxed for 1 h, then added concentrated HCl (ca. 40 ml) over a pH range 2–3, and recrystallized at 4°C. The residue was filtered off, and recrystallized from MeOH to give white crystals (12.26 g, 83.67 mmol, 74%).

Mp>300 °C. 1H NMR (DMSO-d₆): δ 11.20, 12.05 (1 H, br s, NH), 5.72 (1 H, d, H5, J 1.6).

1-Benzyl-6-chlorouracil ( 2 )

6-Chlorouracil (1; 6.79 g, 46.3 mmol) was dissolved in DMF (150 ml) and K₂CO₃ (3.2 g, 23.2 mmol) and benzyl bromide (7.32 ml, 61.6 mmol) were added to the solution, then kept at 70°C for 1 h. Water was added to the mixture and the aqueous mixture was extracted with ethyl acetate. The organic extracts were washed with saturated sodium chloride solution, and dried with sodium sulfate. After removal of the organic solvent, the residual syrup was chromatographed over the column of silica gel (ethyl acetate: hexane =1:1) to give 2 as white crystals (5.97 g, 55%).

HRMS (ESI) Calcd for C₁₁H₉ClN₂NaO₂ [M+Na]⁺: 259.0245. Found 259.0249. Mp 164.6–164.9 °C. ¹H-NMR (DMSO-d₆): δ 11.75 (1 H, br s, NH), 7.26–7.40 (5 H, m, -CH₂C₆H₅), 5.99 (1 H, s, H5), 5.16 (2 H, s, -CH₂C₆H₅).

1-Benzyl-6-chloro-3-(3,5-dimethylbenzyl) uracil ( 3a )

A solution of compound 3 (0.70 g, 2.97 mmol), triphenylphosphine (1.56 g, 5.94 mmol), 3,5-Dimethylbenzylalcohol (787.0 μl, 5.94 mmol) and TMAD (N,N,N',N'-tetramethylazodicarboxamide, 1.02 g, 5.94 mmol) in THF (25.0 ml) was stirred overnight at 50°C. After 12 h stirring, the solution was concentrated to a small volume. The residual solution was purified by silica gel column chromatography (20%→25% ethyl acetate in hexane) to give as a syrup 3a (0.91 g, 2.57 mmol, 86%). HRMS (ESI) Calcd for C₂₀H₁₉ClN₂NaO₂ [M+Na]⁺: 377.1027. Found 377.1037. ¹H NMR (CDCl₃): δ 6.79–7.37 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 5.96 (1 H, s, H5 of uracil), 5.27 (2 H, s, -CH₂), 5.05 (2 H, s, -CH₂), 2.28 (6 H, s, -CH₂×2).

1- Benzyl-3-(3,5-dimethylbenzyl)-6-methoxyuracil (3b)

A mixture of compound 3a (0.52 g, 1.46 mmol) and 28% sodium methoxide in methanol solution (0.48 ml, 2.48 mmol) in MeOH (25.0 ml) was stirred for 0.5 h at room temperature. The mixture was acidified with an acetic acid, and water was added to the mixture and the aqueous mixture was extracted with ethyl acetate. The organic extracts were washed with water, 5% sodium bicarbonate solution and saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was recrystallized from 40% AcOEt in hexane to give white crystals of 3b (0.32 g, 0.92 mmol, 63%).

HRMS (ESI) Calcd for C₂₁H₂₂N₂NaO₃ [M+Na]⁺: 373.1523. Found 373.1535. Mp 93.3–95.7 °C. 1H NMR (CDCl₃): δ 6.84–7.37 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 5.14 (1 H, s, H5 of uracil), 5.07 (2 H, s, -CH₂), 5.05 (2 H, s, -CH₂), 3.85 (3 H, s, -OCH₃), 2.27 (3 H, s, -CH₃), 2.27 (3 H, s, -CH₃).

1-Benzyl-3-(3,5-dimethylbenzyl)-6-thiomethoxyuracil (3c)

A compound 3a (0.50 g, 1.41 mmol) and sodium thiomethoxide (0.13 g, 1.83 mmol) in DMF (10.0 ml) was stirred for 0.5 h at room temperature. The reaction was quenched with AcOH (2.0 ml), and the mixture was evaporated in vacuo and the residue was extracted with benzene. The organic extracts were washed with water, 5% sodium bicarbonate solution and saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residual solution was purified by silica gel column chromatography (33%→50% ethyl acetate in hexane) to white crystals of 3c (0.31 g, 0.85 mmol, 60%).

HRMS (ESI) Calcd for C₂₀H₂₂N₂NaO₂S [M+Na]⁺: 389.1294. Found 389.1305. Mp 119.0–120.4 °C. ¹H NMR (CDCl₃): δ 6.89–7.38 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 5.52 (1 H, s, H5 of uracil), 5.20 (2 H, s, -CH₂), 5.08 (2 H, s, -CH₂), 2.40 (3 H, s, -SCH₃), 2.28 (3 H, s, -CH₃), 2.28 (3 H, s, -CH₃).

1- Benzyl-6-cyano-3-(3,5-dimethylbenzyl) uracil ( 3d )

Compound 3a (0.31 g, 0.86 mmol) was dissolved in DMF (10.0 ml) and potassium cyanide (0.17 g, 2.58 mmol) was added to the solution, then kept at 50°C overnight. After condensation of the solution, the residual syrup was recrystallized from 40% AcOEt in hexane to give white crystals 3d (0.17 g, 0.50 mmol, 59%).

HRMS (ESI) Calcd for C₂₁H₁₉N₃NaO₂ [M+Na]⁺: 368.1370. Found 386.1384. IR (Nujol) cm⁻¹: 2231. Mp 158.1–158.6 °C. ¹H NMR (CDCl₃): δ 7.08–7.68 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 6.88 (1 H, s, H5 of uracil), 5.07 (2 H, s, -CH₂), 4.97 (2 H, s, -CH₂), 2.29 (3H, s, -CH₃), 2.29 (3 H, s, -CH₃).

6-Azido-1-benzyl-3-(3,5-dimethylbenzyl) uracil ( 3e )

A mixture of compound 3a (0.34 g, 0.94 mmol) and sodium azide (0.07 g, 1.03 mmol) in DMF (5.0 ml) was stirred for 0.5 h at room temperature and then water was added to the mixture, and the aqueous mixture was extracted with ethyl acetate. The organic extracts were washed with water, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (25%→33% ethyl acetate in hexane) to give a caramel 3e (0.31 g, 0.85 mmol, 90%).

HRMS (ESI) Calcd for C₂₀H₁₉N₅NaO₂ [M+Na]⁺: 384.1431. Found 386.1439. IR (Nujol) cm⁻¹: 2133. ¹H NMR (CDCl₃): δ 6.89–7.37 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 5.56 (1 H, s, H5 of uracil), 5.06 (2 H, s, -CH₂), 5.04 (2 H, s, -CH₂), 2.27 (3 H, s, -CH₃), 2.27 (3 H, s, -CH₃).

6-Amino-1-benzyl-3-(3,5-dimethylbenzyl) uracil ( 3f )

Compound 3e (0.92 g, 2.53 mmol) was dissolved in dry THF (15 ml), under nitrogen atmosphere. To this stirred solution was carefully added LiAlH₄ (0.10 g, 2.53 mmol) at 0°C. Stirring was continued at 0°C for 0.5 h and quenched by the addition of MeOH (3.0 ml), until no effervescence was observed. Water (20 ml) was then added and the product was extracted with EtOAc. The combined organic extracts were washed with water, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (6.3% MeOH in CHCl₃) to give a caramel 3f (0.67 g, 2.01 mmol, 79%).

HRMS (ESI) Calcd for C₂₀H₂₁N₃NaO₂ [M+Na]⁺: 358.1526. Found 358.1526. ¹H NMR (CDCl₃): δ 6.86–7.43 (8 H, m, -CH₂C₆H₃, -CH₂C₆H₅), 5.17 (2 H, s, -CH₂), 5.09 (2 H, s, -CH₂), 5.00 (1 H, s, H5 of uracil), 4.16 (2 H, br s, -NH₂), 2.29 (6 H, s, -CH₃×2).

5-Chloro-5,6-dihydro-5-fluoro-6-methoxyuracil ( 6 )

5-Fluorouracil (5FU, Sigma-Aldrich, Tokyo, Japan; 6.50 g, 50.0 mmol) was dissolved in MeOH (350.0 ml) and N-Chlorosuccinimide (Sigma-Aldrich, 13.30 g, 100.0 mmol) was added to the solution, and then stirred for overnight at 50°C. The mixture was evaporated, and was recrystallized from 50% EtOH in H₂O to give white crystals of 4 (5.73 g, 29.2 mmol, 58%).

Mp 213.5–215.3 °C. ¹H NMR (DMSO-d₆): δ 11.19 (1 H, br s, 3-NH), 9.21 (1 H, br s, 1-NH), 5.00 (1 H, d, H6, J 4.8), 3.38 (3 H, d, -OMe J 1.2).

3-(3,5-Dimethylbenzyl)-5-fluorouracil ( 9 )

A solution of compound 6 (3.51 g, 17.86 mmol), triphenylphosphine (9.36 g, 35.72 mmol), 3,5-dimethylbenzylalcohol (3.93 ml, 26.79 mmol) and DIAD (diisopropyl azodicarboxylate, WAKO, Osaka, Japan, 3.58 ml, 18.22 mmol) in THF (270.0 ml) was stirred overnight at room temperature. After 12 h stirring, the solution was concentrated to a small volume. The residual solution was purified by silica gel column chromatography (25%→33% ethyl acetate in hexane) to give as a syrup 7 (7.92 g, impurity-containing yellow syrup). Next, Compound 7 (7.92 g) was dissolved in MeOH (200.0 ml) and ammonium chloride (5.39 g, 71.44 mmol) and zinc powder (6.59 g, 71.44 mmol) were added to the solution, and then stirred for 6 h at room temperature. The mixture was extracted with ethyl acetate. The organic extracts were washed with water, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated to give 8 (7.44 g, impurity-containing yellow syrup). Then, 8 was dissolved in MeOH (300.0 ml) and K₂CO₃ (11.02 g, 79.5 mmol) was added to the solution, and then stirred overnight at room temperature. The mixture was filtrated, and was extracted with ethyl acetate. The organic extracts were washed with water, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (40%→50% ethyl acetate in hexane) to give white crystals 9 (3.29 g, 13.27 mmol, 74%).

HRMS (ESI) Calcd for C₁₃H₁₃FN₂NaO₂ [M+Na]⁺: 271.0853. Found 271.0844. Mp 163.0–167.1 °C. UV: max 268.0nm (MeOH) max 298.7nm (NaOH) ¹H NMR (CDCl₃): δ 9.27 (1 H, br s, 1-NH), 7.17 (1 H, s, H6 of uracil), 7.04 (2 H, s, H2+H6 of -C₆H₃), 6.92 (1 H, s, H4 of -C₆H₃), 5.04 (2 H, s, -CH₂C₆H₃), 2.28 (6 H, s, -CH₃×2).

1-Cyanomethyl-3-(3,5-dimethylbenzyl)-5-fluorouracil ( 10a )

Compound 9 (0.25 g, 1.00 mmol) was dissolved in DMF (10.0 ml), and K₂CO₃ (0.42 g, 3.00 mmol) and bromoacetonitrile (0.28 ml, 4.00 mmol) was added to the solution, and then stirred for 2 h at room temperature. The mixture was quenched by the addition of 1.0 M HCl aq (3.0 ml), and then extracted with ethyl acetate. The organic extracts were washed with 5% sodium bicarbonate solution, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (33% AcOEt in hexane) to give white crystals of 10a (0.26 g, 0.89 mmol, 89%).

HRMS (ESI) Calcd for C₁₅H₁₄FN₃NaO₂ [M+Na]⁺: 310.0962. Found 310.0950. Mp 177.1–178.6 °C. ¹H NMR (CDCl₃): δ 7.30 (1 H, d, H6 of uracil, J 6.4), 7.08 (2 H, s, H2+H6 of -C₆H₃), 6.93 (1 H, s, H4 of -C₆H₃), 5.07, (2 H, s, -CH₂), 4.66 (2 H, s, -CH₂), 2.29 (3 H, s, -CH₃), 2.29 (3 H, s, -CH₃).

1-Benzyl-3-(3,5-dimethylbenzyl)-5-fluorouracil ( 10b )

Compound 9 (0.10 g, 0.40 mmol) was dissolved in DMF (4.0 ml), and benzyl bromide (0.19 ml, 1.60 mmol) and K₂CO₃ (0.17 g, 1.20 mmol) was added to the solution, and then stirred for 3 h at room temperature. The mixture was quenched by the addition of 1.0 M HCl aq, and then extracted with ethyl acetate. The organic extracts were washed with 5% sodium bicarbonate solution, saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (20% AcOEt in hexane) to give white crystals of 10b (0.13 g, 0.37 mmol, 92%).

HRMS (ESI) Calcd for C₂₀H₁₉FN₂NaO₂ [M+Na]⁺: 361.1323. Found 361.1320. Mp 103.3–104.8 °C. ¹H NMR (DMSO-d₆): δ 8.36 (1 H, d, H6 of uracil, J 6.4), 7.30–7.38 (5 H, m, -C₆H₄), 6.86 (1 H, s, H4 of -C₆H₃), 6.83 (2 H, s, H2+H6 of -C₆H₃), 4.92, (2 H, s, -CH₂), 4.91 (2 H, s, -CH₂), 2.20 (6 H, s, -CH₃×2).

3-(3,5-Dimethylbenzyl)-1-(2-picolyl)-5-fluorouracil ( 10c )

Compound 9 (0.25 g, 1.00 mmol) was dissolved in DMF (10.0 ml), and K₂CO₃ (1.11 g, 8.00 mmol), NaI (0.03 g, 0.2 mmol), and 2-picolylchloride hydrochloride (0.66 g, 1.60 mmol) was added to the solution, and then stirred for overnight at room temperature. The mixture was quenched by the addition of 1.0 M HCl aq, and then extracted with ethyl acetate. The organic extracts were washed with saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (50% AcOEt in hexane) to give white crystals of 10c (0.26 g, 0.78 mmol, 78%).

HRMS (ESI) Calcd for C₁₉H₁₈FN₃NaO₂ [M+Na]⁺: 362.1275. Found 362.1265. Mp 114.3–115.5 °C. ¹H NMR (DMSO-d₆): δ 8.52 (1 H, ddd, H6 of 2-picolyl, J 4.8, 1.6 and 0.8), 8.34 (1 H, d, H6 of uracil, J 6.4), 7.80 (1 H, ddd, H4 of 2-picolyl, J 7.6, 7.6 and 1.6),7.37 (1H, d, H3 of 2-picolyl, J 7.6) 7.32 (1H, ddd, H5 of 2-picolyl, J 7.6, 4.8 and 0.8) 6.87 (1 H, s, H4 of -C₆H₃), 6.82 (2 H, s, H2+H6 of -C₆H₃), 5.04, (2 H, s, -CH₂), 4.92 (2 H, s, -CH₂), 2.21 (6 H, s, -CH₃×2).

3-(3,5-Dimethylbenzyl)-1-(4-picolyl)-5-fluorouracil ( 10d )

Compound 9 (0.25 g, 1.00 mmol) was dissolved in DMF (10.0 ml), and K₂CO₃ (1.11 g, 8.00 mmol), NaI (0.03 g, 0.2 mmol), and 4-picolylchloride hydrochloride (0.66 g, 1.60 mmol) was added to the solution, and then stirred overnight at room temperature. The mixture was quenched by the addition of 1.0 M HCl aq, and then extracted with ethyl acetate. The organic extracts were washed with saturated sodium chloride solution, and dried with sodium sulfate, and then evaporated. The residue was purified by silica gel column chromatography (33% AcOEt in hexane) to give a caramel 10d (0.30 g, 0.88 mmol, 88%).

HRMS (ESI) Calcd for C₁₉H₁₈FN₃NaO₂ [M+Na]⁺: 362.1275. Found 362.1277. ¹H NMR (DMSO-d₆): δ 8.60 (1 H, dd, H2+H6 of 4-picolyl, J 4.4 and 1.6), 8.43 (1 H, d, H6 of uracil, J 6.4), 7.36 (2 H, dd, H3+H5 of 4-picolyl, J 4.4 and 1.6), 6.93 (1 H, s, H4 of -C₆H₃), 6.91 (2 H, s, H2+H6 of -C₆H₃), 5.02, (2 H, s, -CH₂), 4.98 (2 H, s, -CH₂), 2.27 (6 H, s, -CH₃×2).

Results

Structure-activity relationship

Antiviral activity of the 6-substituted 1-benzyl-3-(3,5-dimethylbenzyl)uracils (3b-3f) and 1-substituted 3-(3,5-dimethybenzyl)-5-fluorouracils (10a–10d) were examined for their inhibitory effects on HIV-1-induced cytopathogenicity and cell viability in MT-4 cells. As comparison, MKC-442 (emivirine), NVP and 4′-ethynylstavudine (4′-Ed4T) were also tested [19–21]. In a series of 6-substituted analogues (3b–3f), 6-methoxy derivative (3b) provided a comparable result to BBF-29, while in 6-thiomethoxy analog (3c), anti-HIV-1 activity was reduced as compared to BBF-29 (Figure 4), suggesting that smaller electronegativity in the chalcogen atom at the 6-position in the uracil skeleton caused reduced anti-HIV-1 activity. Next, introduction of linear group, such as cyano and azido groups at 6-carbon, was investigated for the structure-activity relationship. It is interesting that the 6-azido derivative (3e) showed excellent anti-HIV-1 activity and emerged as the most potent inhibitor in this series with 50% effective concentration (EC₅₀) of 0.067 ±0.011 μM (50% cytotoxic concentration [CC₅₀] 45.9 ±0.7 μM and selectivity index [SI] 685). By contrast, the 6-cyano compound (3d) showed only moderate anti-HIV-1 activity. These wildly divergent results could be explained by the following. Since the steric effect is similar for both groups, the difference presumably comes from the anti-HIV-1 activity of their metabolites. To investigate the possibility that 6-azido derivative (3e) showed strong antiviral activity after converting metabolically to the 6-amino congener, the anti-HIV-1 activity of compound 3f was evaluated and found that it had comparable potency to that of 6-azido derivative (3e) with EC₅₀ of 0.069 ±0.006 μM (CC₅₀ 45.6 ±0.9 μM and SI 661). This result supports our assumption that 3e is reduced to 3f in cell cultures, and the metabolite (3f) inhibits HIV-1 RT. To prove this hypothesis, it is necessary to perform the anti-HIV-1 assay for 3e and 3f in a cell-free system.

Figure 4.

Antiviral activity of 1,3-disubstituted uracils against HIV-1

5-Fluorouracil analogues (10a–10d) had reduced anti-HIV-1 activity and cytotoxicity in MT-4 cell (Figure 4), suggesting that the introduction of fluorine into C5-position of uracil ring results in reduced anti-HIV-1 activity, presumably because the strong electron-withdrawing effect of fluorine causes electron deficiency of uracil ring and reduced affinity to the HIV-1 RT.

Discussion

We have previously demonstrated that some novel 1,3-disubstituted uracils selectively inhibit HIV-1 replication in cell cultures [6,7]. In this report, hydrogen bonding interaction (H-bond) between ligand and amide group of Lys 101 residue as well as the hydrophobic interaction is important for the binding of uracil derivatives to HIV-1 RT. Strong anti-HIV-1 activity of the 6-amino derivative (3f) can be explainable by the H-bond formed between 6-amino group of 3f and amide group of HIV-1 RT. Accordingly, the discrepancy of anti-HIV-1 activity between the 6-azido (3e) and 6-cyano analogues (3d) is caused by the antiviral activity of their metabolite. The 6-azido analogue (3e) may be converted to 6-amino congener (3f) and exert the anti-HIV-1 activity. To develop these compounds for the treatment of HIV-1, the antiviral activity against drug-resistant mutants is important. Therefore, it is interesting to know weather 3e and 3f inhibit the replication of NNRTI-resistant strains.

In conclusion, we discovered two 6-substituted 1-benzyl-3-(3,5-dimethylbenzyl) uracils as novel anti-HIV agents. These compounds should be further pursued for their toxicity and pharmacokinetics in vivo as well as antiviral activity against NNRTI-resistant strains.