Antiviral Chemistry & Chemotherapy 9: Regiospecific ...

A key target enzyme in the search for effective drugs use-ful for AIDS therapy is the human immunodeficiency virus(HIV) reverse transcriptase (RT) (Kaltz & Skalka, 1993;Vaishnav & Wong-Staal, 1991). A great number ofinhibitors of HIV RT have been developed (Schinazi,1993; Young, 1993; De Clercq, 1995). Among the RTinhibitors, the so-called non-nucleoside RT inhibitors(NNRTIs) represent a group of highly potent and specificinhibitors of HIV-1 replication (De Clercq, 1993, 1996)that interact non-competitively with the enzyme at anallosteric and highly hydrophobic non-substrate bindingsite. Mutational, modelling and crystallographic studiessuggest that this particular site is distinct from, but func-tionally and spatially associated with, the substrate bindingsite (Jacobo-Molina et al., 1993; Tantillo et al., 1994;Nanni et al., 1993; Smerdon et al., 1994). This pocket isexclusively found in the RT of HIV-1 and hence NNRTIsare in principle only inhibitory to HIV-1 and not to HIV-2 or other retroviruses (De Clercq, 1993, 1996).

Considerable effort has been focused over the pastdecade on the development of new NNRTIs. Such effortin our group has led to the discovery of the [2′,5′-bis-O-

(tert-butyldimethylsilyl)-β-D-ribofuranosyl]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) nucleosides(TSAO) (Balzarini et al., 1992a,b,c, 1993a,b; Camarasa etal., 1992; Pérez-Pérez et al., 1992). The prototype com-pound of this family is the thymine derivative, designatedTSAO-T (1) (Figure 1). Within the NNRTIs, the TSAOnucleosides occupy a unique position in that they interfereat the interface between the p51 and p66 subunits of RT.Well-defined amino acids at both the p51 and p66 RTsubunits are needed for an optimum interaction of TSAOwith the HIV-1 RT (Balzarini et al., 1993b, 1994;Camarasa et al., 1995; Jonckheere et al., 1994; Boyer et al.,1994). Our experimental data strongly suggest a specificinteraction of the 3′-spiro moiety of TSAO molecules withthe glutamic acid residue at position 138 (Glu-138) of thep51 subunit of HIV-1 RT (Balzarini et al., 1994;Jonckheere et al., 1994; Alvarez et al, 1997).

Structure–activity relationship (SAR) studies with theTSAO class of compounds have revealed that stringentrequirements exist with regard to the structural determi-nants for optimum anti-HIV activity in cell culture. Thesugar part of the TSAO molecules plays a crucial role in

Antiviral Chemistry & Chemotherapy 9: 481-489

Regiospecific synthesis and anti-human immuno-deficiency virus activity of novel 5-substituted N-alkylcarbamoyl and N,N-dialkyl carbamoyl 1,2,3-triazole-TSAO analoguesS Velázquez1, R Alvarez1, C Pérez2, F Gago2, E De Clercq3, J Balzarini3 and M-J Camarasa1*

1Instituto de Química Médica (CSIC), Juan de la Cierva 3, 28006 Madrid, Spain2Departamento de Farmacología, Universidad de Alcalá, 28871 Alcalá de Henares, Madrid, Spain3Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuven, Belgium

*Corresponding author: Tel: +34 91 5 62 29 00 Fax: +34 91 5 64 48 53 E-mail: [email protected]

Several 5-N-alkyl and 5-N,N-dialkylcarbamoylsubstituted analogues of the anti-human immun-odeficiency virus (HIV) type 1 lead compound[1-[2′,5′-bis-O-(tert-butyldimethylsilyl)-β-D-ribofura-nosyl]-5-(N,N-dimethylcarbamoyl)-1,2,3-triazole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-diox-ide) have been prepared and evaluated asinhibitors of HIV-1 replication. A new regiospecific synthetic procedure is described. Thecompounds were prepared by cycloaddition of theappropriate glycosylazide to 2-oxo-

alkylidentriphenyl-phosphoranes, followed bytreatment with primary or secondary amines, toyield, exclusively, 5-substituted 1,2,3-triazole-TSAO analogues. Several 5-substituted 1,2,3-tria-zole-TSAO derivatives proved to be potentinhibitors of HIV-1 replication with higher antivi-ral selectivity than that of the parent TSAO proto-type.

Keywords: AIDS; non-nucleoside HIV-1 RTinhibitors; TSAO-1,2,3-triazoles; 3’-spironucleosides

Introduction

©1998 International Medical Press 0956-3202/98/$17.00 481

the interaction of the TSAO compounds with their targetenzyme. The presence of tert-butyldimethylsilyl (TBDMS)groups at both the C-2′ and C-5′ positions of the riboseand the unique 3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) moiety of the nucleoside in the D-ribo con-figuration are prerequisites for anti-HIV-1 activity(Balzarini et al., 1992c; Camarasa et al., 1992, 1995; Pérez-Pérez et al., 1992; Velázquez et al., 1993, 1994). In con-trast, the nature of the heterocyclic base is less critical foranti-HIV-1 activity (Balzarini et al., 1992c; Pérez-Pérez etal., 1992; Velázquez et al., 1993; Alvarez et al., 1994).However, the role that the base moiety plays in the inter-action of TSAO compounds with HIV-1 RT is, as yet,unclear (Balzarini et al., 1993b; Camarasa et al., 1995).

As part of our ongoing programme to explore theimportance of substituent effects on the anti-HIV-1 activ-ity of TSAO derivatives and, in particular, to determine therole of the nucleobase in this interaction, a series of novelTSAO compounds, in which the base part of the prototypecompound TSAO-T was replaced by 4- and/or 5-substi-tuted 1,2,3-triazoles, were designed, synthesized and eval-uated for their activity against HIV-1 replication (Alvarezet al., 1994). Several members of this class of compoundsshowed potent anti-HIV-1 activity comparable to that ofthe TSAO prototype derivative. Among these derivativesthe 5-substituted amido, methylamido and dimethylamidoTSAO 1,2,3-triazoles 2, 3 and 4 (Figure 1) were endowedwith potent anti-HIV-1 activity [50% effective concentra-tion (EC50) 0.056–0.52 µM]. In particular, compound 4emerged as the most active triazole TSAO derivative andcan be considered as a novel lead compound to develop fur-ther TSAO derivatives with potent anti-HIV-1 activity.When the unsubstituted and methyl- and dimethyl-substi-tuted carbamoyl functions were introduced at C-4, insteadof C-5, of the triazole moiety (5, 6 and 7), the antiviralactivities were decreased by 10-fold (Alvarez et al., 1994).

The molecular basis for the markedly higher antiviral

activities of the 5-substituted compounds 2, 3 and 4, rela-tive to the 4-substituted compounds 5, 6 and 7, was unclearat the time these analogues were synthesized. Subsequentmolecular modelling studies suggested a binding orienta-tion for the triazole ring compatible with these binding dif-ferences and provided new ideas for further structural mod-ifications. The 1,2,3-triazole-TSAO series of compoundsreported herein was conceived on the premise of an addi-tional hydrophobic interaction in a region of the NNRTIbinding site that has actually been explored for otherknown inhibitors (Esnouf et al., 1997).

Our reported procedure for the synthesis of the car-bamoyl substituted-1,2,3-triazole-TSAO analogues(Alvarez et al., 1994) gave the most active 5-isomers asminor products in low yields. An improved synthetic pro-cedure is now reported that guides the synthesis of these1,2,3-triazole 3′-spiro nucleosides exclusively to the desired5-isomers. With the aim of improving the antiviral poten-cy and/or selectivity of these TSAO-triazole analogues andstudying the role of the N-alkyl substituent of the 5-amidomoiety in the biological activity of these new 1,2,3-tria-zole-TSAO analogues, this new synthetic route has beenextended to other 5-substituted N-alkylcarbamoyl andN,N-dialkyl carbamoyl analogues of the new lead com-pound 4. The synthesis and anti-HIV-1 activity of thesenovel 1,2,3-triazole TSAO analogues are described.

Materials and Methods: Chemistry

Microanalyses were obtained with a Heraeus CHN-O-RAPID instrument. 1H NMR spectra were recorded with aVarian Gemini, a Varian XL-300 and a Bruker AM-200spectrometer operating at 300 and 200 MHz with Me4Si asthe internal standard. 13C NMR spectra were recorded witha Bruker AM-200 spectrometer operating at 50 MHz, withMe4Si as the internal standard. Analytical TLC was per-formed on silica gel 60 F254 (Merck). Separations on silica gel

482 ©1998 International Medical Press

S Velázquez et al.

Figure 1. Chemical structures of TSAO-T and 1,2,3-triazole TSAO derivatives 2–7

HN

N

O

S O OSi

SiO

O O

CH3

O

O

H2NO

S O OSi

SiO

O O

H2N

N

NN CONRR'

1 TSAO-T

O

S O OSi

SiO

O O

H2N

N

N

N

2 R; R'; H3 R; H; R'; CH34 R; R'; CH3

CONRR'

5 R; R'; H6 R; H; R'; CH37 R; R'; CH3

were performed by preparative centrifugal circular TLC(CCTLC) on a chromatotron (Kiesegel 60 PF 254 gip-shaltig; Merck), layer thickness 1 mm, flow rate 5 ml/min.Flash column chromatography was performed with silica gel60 (230–400 mesh; Merck). Analytical HPLC was per-formed in Waters Novapak C18 column using 70% CH3CNand 30% H2O, with UV detection at 254 nm and a flow rateof 1 ml/min.

The azide derivative 8 was prepared according to ourreported protocol (Alvarez et al., 1994). The 2-oxoalkyli-den phosphorane 9 was synthesized according to a proce-dure described previously (Le Corre, 1970).

[1-[2′,5′-bis-O-(tert-butyldimethylsilyl)-β-D-ribofu-ranosyl]-5-[(ethyloxy)carbonyl]-1,2,3-triazole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide)(10)A solution of azide 8 (0.50 g, 0.99 mmol) and phosphorane9 (0.45 g, 1.18 mmol) in dry xylene (7 ml) was heated toreflux under N2. After complete disappearance (HPLC) ofthe starting material (approximately 12 h) the solvent wasevaporated in vacuo. The residue was purified by columnchromatography (hexane:ethyl acetate, 3:1) to give 0.32 g(54%) of 10 (Alvarez et al., 1994) as a white foam.

General procedure for the synthesis of [1-[2′,5′-bis-O-(tert-butyldimethylsilyl)-β-D-ribofura-nosyl]1,2,3-triazole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) nucleosides (11–17)A solution of 10 (0.1 mmol) in ethanol (1 ml) was treatedwith the corresponding amine (1 ml). The reaction mixturewas stirred at room temperature until complete reaction ofthe starting material (45 min–24 h). The solution wasevaporated to dryness and the residue was purified byCCTLC on the chromatotron (hexane:ethyl acetate, 1:1).

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-ethylcarbamoyl]-1,2,3-triazole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (11)Following the general procedure a solution of compound10 (0.06 g, 0.1 mmol) in ethanol (1 ml) was reacted withethylamine (1 ml) at room temperature for 24 h.Purification of the residue yielded 0.038 g (63%) of 11 as awhite foam. 1H NMR [(CD3)2CO, 300 MHz] δ –0.30,–0.03, 0.11, 0.17 (4s, 12H, 4CH3Si), 0.73, 0.94 (2s, 18H,2t-Bu), 1.18 (t, 3H, CH3CH2, J 7.2 Hz), 3.40 (m, 2H,CH3CH2), 3.98 (d, 2H, 2H-5′, J4′,5′ 2.6 Hz), 4.38 (t, 1H,H-4′), 5.69 (d, 1H, H-2′, J1′,2′ 7.7 Hz), 5.79 (s, 1H, H-3′′),6.60 (bs, 2H, NH2-4′′), 6.87 (d, 1H, H-1′), 8.13 (bs, 1H,CONH), 8.19 (s, 1H, H-4). Anal. calcd forC24H45N5O7SSi2: C, 47.74; H, 7.51; N, 11.60. Found: C,47.60; H, 7.60; N, 11.51.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-isopropylcarbamoyl]-1,2,3-tria-zole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (12)According to the general procedure, triazole 10 (0.03 g, 0.05mmol) was reacted with isopropylamine (1 ml) at room tem-perature for 24 h. The residue was chromatographed to give12 (0.024 g, 80%) as a white foam. 1H NMR [(CD3)2CO,300 MHz] δ –0.31, –0.02, 0.11, 0.17 (4s, 12H, 4CH3Si),0.73, 0.94 (2s, 18H, 2t-Bu), 1.20, 1.22 (2d, 6H, 2CH3, J 6.6Hz), 3.99 (d, 2H, 2H-5′, J4′,5′ 2.4 Hz), 4.19 (m, 1H, CH),4.38 (t, 1H, H-4′), 5.67 (d, 1H, H-2′, J1′,2′ 7.8 Hz), 5.79 (s,1H, H-3′′), 6.61 (bs, 2H, NH2-4′′), 6.87 (d, 1H, H-1′),7.94 (bd, 1H, CONH), 8.18 (s, 1H, H-4). 13C NMR[(CD3)2CO, 50 MHz] δ –5.97, –5.21, –4.91, –4.41(4CH3Si), 18.45, 19.11 (CH3CSi), 25.82, 26.49 (t-Bu),22.43, 22.48 (2CH3), 42.62 (CH), 63.23 (C-5′), 76.63 (C-2′), 86.58, 86.90, 92.97 (C-1′, C-4′, C-3′′), 94.05 (C-3′),134.07 (C-5), 135.05 (C-4), 152.03 (C-4′′), 157.36(CONH). Anal. calcd for C25H47N5O7SSi2: C, 48.60; H,7.67; N, 11.33. Found: C, 48.49; H, 7.60; N, 11.19.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-cyclopropylcarbamoyl]-1,2,3-tri-azole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (13)Following the general procedure, compound 10 (0.04 g,0.066 mmol) was treated with cyclopropylamine (1 ml) atroom temperature for 24 h. Purification of the residueyielded 0.029 g (75%) of 13 as a white foam. 1H NMR[(CD3)2CO, 200 MHz] δ –0.29, –0.02, 0.11, 0.16 (4s,12H, 4CH3Si), 0.74, 0.94 (2s, 18H, 2t-Bu), 0.57–0.82(m, 4H, 2CH2), 2.91 (m, 1H, CH), 3.99 (d, 2H, 2H-5′,J4′,5′ 2.6 Hz), 4.39 (t, 1H, H-4′), 5.68 (d, 1H, H-2′, J1′,2′

7.8 Hz), 5.77 (s, 1H, H-3′′), 6.55 (bs, 2H, NH2-4′′),6.86 (d, 1H, H-1′), 8.09 (bs, 1H, CONH), 8.14 (s, 1H,H-4). 13C NMR [(CD3)2CO, 50 MHz] δ –6.09, –5.25,–4.96, –4.48 (4CH3Si), 18.40, 19.78 (CH3CSi), 25.74,26.43 (t-Bu), 6.30, 6.34 (2CH2), 23.51 (CH), 63.16 (C-5′), 76.39 (C-2′), 86.48, 88.63, 92.88 (C-1′, C-4′, C-3′′), 94.21 (C-3′), 133.65 (C-5), 135.15 (C-4), 151.89(C-4′′), 159.15 (CONH). Anal. calcd forC25H45N5O7SSi2: C, 48.75; H, 7.36; N, 11.37. Found: C,48.67; H, 7.31; N, 11.30.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-cyclobutylcarbamoyl]-1,2,3-tria-zole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (14)Compound 10 (0.04 g, 0.066 mmol) reacted withcyclobutylamine (1 ml) at room temperature for 24 h. Theresidue was chromatographed to give 0.029 g (71%) of 14as a white foam. 1H NMR [(CD3)2CO, 200 MHz] δ

Novel 5-substituted 1,2,3-triazole-TSAO analogues

483Antiviral Chemistry & Chemotherapy 9:6

–0.31, –0.03, 0.11, 0.16 (4s, 12H, 4CH3Si), 0.73, 0.93(2s, 18H, 2t-Bu), 1.73, 2.10, 2.30 (3m, 6H, 3CH2), 3.98(d, 2H, 2H-5′, J4′,5′ 2.4 Hz), 4.37 (t, 1H, H-4′), 4.51 (m,1H, CH), 5.68 (d, 1H, H-2′, J1′,2′ 7.9 Hz), 5.78 (s, 1H,H-3′′), 6.59 (bs, 2H, NH2-4′′), 6.85 (d, 1H, H-1′), 8.20(s, 1H, H-4), 8.28 (bd, 1H, CONH). Anal. calcd forC26H47N5O7SSi2: C, 49.58; H, 7.52; N, 11.12. Found: C,49.51; H, 7.48; N, 11.09.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-azetidinecarbamoyl]-1,2,3-tria-zole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (15)Following the general procedure, compound 10 (0.042 g,0.07 mmol) was treated with azetidine (1 ml). The reac-tion mixture was stirred at 0°C for 45 min. Purification ofthe residue yielded 0.035 g (81%) of 15 as a white foam.1H NMR [(CD3)2CO, 300 MHz] δ –0.42, –0.16, 0.01,0.03 (4s, 12H, 4CH3Si), 0.61, 0.81 (2s, 18H, 2t-Bu),2.30 (m, 2H, CH2), 3.85 (d, 2H, 2H-5′, J4′,5′ 2.4 Hz),3.99, 4.34 (2m, 4H, 2CH2), 4.23 (t, 1H, H-4′), 5.56 (d,1H, H-2′, J1′,2′ 7.8 Hz), 5.65 (s, 1H, H-3′′), 6.46 (bs, 2H,NH2-4′′), 6.62 (d, 1H, H-1′), 7.98 (s, 1H, H-4). 13CNMR [(CD3)2CO, 50 MHz] δ –5.97, –5.19, –4.94,–4.42 (4CH3Si), 18.47, 19.11 (CH3CSi), 25.81, 26.48 (t-Bu), 16.43 (CH2), 49.33, 53.27 (2CH2), 63.22 (C-5′),76.55 (C-2′), 86.60, 89.04, 92.97 (C-1′, C-4′, C-3′′),93.97 (C-3′), 131.35 (C-5), 135.46 (C-4), 152.09 (C-4′′), 158.62 (CONH). Anal. calcd for C25H45N5O7SSi2:C, 48.75; H, 7.36; N, 11.37. Found: C, 48.87; H, 7.37;N, 11.40.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-pyrrolidinecarbamoyl]-1,2,3-tri-azole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (16)Compound 10 (0.04 g, 0.07 mmol) reacted with pyrroli-dine (1 ml) for 2 h at room temperature according to thegeneral procedure. The residue was purified to give com-pound 16 (0.036 g, 78%) as a white foam. 1H NMR[(CD3)2CO, 50 MHz] δ –0.27, –0.03, 0.12, 0.16 (4s, 12H,4CH3Si), 0.74, 0.93 (2s, 18H, 2t-Bu), 1.97 (m, 4H,2CH2), 3.56, 3.64 (2m, 4H, 2CH2), 3.98 (d, 2H, 2H-5′,J4′,5′ 2.6 Hz), 4.36 (t, 1H, H-4′), 5.69 (d, 1H, H-2′, J1′,2′ 7.8Hz), 5.78 (s, 1H, H-3′′), 6.50 (d, 1H, H-1′), 6.60 (bs, 2H,NH2-4′′), 8.15 (s, 1H, H-4). 13C NMR [(CD3)2CO, 50MHz] δ –6.04, –5.21, –4.98, –4.48 (4CH3Si), 18.43, 19.06(CH3CSi), 25.76, 26.41 (t-Bu), 24.69, 26.72 (2CH2),47.11, 49.26 (2CH2), 63.13 (C-5′), 76.85 (C-2′), 86.53,89.03, 92.66 (C-1′, C-4′, C-3′′), 93.90 (C-3′), 134.13 (C-5), 134.91 (C-4), 151.96 (C-4′′), 157.73 (CONH). Anal.calcd for C26H47N5O7SSi2: C, 49.58; H, 7.52; N, 11.12.Found: C, 49.83; H, 7.74; N, 11.29.

[1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribo-furanosyl]-5-(N-ethylmethylcarbamoyl]-1,2,3-tri-azole]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (17)Following the general procedure, compound 10 (0.08 g,0.13 mmol) was treated with ethylmethylamine (1 ml)and heated to 50°C for 7 h in a sealed tube. The residuewas purified to give compound 17 (0.020 g, 24%) as awhite foam. 1H NMR [(CD3)2CO, 300 MHz] T(˚K) 303δ –0.20, –0.02, 0.16, 0.16 (4s, 12H, 4CH3Si), 0.77, 0.94(2s, 18H, 2t-Bu), 1.21 (m, 3H, NCH2CH3), 3.08, 3.13(2s, 3H, NCH3), 3.45, 3.59 (2m, 2H, NCH2CH3), 3.99(s, 2H, 2H-5′), 4.38 (m, 1H, H-4′), 5.74, 5.76 (2d, 1H,H-2′, J1′,2′ 7.8, 7.7 Hz), 5.81 (s, 1H, H-3′′), 6.16, 6.25(2d, 1H, H-1′), 6.59 (bs, 2H, NH2-4′′), 8.01, 8.07 (2s,1H, H-4). 1H NMR [(CD3)2SO, 300 MHz] T(˚K) 343 δ–0.32, –0.31, –0.10, –0.09, 0.03, 0.05, 0.06 (7s, 12H,4CH3Si), 0.70, 0.85, 0.86 (3s, 18H, 2t-Bu), 1.13 (m, 3H,NCH2CH3), 3.00, 3.02 (2s, 3H, NCH3), 3.27, 3.52 (2m,2H, NCH2CH3), 3.83 (m, 2H, 2H-5′), 4.29 (m, 1H, H-4′), 5.61 (m, 1H, H-2′), 5.83 (s, 1H, H-3′′), 5.95, 6.06(2d, 1H, H-1′, J1′,2′ 7.5, 7.8 Hz), 6.89 (bs, 2H, NH2-4′′),8.15, 8.21 (2s, 1H, H-4). 1H NMR [(CD3)2SO, 300MHz] T(˚K) 353 δ –0.27, –0.06, 0.04, 0.06 (4s, 12H,4CH3Si), 0.72, 0.86 (2s, 18H, 2t-Bu), 1.18 (t, 3H,NCH2CH3, J 7.0 Hz), 3.01 (s, 3H, NCH3), 3.43 (bm, 2H,NCH2CH3), 3.86 (d, 2H, 2H-5′), 4.30 (t, 1H, H-4′),5.62 (d, 1H, H-2′, J1′,2′ 7.6 Hz), 5.77 (s, 1H, H-3′′), 6.03(bd, 1H, H-1′), 6.72 (bs, 2H, NH2-4′′), 8.09 (s, 1H, H-4). Anal. calcd for C25H47N5O7SSi2: C, 48.60; H, 7.67; N,11.33. Found: C, 48.31; H, 7.49; N, 11.20.

Model building and charge derivationThe substituted triazole rings present in molecules 3, 4, 6and 7 were model-built in Insight-II [Insight II, release95.0 (1995), Biosym/Molecular Simulations, San Diego,Calif., USA]. Attachment of a methyl group to the ring N1nitrogen in place of the C1′ of the ribofuranosyl ring ofTSAO produced the reduced models used in the dockingexperiments described below. The geometries obtainedwere optimized at the semi-empirical level using the AM1hamiltonian (Dewar et al., 1985) as implemented in pro-gram MOPAC 93 (Stewart, 1993). Atom centred chargesfor all the fragments were derived by fitting the molecularelectrostatic potential to a monopole–monopole expression(Besler et al., 1990).

Automated docking procedureThe search for the most favourable interactions possiblebetween the triazole moieties of molecules 3, 4, 6 and 7 andthe non-nucleoside binding site of HIV-1 RT was per-formed by means of a Monte Carlo simulated annealingtechnique, as implemented in the program AutoDock

484 ©1998 International Medical Press

S Velázquez et al.

(Goodsell & Olson, 1990, 1996; Morris et al., 1996). Thismethod provides a rapid energy evaluation of each config-uration explored by precalculating atomic affinity poten-tials for carbon, oxygen, nitrogen and hydrogen atoms inthe non-nucleoside binding site using a three-dimensionalgrid (Goodford, 1985). The grid spacing was 0.25 Å andthe grid dimensions 12.5×12.5×12.5 Å3 centred on Leu-A100 of HIV-1 RT, as found in its complex with nevirap-ine (Ren et al., 1995).

Materials and Methods: Virology

Cells and virusesMT-4 cells were kindly provided by N Yamamoto (Tokyo,Japan); CEM cells were obtained from the ATCC(Rockville, MD). HIV-1IIIB was originally obtained fromthe culture supernatant of persistently HIV-1-infected H9cells and was provided by RC Gallo and M Popovic (at thattime at National Institutes of Health, Bethesda, Md.,USA). HIV-2ROD was a kind gift of L Montagnier (PasteurInstitute, Paris, France).

Antiretrovirus assaysCEM cells were suspended at 250000 cells/ml culture medi-um and infected with HIV-1 or HIV-2 at approximately 20and 100 CCID50/ml, respectively. Then, 100 µl of theinfected cell suspension were added to 200 µl plate wellscontaining 100 µl of an appropriate dilution of the test com-pounds. After 4 days of incubation at 37°C, the cell cultureswere examined for giant cell formation. MT-4 cells weresuspended at 300000 cells/ml culture medium and infected

with HIV-1 or HIV-2 at approximately 100 CCID50. Then,100 µl of the infected cell suspension were added to 200 µlplate wells containing 100 µl of the appropriate dilution ofthe test compounds. After 5 days of incubation at 37°C, theviability of the cell cultures was recorded microscopicallyupon trypan blue staining of the cell cultures.

The 50% effective concentration (EC50) was determinedas the compound concentration required to inhibit HIV-induced cytopathicity (giant cell formation) in CEM cellcultures by 50% or as the compound concentrationrequired to inhibit HIV-induced MT-4 cell destruction by50%. The 50% cytotoxic concentration (CC50) was definedas the compound concentration required to inhibit CEMcell proliferation or MT-4 cell viability by 50%.

Results

ModellingThe more favourable binding orientations suggested by theautomated docking procedure for the substituted triazolerings present in molecules 3 and 4 placed the methyl groupsof the amides in two well-defined small hydrophobic pock-ets made up by the side chains of residues Pro-A236, Pro-A225 and Phe-A227, and Phe-A227, Tyr-A188 and Leu-A234 (Figure 2). In this orientation the triazole ring isfound in a cavity delineated by the side chains of residuesLeu-A100, Lys-A103, Val-A106 and Leu-A234, the car-bonyl groups of the main peptide chain of Lys-A101, His-A235 and Pro-A236 and the π ring of Tyr-A318. Thisbinding mode for the triazole part of the TSAO derivativeis compatible with the binding mode proposed for the

Novel 5-substituted 1,2,3-triazole-TSAO analogues

485Antiviral Chemistry & Chemotherapy 9:6

Figure 2. Proposed interaction between the amide substituents on the 1,2,3-triazole moiety of TSAO analogue 4 and thehydrophobic zones delineated by the RT residues shown

Phe-227

Tyr-188

Leu-234Pro-236

Pro-225

other parts of the molecule (Alvarez et al., 1998) and withprevious docking studies performed with the thymine ringof TSAO-T (Velázquez et al., 1998). In contrast, for thesubstituted triazole rings present in molecules 6 and 7 thecomputational search did not produce any valid solutions.On the basis of this possible mechanism of interaction forthe 5-isomers, the present series was designed in anattempt to increase the extent of hydrophobic contact anddispersion interactions with the putative binding pocket.

ChemistryOur two-step reported procedure for the synthesis of thecarbamoyl substituted TSAO-triazoles (Alvarez et al.,1994) involved the 1,3-dipolar cycloaddition of the suitablyfunctionalized and protected ribofuranosyl azide intermedi-ate 8 to unsymmetrical alkyl propiolates to give a mixture ofthe two possible 4- (major) and 5- (minor) ester substitutednucleosides followed by aminolysis of these ester derivativeswith the appropriated amine. Following this approach themost active 5-isomers were obtained as minor products(Alvarez et al., 1994), owing to steric and electronic factorsin the cycloaddition reaction (Bastide et al., 1973; L’abbé &

Hassner, 1971; García-López et al., 1969; Alonso et al.,1980). Therefore, in order to obtain exclusively the mostactive 5-isomers, we devised a new synthetic strategy basedon the reported cycloaddition of alkylazides (Harvey, 1966;L’abbé & Hassner, 1971; Zbiral, 1974) and glycosylazides(Schörkuber & Zbiral, 1980, 1981; Hammerschmidt et al.,1995) to 2-oxo-alkylidentriphenyl-phosphoranes, which,upon concomitant elimination of triphenylphosphaneoxide, yield 5-substituted 1,2,3-triazoles.

Thus, the target 5-N-alkyl carbamoyl substitutedTSAO triazoles were obtained in two steps. Reaction ofthe azide intermediate 8 (Alvarez et al., 1994) (Figure 3)with 2-oxo alkylidentriphenyl phosphorane (9) (Le Corre,1970) in refluxing xylene afforded, exclusively, the 5-sub-stituted 1,2,3-triazole derivative 10 (Alvarez et al., 1994) in54% yield. Compound 10 was used as the common precur-sor for the synthesis of the desired 5-substituted N-alkyland N,N-dialkyl carbamoyl TSAO-1,2,3-triazoles.

Treatment of 10 with ethyl-, isopropyl-, cyclopropyl-and cyclobutyl-amines at room temperature gave the corre-sponding carbamoyl derivatives in good yields [11 (63 %),12 (80%), 13 (75%) and 14 (71%)] (Figure 3).

486 ©1998 International Medical Press

S Velázquez et al.

Figure 3. Synthesis of 5-N-alkylcarbamoyl-1,2,3-triazole TSAO derivatives 11–17

N3O

S O OSi

SiO

O O

H2N

O

S O OSi

SiO

O O

H2N

N

NN

OEt

O

Sugar

N

NN

N

O

H

CH2CH3

Sugar

N

N

NN

O

H

CH(CH3)2

Sugar

N

N

NN

O

H

Sugar

N

N

NN

O

H

Sugar

N

NN

N

O

Sugar

N

NN

N

OSugar

N

NN

N

O

10

CH3

CH2 CH3

8

(9)

Xylene / reflux

11

12

13

14

15

16

17

Ph3 P=CH-CO-CO2Et

Reaction of 10 with the highly reactive secondary amine,azetidine, at 0°C gave the 5-N,N-azetidine carbamoylderivative 15 in 81% yield. Similarly, treatment of 10 withpyrrolidine gave 16 in 78% yield. However, aminolysis of 10with ethylmethylamine was more sluggish and requiredmore severe reaction conditions (50°C for 7 h in a sealedtube) to give the desired derivative 17 in low yield (24%).Finally, attempts to prepare the 5-N,N-diethyl amine car-bamoyl derivative by reaction of 10 with the poorly reactive,diethylamine were unsuccessful and only decompositionproducts were obtained from the reaction mixture.

Structures of the new compounds were assigned on thebasis of the corresponding analytical and spectroscopic data.The 1H NMR spectra of compound 17, in DMSO anddeuterated acetone at room temperature, showed double sig-nals for the amide protons [CON (CH2CH3)(CH3)], theH-4 proton and for the sugar protons H-1′ and H-2′, sug-gesting a possible dynamic process. When the sample washeated at 343 or 353 K (in DMSO) we observed averagedspectra with sharpened resonances corresponding to the pro-posed structure (see Methods) . The dynamic process may bedue to restricted rotation about the amide bond.

Antiviral activityA variety of 1,2,3-triazole TSAO drivatives substituted atthe 5-position of the triazole moiety have been evaluatedfor their inhibitory effects against HIV-1- and HIV-2-induced cytopathicity in CEM and MT-4 cell cultures.The most active TSAO derivatives were those that con-tained either an N-ethyl- or N-cyclopropylcarbamoylfunction [EC50 (MT-4) 0.17 and 0.19 µM, respectively]or an azetidine carbamoyl function at the 5-position ofthe triazole moiety [EC50 (MT-4) 0.14 µM]. These val-ues were, as a rule, slightly lower in CEM cells, and werecomparable with the inhibitory activity of the prototypeTSAO-T and its N,N-dimethylcarbamoyl derivative

(reported earlier). However, in contrast with these refer-ence substances, several of the new compounds, such ascompound 11, were devoid of any cytotoxicity at 250 µM,thus resulting in a selectivity index (SI) (1470) thatexceeded the SI of the prototype compound TSAO-T bysix to sevenfold. Also, compound 14, which was only fourto fivefold less active against HIV-1 than TSAO-T,showed a markedly increased SI (approximately fivefold).Strikingly, a bulky substituent at the nitrogen of the car-bamoyl moiety (N-isopropyl, N-pyrrolidine or N-ethyl-methyl) resulted in a more pronounced loss of antiviralactivity (10-fold as compared to TSAO-T) in MT-4cells. The loss of antiviral activity was less pronounced inCEM cells. None of the test compounds showed anymeasurable anti-HIV-2 activity in MT-4 or CEM cells.

Marked differences were noted in the toxicity of some ofthe new 1,2,3-triazole TSAO derivatives. For example,whereas 16 containing a pyrrolidine carbamoyl substituentwas virtually devoid of cytotoxicity, compound 15 (con-taining an azetidine carbamoyl substituent) was clearlycytotoxic (CC50 14 µM). Also, whereas the 5-N,N-dimethyl carbamoyl triazole derivative of TSAO had aCC50 of 20 µM, the N-monoethyl, N-isopropyl and evenN,N-ethylmethyl derivatives were clearly less cytotoxic.These differences in cytotoxicity obviously influence theSIs of the TSAO derivatives.

Discussion

The differences in activity found among several previouslyreported triazole analogues of TSAO-T with different sub-stitution patterns on C-4 and C-5 led us to study the modeof binding of these molecules. An extensive conformation-al search of possible binding modes highlighted some cleardifferences between the two families of compounds, givingcredence to a model in which the substituents on the amide

Novel 5-substituted 1,2,3-triazole-TSAO analogues

487Antiviral Chemistry & Chemotherapy 9:6

Table 1. Inhibitory activity of TSAO derivatives against HIV-1 and HIV-2 replication in MT-4 and CEM cell cultures

EC50 (µM)*MT-4 CEM CC50 (µM)†

Compound HIV-1 HIV-2 HIV-1 HIV-2 MT-4 CEM SI‡11 0.17±0.02 > 250 0.13±0.04 > 250 ≥ 250 ≥ 250 147012 0.72±0.08 > 250 0.18±0.04 > 50 > 250 > 250 34713 0.19±0.03 > 250 0.08±0.0 > 50 168±20 46±17 88414 0.35±0.21 > 250 0.21±0.09 > 250 > 250 > 250 119015 0.14±0.016 > 50 0.08±0.0 > 50 14±1.8 14±4.1 10116 0.86±0.10 > 250 0.16±0.08 > 50 ≥ 250 ≥ 250 29117 0.72±0.02 > 250 0.33±0.11 > 50 88±31 ≥ 250 126

4 0.06±0.06 > 30 0.12±0.012 > 7 20±7.5 – 333TSAO-T 0.06±0.03 > 20 0.06±0.01 > 20 14±2 – 233

*50% Effective concentration, or compound concentration required to inhibit virus-induced cytopathicity by 50%.†50% Cytotoxic concentration, or compound concentration required to inhibit CEM cell proliferation or MT-4 cell viability by 50%.‡Selectivity index or ratio of CC50 (MT-4) to EC50 (MT-4).

at C-5 are located in two adjacent hydrophobic zones sep-arated by Phe-A227, a binding mode which is not readilyaccessible to the 4-isomers. A new series of analogues pos-sessing different hydrocarbon shapes capping the amide atC-5 was then synthesized using a novel regiospecific syn-thetic procedure in attempts to exploit the compound’sinteractions with this region.

The antiviral potency and selectivity of a number ofnewly synthesized 1,2,3-triazole TSAO derivatives provesthat the antiviral activity of TSAO is not dependent on thepresence of either a pyrimidine or purine moiety in thesenucleoside analogues. Moreover, mainly because of a lowercytotoxic (cytostatic) activity, some of the novel 1,2,3-tria-zole TSAO derivatives show superior antiviral selectivitycompared to the parent prototype TSAO derivatives.

Acknowledgements

We thank Lizette van Berckelaer and Ann Absillis forexcellent technical assistance. This research was supportedin part by grants from the Spanish CICYT (ProjectSAF97-0048-C02-01), the NATO CollaborativeResearch Grant no. CRG 920777 and the BiomedicalResearch Programme of the European Commission (pro-ject BMH4- CT97-2161).

References

Alonso R, Camarasa MJ, Alonso G & De las Heras FG (1980)Alkylating nucleosides. 5. Synthesis and cytostatic activity of N-ribosyl-halomethyl-1,2,3-triazoles. European Journal of MedicinalChemistry 15:105–109.

Alvarez R, Jimeno ML, Pérez-Pérez MJ, De Clercq E, Balzarini J &Camarasa MJ (1997) Synthesis and anti-human immunodeficiencyvirus type 1 activity of novel 3′-spiro nucleoside analogues ofTSAO-T. Antiviral Chemistry & Chemotherapy 8:507–509.

Alvarez R, Jimeno ML, Gago F, Balzarini J, Pérez-Pérez MJ &Camarasa MJ (1998) Novel 3′-spiro nucleoside analogues ofTSAO-T. Part II. A comparative study based on NMR conforma-tional analysis in solution and theoretical calculations. AntiviralChemistry & Chemotherapy 9:123–133.

Alvarez R, Velázquez S, San-Félix A, Aquaro S, De Clercq E,Perno C-F, Karlsson A, Balzarini J & Camarasa MJ (1994) 1,2,3-Triazole-[2′,5′-bis-O-(tert-butyldimethylsilyl-β-D-ribofuranosyl]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) (TSAO)analogues. Synthesis and anti-HIV-1 activity. Journal of MedicinalChemistry 37:4185–4194.

Balzarini J, Pérez-Pérez MJ, San-Félix A, Camarasa MJ, BathurstIC, Barr PJ & De Clercq E (1992a) Kinetics of inhibition ofhuman immunodeficiency virus type 1 (HIV-1) reverse transcrip-tase by the novel HIV-1 specific nucleoside analogue 2′,5′-bis-O-(t-butyldimethylsilyl)-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) thymine (TSAO-T). Journal of Biological Chemistry267:11831–11838.

Balzarini J, Pérez-Pérez MJ, San-Félix A, Schols D, Perno CF,Vandamme AM, Camarasa MJ & De Clercq E (1992b) 2′,5′-Bis-O-(tert-butyldimethylsilyl)-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathi-ole-2′′,2′′-dioxide)pyrimidine (TSAO) nucleoside analogues: high-ly selective inhibitors of human immunodeficiency virus type 1

that are targeted at the viral reverse transcriptase. Proceedings of theNational Academy of Sciences, USA 89:4392–4396.

Balzarini J, Pérez-Pérez MJ, San-Félix A, Velázquez S, CamarasaMJ & De Clercq E (1992c) [2′,5′-Bis-O-(tert-butyldimethylsi-lyl)]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide)(TSAO) derivatives of purine and pyrimidine nucleosides aspotent and selective inhibitors of human immunodeficiency virustype 1. Antimicrobial Agents and Chemotherapy 36:1073–1080.

Balzarini J, Karlsson A, Pérez-Pérez MJ, Vrang L, Walbers J, ZhangH, Öberg B, Vandamme AM, Camarasa MJ & De Clercq E(1993a) HIV-1-specific reverse transcriptase inhibitors show dif-ferential activity against HIV-1 mutant strains containing differentamino acid substitutions in the reverse transcriptase. Virology192:246–253.

Balzarini J, Camarasa MJ & Karlsson A (1993b) TSAO derivatives:Highly specific human immunodeficiency virus type 1 (HIV-1)reverse transcriptase inhibitors. Drugs of the Future 18:1043–1055.

Balzarini J, Kleim JP, Riess G, Camarasa MJ, De Clercq E &Karlsson A (1994) Sensitivity of (138 GLU→ LYS) mutatedhuman immunodeficiency virus type 1 (HIV-1) reverse transcrip-tase (RT) to HIV-1-specific RT inhibitors. Biochemical andBiophysical Research Communications 201:1305–1312.

Besler BH, Merz KM & Kollman PA (1990) Atomic chargesderived from semiempirical methods. Journal of ComputationalChemistry 11:431–439.

Bastide J, Hamelin J, Texier F & Quang YV (1973) Cycloadditiondipolaire-1,3 aux alcynes. Bulletin de la Societe Chimique de France2998–3003.

Boyer PL, Ding J, Arnold E & Hughes SH (1994) Subunit speci-ficity of mutations that confer resistance to nonnucleosideinhibitors in human immunodeficiency virus type 1 reverse tran-scriptase. Antimicrobial Agents and Chemotherapy 38:1909–1914.

Camarasa MJ, Pérez-Pérez MJ, San-Félix A, Balzarini J & DeClercq E (1992) 3′-Spironucleosides (TSAO derivatives), a newclass of specific human immunodeficiency virus type 1 inhibitors:synthesis and antiviral activity of 3′-spiro-5′′-[4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide]pyrimidine nucleosides. Journal ofMedicinal Chemistry 35:2721–12727.

Camarasa MJ, Pérez-Pérez MJ, Velázquez S, San-Félix A, AlvarezA, Ingate S, Jimeno ML, Karlsson A, De Clercq E & Balzarini J(1995) TSAO derivatives: highly specific inhibitors of humanimmunodeficiency virus type-1 (HIV-1) replication. Nucleosidesand Nucleotides 14:585–594.

De Clercq E (1993) HIV-1 specific RT inhibitors: highly selectiveinhibitors of human immunodeficiency virus type 1 that are specif-ically targeted at the viral reverse transcriptase. Medicinal ResearchReviews 13:229–258.

De Clercq E (1995) Towards improved anti-HIV chemotherapy:therapeutic strategies for intervention with HIV infections. Journalof Medicinal Chemistry 38:2491–2517.

De Clercq E (1996) Nonnucleoside reverse transcriptase inhibitors(NNRTIs) for the treatment of human immunodeficiency virustype 1 (HIV-1) infections: strategies to overcome drug resistancedevelopment. Medicinal Research Reviews 16:125–157.

Dewar MJS, Zoebisch EG, Healy E F & Stewart JJP (1985) AM1:A new general purpose quantum mechanical molecular model.Journal of the American Chemical Society 107:3902–3909.

Esnouf RM, Ren J, Hopkins AL, Ross CK, Jones EY, StammersDK & Stuart DI (1997) Unique features in the structure of thecomplex between HIV-1 reverse transcriptase and the bis(het-eroaryl)piperazine (BHAP) U-90152 explain resistance mutationsfor this nonnucleoside inhibitor. Proceedings of the NationalAcademy of Sciences, USA 94:3984–3989.

García-López MT, García-Muñoz G, Iglesias J & Madroñero R(1969) Heterocyclic N-glycosides. III. Synthesis of N-glycosyl-v-

488 ©1998 International Medical Press

S Velázquez et al.

triazoles from glycosyl azides and phenylacetylenes. Journal ofHeterocyclic Chemistry 6:639–642.

Goodford PJ (1985) A computational procedure for determiningenergetically favorable binding sites on biologically importantmacromolecules. Journal of Medicinal Chemistry 28:849–857.

Goodsell DS & Olson AJ (1990) Automated docking of substratesto proteins by simulated annealing. Proteins Structure Function andGenetics 8:195–202.

Goodsell DS, Morris GM & Olson AJ (1996) Docking of flexibleligands: applications of autoDock. Journal of Molecular Recognition9:1–15.

Harvey GR (1966) The reactions of phosphorous compounds. XII.A new synthesis of 1,2,3-triazoles and diazo esters from phospho-rous ylids and azides. Journal of Organic Chemistry 31:1587–1590.

Hammerschmidt F, Polsterer JP & Zbiral E (1995) Synthesis of 1-(D-apio-β-D-furanosyl)-1,2,3-triazoles. Synthesis 415–418.

Jacobo-Molina A, Ding J, Nanni RG, Clark AD Jr, Lu X, TantilloC, Williams RL, Kamer G, Ferris AL, Clark P, Hizi A, HughesSH & Arnold E (1993) Crystal structure of human immunodefi-ciency virus type 1 reverse transcriptase complexed with double-stranded DNA at 3.0 Å resolution shows bent DNA. Proceedingsof the National Academy of Sciences, USA 90:6320–6324.

Jonckheere H, Taymans JM, Balzarini J, Velázquez S, Camarasa MJ,Desmyter J, De Clercq E & Anné J (1994) Resistance of HIV-1reverse transcriptase against [2′,5′-bis-O-(tert-butyldimethylsilyl)-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide)] (TSAO)derivatives is determined by the mutation Glu138 → Lys on thep51 subunit. Journal of Biological Chemistry 269:25255–25258.

Katz RA & Skalka AM (1993) The retroviral enzymes. AnnualReview of Biochemistry 63:133–173.

L’abbé G & Hassner A (1971) Stereochemistry XLIV.Regiochemistry in cycloaddition reactions of vinylazides withphenylacetylene. Bulletin des Societes Chimiques Belges 80:209–210.

Le Corre M (1970) Nouvelle voie d’accés aux esters α-cétoniques β-éthyléniques. Comptes Rendus de L’Academie des Sciences Serie C270:1312–11314.

Morris GM, Goodsell DS, Huey R & Olson AJ (1996) AutoDock:automated docking of flexible ligands to receptors, version 2.4. La Jolla:The Scripps Research Institute.

Nanni RG, Ding J, Jacobo-Molina A, Hughes SH & Arnold E(1993) Review of HIV-1 reverse transcriptase three-dimensionalstructure: implications for drug design. Perspectives in DrugDiscovery and Design 1:129–1150.

Pérez-Pérez MJ, San-Félix A, Balzarini J, De Clercq E & CamarasaMJ (1992) TSAO-analogues. Stereospecific synthesis and anti-HIV-1 activity of 1-[2′,5′-Bis-O-(tert-butyldimethylsilyl)-β-D-ribofuranosyl]-3′-spiro-5′′-(4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide) pyrimidine and pyrimidine-modified nucleosides. Journalof Medicinal Chemistry 35:2988–2995.

Ren JS, Esnouf R, Garman E, Jones Y, Somers D, Ross C, Kirby I,Keeling J, Darby G, Stuart D & Stammers D (1995) High resolu-tion structure of HIV-1 RT: insights from four RT–inhibitorcomplexes. Nature Structural Biology 2:293–302.

Schörkuber W & Zbiral E (1980) Glycosylazide als Ausgangsbasiszur Gewinnung von Nucleosidanalogen. Liebigs Annalen1455–1469.

Schörkuber W & Zbiral E (1981) Zur synthese von 5-(x-methyl)ribofuranosyl-1,2,3-triazol-nucleosides.Chemische Berichte114:3165–3169.

Smerdon SJ, Jäger J, Wang J, Kohlstaedt LA, Chirino AJ, FriedmanJM, Rice PA & Steitz TA (1994) Structure of the binding site fornon-nucleoside inhibitors of the reverse transcriptase of humanimmunodeficiency virus type 1. Proceedings of the National Academyof Sciences, USA 91:3911–3915.

Schinazi RF (1993) Competitive inhibitors of human immunodefi-ciency virus reverse transcriptase. Perspectives in Drug Discovery andDesign 1:151–180.

Stewart JJP (1993) MOPAC 93. Tokyo: Fujitsu.

Tantillo C, Ding J, Jacobo-Molina A, Nanni RG, Boyer PL,Hughes SH, Pauwels R, Andries K, Janssen PAJ & Arnold E(1994) Locations of anti-AIDS drug binding sites and resistancemutations in the three-dimensional structure of HIV-1 reversetranscriptase. Implications for mechanisms of drug inhibition andresistance. Journal of Molecular Biology 243:369–387.

Young S D (1993) Nonnucleoside inhibitors of HIV-1 reverse tran-scriptase. Perspectives in Drug Discovery and Design 1:181–192.

Vaishnav YN & Wong-Staal F (1991) The biochemistry of AIDS.Annual Review of Biochemistry 60:577–630.

Velázquez S, Chamorro C, Pérez-Pérez MJ, Alvarez R, Jimeno ML,Martín-Domenech A, Pérez C, Gago F, De Clercq E, Balzarini J,San-Félix A & Camarasa MJ (1998) Abasic analogues of TSAO-T as the first sugar derivatives that specifically inhibit HIV-1reverse tanscriptase. Journal of Medicinal Chemistry 41:4636–4647

Velázquez S, Jimeno ML, Camarasa MJ & Balzarini J (1994)Synthesis of [1-[3′,5′-bis-O-(tert-butyldimethylsilyl-β-D-arabino-and ribofuranosyl]cytosine]-2′-spiro-5′′-(4′′-amino-1′′,2′′-oxathi-ole-2′′,2′′-dioxide). Analogues of the highly specific anti-HIV-1agent TSAO-T. Tetrahedron 50:11013–11022.

Velázquez S, San-Félix A, Pérez-Pérez MJ, Balzarini J, De Clercq E& Camarasa MJ (1993) TSAO analogues. 3. Synthesis and anti-HIV-1 activity of [2′,5′-bis-O-(tert-butyldimethylsilyl-β-D-ribofu-ranosyl]-3′-spiro-5′′-[4′′-amino-1′′,2′′-oxathiole-2′′,2′′-dioxide]purine and purine-modified nucleosides. Journal of MedicinalChemistry 36:3230–3239.

Zbiral E (1974) Synthesen von heterocyclen mit hilfe von alkyliden-phosphoranen phosphin-imiden, vinyl phosphonium-salzen undvinylphosphin-oxiden. Synthesis 775–800.

Novel 5-substituted 1,2,3-triazole-TSAO analogues

489Antiviral Chemistry & Chemotherapy 9:6

Received 30 July 1998; accepted 1 September 1998