A ‘click chemistry’ approach to the straightforward synthesis of new 4-aryl-1,2,3-triazolocarbanucleosides

The synthesis and biological evaluation as antiviral agents of a series of racemic 4-aryl-1,2,3-triazolyl carbanucleosides of type (±)- 10 /(±)- 11 related to the broad spectrum antiviral agent ribavirin 1 are described. These compounds were produced using a “click chemistry” strategy starting from readily available protected alcohol 13b . The synthetic approach made use of olefin-based organic reactions for the stereoselective construction of the appropriately functionalized cyclopentane ring moiety followed by copper (I) catalyzed Huisgen 1,3-dipolar cycloaddition of azides and alkynes for the regioselective construction of the heterocyclic triazole moiety


Introduction
Ribavirin (Virazole ® , 1, Figure 1) 1 is a broad spectrum antiviral in clinical use for the treatment of RSV infections, lassa fever, hepatitis (A, B, and C), measles and mumps.The structure of this nucleoside analogue consists in a β-D-ribose ring attached to a 1,2,4-triazole derivative replacing the classic purine or pyrimidine base as the aglycon.Other nucleoside analogues owning 5membered heterocyclic bases, such as imidazoles and triazoles, have displayed interesting biological properties; for instance, brenedin (Mizoribine ® , 2, Figure 1) is currently in clinical use as an immunosuppressor for the treatment of transplant patients 2 and EICAR (3, Figure 1), causes depletion of purine nucleotides resulting in a broad spectrum of activity against RNA and DNA viruses and tumour cell proliferation. 3n this context, a great number 1,2,3-triazole derivatives have shown a great potential as antiviral, antibacterial or antiproliferative agents, 4 with 1,2,3-triazolyl nucleoside derivatives such as compounds 4 5 and 5 6 (Figure 1) exhibiting, respectively, interesting antiviral and cytostatic activities.Furthermore, we and others have recently reported several examples of 1,2,3-triazolyl carbanucleosides owning promising biological activities (Figure 2). 7For example, compound (±)-6 (R = 2-C6H4OMe) exhibited specific inhibitory potential against TK + VZV (EC50 = 11 M), 8 7 was found to own moderate activity against HIV-1 (IC50 = 43.8M), 9 and 8 10 displayed a potent antiviral activity against vaccinia virus (EC50 = 0.4 M).In recent times we have started a research program devoted to the synthesis and biological evaluation as antiviral and antitumoral agents of carbanucleosides produced using, in some extent, the postulates of "click chemistry".In our previous work, 8 we reported on a series of racemic 4-aryl-1,2,3-triazolyl 2',3'-dideoxy-2'-iodocarbanucleosides of type (±)-6 (Figure 2) and 4-aryl-1,2,3-triazolyl 2',3'-dideoxy-2',3'-didehydrocarbanucleosides of type (±)-9 (Figure 3).These triazolyl carbanucleosides, structurally related to ribavirin 1, were produced using a iodoazidation reaction as the key step for the stereoselective construction of the desired functionalized cyclopentane ring, followed by the regioselective assembly of the heterocyclic moiety by a Cu(I) catalysed Huisgen 1,3-dipolar cycloaddition. 11e present here our advances on the topic, reporting the synthesis and biological evaluation as antiviral agents of a series of 4-aryl-1,2,3-triazolyl 3'-deoxycarbanucleosides, of type (±)-10 and (±)-11 (Figure 3).These compounds were designed to explore the effect on the antiviral activities of a different pattern of substitution at position 2' of the carbocycle, as well as the modification of the relative stereochemical configuration of the base with regard to the hydroxymethyl group at position 4' of the pseudosugar.

Results and Discussion
Even when the number of references related to the "click chemistry" topic is increasing exponentially since the seminal review by Kolb, Finn and Sharpless, 12 the original postulates are (in most of the cases) restricted to the use of the ubiquitous Cu(I) catalysed Huisgen's 1,3-dipolar cycloaddition. 11In a recent review by Moorhouse and Moses, 13 intelligently entitled: Click Chemistry and Medicinal Chemistry: A Case of "Cyclo-Addiction", the authors stated: It is important to remember that "click chemistry" was originated before the evolution of the Cu(I) catalyst modification of the Huisgen cycloaddition, and that there are other examples of reactions that meet the "click chemistry" criteria, including mainly olefin based reactions.
Taking this considerations into account, we felt that our strategy for the rapid synthesis of carbanucleoside derivatives could not be only restricted to the use of the Huisgen cycloaddition for the construction of the heterocyclic base.For this reason, the designed strategy (Scheme 1) for the synthesis of our target compounds was intended to use olefin-based reactions as the key steps for the synthesis of the corresponding pseudosugar scaffold.As shown in Scheme 1, the synthesis of target compounds (±)-10/(±)-11 could be tackled by epoxidation of readily available protected derivatives of cyclopent-3-enylmethanol of general structure 13, leading to a mixture of epimers 14 that, upon nucleophilic ring opening of the oxirane ring, would lead to a mixture of stereoisomers (±)-15.Construction of the heterocyclic base using Cu(I) catalysed Huisgen 1,3-dipolar cycloaddition, 11  and further deprotection of the corresponding hydroxymethyl group, would lead to the target compounds.This synthetic plan would enable us to prepare several racemic 1,2,3-triazolocarbanucleoside derivatives differing in the relative stereochemical configuration of the functional groups attached to positions 1', 2' and 4' in the cyclopentane ring, with derivatives of type (±)-10 having a cis relationship between the base and the hydroxymethyl group at position 4' (the same relative configuration of natural nucleosides), and derivatives of type (±)-11, having a trans relationship between the above mentioned substituents.
Starting from readily available known alkenes 13a 15 and 13b 8 (Scheme 2), epoxidation of 13a using MCPBA led, after purification by flash column chromatography, to a mixture of cis/trans epimers 14a/b in a 1:8 ratio as previously reported, 14,15  with an improved yield of 91% (69% in the original paper).15  The reaction of 13b with MCPBA produced, after chromatographic purification, an 81% yield of a mixture of cis/trans epimers 14c/d in a 1:3.8 ratio (by 1 H-NMR).16  In both cases, attempts of separation of the mixtures were unsuccessful, leading only to partial resolution of epimers after successive flash column cromatographies.Concerning the nucleophilic ring opening of the resulting epoxides 14a/b and 14c/d, we selected the reaction conditions developed by Crotti et al, 17 that produced mixtures of racemic monoprotected azidoalcohols (±)-15a/b in a 86% yield and (±)-15c/d in a 94% yield (Scheme 2).Resolution of those by flash column chromatography was much easier than for the mixtures of epimeric epoxides 14a/b and 14c/d (see experimental section), so the needed separation of the corresponding isomers was performed at this stage of the synthetic route.
Since we wished to evaluate the effect on the biological activities of the relative configuration of the base and the hydroxymethyl group on position 4' of the cyclopentane ring of target compounds (±)-10/(±)-11, we consequently decided to follow the synthetic plan using the TBDPS diastereomeric precursors ((±)-15c/d), as they could be produced in more equitable quantities than their TBDMS ((±)-15a/b) counterparts.
b Isolated yields after flash column chromatography.c For entries 3 and 4, a 42 % of the starting material (±)-15c was also recovered.
Compounds (±)-10a-c, (±)-11a-c, (±)-16a-c, and (±)-18a/c were also evaluated for its inhibitory activities against Cytomegalovirus (CMV Davis strain) in HEL cell lines, and the results contrasted with those of ganciclovir and cidofovir.Likewise, these compounds were also evaluated for their inhibitory activities against influenza viruses (Influenza A (H1N1/ H3N2 subtypes) and Influenza B) in MDCK cell lines, and the results compared with those of oseltamivir carboxylate and ribavirin.
In all the cases, these 1,2,3-triazolocarbanucleosides did not show any specific antiviral effects (i.e.minimal antiviral effective concentration ≤ 5-fold lower than the minimal cytotoxic concentration for the host cell) against any of the viruses in the assay systems used.
More promising results were obtained in the antiviral evaluation of the silylated derivatives (±)-16a-c and (±)-18a/c against varicella-zoster virus (TK + VZV, thymidine kinase positive strain, and TK − VZV, thymidine kinase deficient strain) in human embryonic lung (HEL) cells.Even when, strictly speaking, no specific antiviral effects were noted, if the data for the TK + VZV (OKA strain) are analysed in more detail, derivative 18a can be interpreted as specifically antivirally active, if based on a comparison of its EC50 = 5.3 M with its MCC > 160 M. 21

Experimental Section
General Procedures.All chemicals used were of reagent grade, obtained from Aldrich Chemical Co. and used without further purification.Starting materials 13a/b were prepared following previously reported methods. 8,15Melting points were measured in a Reichert Kofler Thermopan and are uncorrected.Infrared spectra were recorded in a Perkin-Elmer 1640 FTIR spectrophotometer. 1 H and 13 C NMR spectra were recorded in a Bruker AMX 300 spectrometer at 300 and 75.47 MHz, respectively, using TMS as internal standard (chemical shifts in δ values, J in Hz).Mass spectra were recorded on a Micromass Autoespec (EI and HRMS) and on a Bruker Microtof (ESI-TOF) spectrometers.Microanalyses were performed in a Perkin-Elmer 240B Elemental Analyser at the University of Santiago Microanalysis Service.Analyses indicated by the symbols of elements were within ±0.4% of the theoretical values.Flash chromatography was performed on silica gel (Merck 60, 230-240 mesh) and analytical TLC on pre-coated silica gel plates (Merck 60 F254, 0.25 mm).X-Ray crystal structure determination.Single crystals of (±)-16a suitable for X-ray diffractometry were obtained by dissolving crystals of the already purified material in the minimum quantity of cold THF in an open vial that was then placed in a larger container with a little hexane in its bottom; the container was closed, and after a few days in a cool, dark place free from vibrations afforded the desired single crystals.This were mounted in an inert oil and transferred to the cold gas stream of the diffractometer.Empirical formula: C30H35N3O2Si; formula weight: 497.70; 0.45 × 0.23 × 0.10 mm 3 ; crystal colour: colourless; habit: prismatic; crystal system: triclinic; lattice type: plate; lattice parameters: a = 7.4396( 17 A solution of the corresponding alkene 13a 15 or 13b 8 (1 mmol) in dry CH2Cl2 (5 mL) was slowly added to a well stirred suspension of MCPBA (1.2 mmol) in dry CH2Cl2 (5 mL).The reaction mixture was then heated at reflux until complete disappearance of the starting material was detected by TLC.The solution was then cooled to 0ºC, the solid in suspension filtered and washed with a small amount of cold CH2Cl2.The filtrate was washed with a 10% aqueous solution of Na2SO3 (10 mL), a 10% aqueous solution of NaHCO3 (10 mL) and brine (10 mL).The organic layer was dried (Na2SO4) and the solvent concentrated under reduced pressure.The corresponding residue was purified by flash column chromatography (see below for further details).
Once the reaction was complete, the mixture was cooled to room temperature, diluted with water 50 mL and extracted with ether (3 x 25 mL).The organic layer was dried (Na2SO4) and the solvents evaporated under reduced pressure yielding an oily residue that was purified by flash column chromatography (see below for further details).