Assembly of CF3-Pyrazole-Triazole Hybrids through (3 + 3)-Cycloaddition/Ring Contraction and Click Chemistry.

We commenced our study with 1-aryl-3-trifluoromethylpyrazoles 3 , prepared according to the protocol recently developed in our laboratory. The method relies on a one-pot (3 + 3)-cycloaddition of mercaptoacetaldehyde with the nitrile imine generated in situ via base-mediated dehydrohalogenation of the corresponding bromides 2 , followed by a spontaneous Eschenmoser-type ring contraction of the first formed 1,3,4-thiadiazine intermediate ( Scheme ). Thus, starting from 1,4-dithiane-2,5-diol ( 4 ) (the dimer of mercaptoacetaldehyde) and nitrile imine precursors 2a – 2g , we obtained a series of eight model pyrazoles 3a – 3g bearing selected para substituents, i.e., X = Me (96%), i Pr (85%), OMe (92%), OBn (97%), Cl (86%), CF 3 (70%), and CN (71%). All products were isolated in high yield. In the synthesis of the key azides 5 , we benefited from the pronounced acidity of the C(5)– H in pyrazoles 3 . Selective deprotonation with a slight excess of n -BuLi at −78 °C furnished the corresponding lithium 5-pyrazolides, which upon treatment with p -toluenesulfonyl azide as an N 3 -transferring reagent underwent smooth functionalization to afford the desired series of 5-azidopyrazoles 5a – 5g . No significant influence of electronic effects on the reaction efficiency was observed; pyrazoles with aryl residues bearing strongly electron-donating ( 5c ; X = OMe) and strongly electron-withdrawing ( 5f ; X = CF 3 ) substituents delivered the expected azides in 94% and 80% yield, respectively. The chloro- and cyano-substituted substrates also provided the desired azides 5e (60%) and 5g (37%) in somewhat lower yields, presumably due to side reactions with n -BuLi (i.e., halogen–lithium exchange and nucleophilic addition, respectively). In the latter case, the formation of an unidentified byproduct was noted; despite repeated attempts, it could not be removed by standard chromatography or recrystallization due to limited stability of 5g , and the material (ca. 90% purity) was therefore used in the next step without further purification. The reaction proved to be readily scalable, delivering 5c (X = OMe) in a comparable 90% yield when performed on a gram scale (1.9 g, 6.6 mmol). For the initial click experiments, we employed azide 5a and phenylacetylene ( Scheme ). A brief screening of literature-reported conditions, , during which the solvent (DCM, MeCN, aqueous MeOH) and the copper source (CuSO 4 , (AcO) 2 Cu, CuI/DIPEA) were evaluated, allowed us to identify optimal reaction parameters. Thus, running the reaction with a slight excess of the alkyne and a CuSO 4 /ascorbate catalytic system, in a MeOH/H 2 O (10:1 mixture) at 55 °C, ensured complete consumption of the starting azide. The analytically pure sample of the expected (3 + 2)-cycloadduct 1aa was isolated in high 85% yield after simple filtration of the crude product through a short plug of silica gel, followed by recrystallization from hexanes. With these conditions in hand, a first set of phenylacetylene-derived products 1aa – 1ag were obtained in good yields (67–86%), regardless of the electronic nature of the substituents or functional groups present in the starting azides. Next, we examined the scope with respect to the alkyne reaction partners using 4-methoxyphenyl-substituted azide 5c ( Scheme ). A series of para -substituted phenylacetylenes was checked, affording the corresponding 1,2,3-triazoles bearing primary amino ( 1ba ), alkyl ( 1bc - 1bd , Me, and n-Pent), halogen ( 1be , Cl), or cyano ( 1bf ) groups. Introduction of an electron-rich OMe donor at the para , meta , or ortho position of the aromatic ring exerted only a marginal influence on the reaction efficiency, with the ortho -substituted product 1bg isolated in 91% yield and an excellent 79% overall yield (for three steps). The structure of para -functionalized isomer 1bb was unambiguously confirmed by X-ray diffraction (CCDC 2425718 ). Both increased substitution, as in 1bi (2,4,6-trimethyl), and extension of the π system, as in naphthyl derivative 1bj , were well-tolerated. Electron-deficient fluorinated phenylacetylenes bearing one or two substituents, i.e., fluorine atoms or fluoroalkyl groups (CF 3 , OCF 3 ), did not hamper the reaction, consistently providing products 1bk – 1bo in yields exceeding 80%. In the case of 1,4-diethynylbenzene, the applied conditions furnished a mixture of two products, 1bt (53%) and 1bu (15%). Fine-tuning of the reactant stoichiometry provided excellent control over the reaction outcome, enabling the selective formation of the corresponding monocycloadducts (using 1.6 equiv of alkyne) and biscycloadducts (using 2.5 equiv of azide), which were isolated in 83% and 56%, respectively. Incorporation of pyridyl and thienyl units, as exemplary six- and five-membered heteroaryl substituents, furnished 1br and 1bs in 73% and 87% yield, respectively. Moreover, reaction of an acetylene bearing ferrocene moiety, selected as a representative organometallic counterpart, also proceeded smoothly to afford the expected cycloadduct ( 1bq , 66%). Aliphatic alkynes were checked as well; the (3 + 2)-cycloaddition reactions of 1-octyne and tert -butylacetylene delivered the expected products 1ca and 1cb , although in the latter case, the sterically crowded product was obtained in diminished yield. Trimethylsilylacetylene was used as a surrogate for acetylene; due to spontaneous cleavage of the C–Si bond during the aqueous workup, the desired hybrid 1cc , lacking a substituent at C(4) of the 1,2,3-triazole ring, was isolated. Introduction of functional groups such as hydroxymethyl, diethoxymethyl, and methoxycarbonyl was readily accomplished by employing the respective acetylenes, although product 1ce bearing a masked carbonyl group was accessed in a markedly lower yield (35%) due to partial hydrolysis of the acetal moiety. In addition, two multifunctional building blocks, BMK-alkyne and hydroxyl-diazirine-alkyne, often used as probes for selective modification of biological targets, were checked in reactions with model azide 5c under the optimal reaction conditions (CuSO 4 /sodium ascorbate, MeOH/H 2 O, 55 °C, 5 h). Accordingly, new derivatives 1cg (41%) and 1ch (81%) suitable for potential 19 F labeling of biomolecules were obtained. To further assess the synthetic potential of 5-azido-CF 3 -pyrazole 5c in the modification of more complex systems, a series of well-known pharmaceuticals bearing terminal alkyne functionality was involved in the study ( Scheme ). Two monoamine oxidase inhibitors, pargyline and rasagiline, featuring secondary and tertiary amine groups, respectively, provided the expected cycloadducts 1da (68%) and 1dc (77%). Notably, the optical purity of the latter substrate remained unchanged during the (3 + 2)-cycloaddition step, leading to an enantiopure pyrazole–triazole hybrid. Moreover, erlotinib, a tyrosine kinase inhibitor used as first-line therapy for advanced nonsmall-cell lung cancer, as well as gestrinone, a synthetic steroid hormone approved for the treatment of endometriosis, provided the expected cycloadducts in excellent yields of 84% ( 1db ) and 87% ( 1dd ). The presence of hydroxyl and cyano functionalities in the pyrazole–triazole hybrids 1bp and 1bf , respectively, offers several opportunities for postcyclizative modifications of the heterocyclic framework ( Scheme ). For example, straightforward acylation of the hydroxy group in 1bp with Ac 2 O afforded ester 1bv in 91% yield, whereas oxidation with the Corey–Suggs reagent furnished the corresponding aldehyde 1bw almost quantitatively. The methoxycarbonyl moiety in 1bx was introduced efficiently by permanganate-mediated oxidation in wet MeCN, followed by Fischer esterification, giving the product in an overall 93% yield. a Yields refer to analytically pure samples obtained by chromatography, followed by recrystallization. b Using 1.6 equiv of 1,4-diethynylbenzene. c Using 2.5 equiv of azide 5c . d TMS-acetylene was used as a reaction partner. e 1.5 equiv of azide 5c , reaction time of 40 h. f 1.5 equiv of azide 5c , reaction time of 16 h. Yields refer to analytically pure samples obtained by chromatography, followed by recrystallization. Using 1.6 equiv of 1,4-diethynylbenzene. Using 2.5 equiv of azide 5c . TMS-acetylene was used as a reaction partner. 1.5 equiv of azide 5c , reaction time of 40 h. 1.5 equiv of azide 5c , reaction time of 16 h. For the transformation of the cyano group into the primary amide, we employed classical conditions of the Radziszewski-type reaction, relying on H 2 O 2 -induced hydrolysis under basic conditions. Thus, treatment of 1bf with an excess of hydrogen peroxide in the presence of Na 2 CO 3 at room temperature furnished amide 1by in an excellent yield of 94%. Exhaustive reduction of the cyano group, proceeding via reductive deamination of the initially formed amine, was observed upon catalytic hydrogenation at elevated pressure (70 psi) using palladium on charcoal as the catalyst. The corresponding product bearing the p -tolyl substituent was obtained quantitatively, and its identity with the original sample 1bc , prepared independently through azide–alkyne (3 + 2)-cycloaddition reaction ( Scheme ) was confirmed by NMR analysis. The desired amine was accessible under milder hydrogenation conditions (H 2 , slight positive pressure from balloon) employing Raney-Ni as the catalyst; the first formed product was subsequently acylated with isobutyryl chloride to furnish the corresponding amide 1bz in a fair overall yield of 74%. These experiments clearly demonstrate the remarkable stability of the designed 1-aryl-5-(1,2,3-triazol-1-yl)-3-CF 3 -pyrazole core under both harsh reductive and oxidative conditions and establish these pyrazole hybrids as robust building blocks for construction of more elaborate molecular architectures. a Ac 2 O, DCM, 40 °C, 90 min. b PCC, DCM, rt, 3 h. c KMnO 4 , MeCN, 80 °C, 2 h, then MeOH, H 2 SO 4 , reflux 12 h. d H 2 O 2 , Na 2 CO 3 , acetone rt, 2 h. e H 2 (70 psi), Pd/C, THF, rt, 5 h. f H 2 , Raney-Ni, NH 3 aq, rt, 2 h, then i-PrCOCl, Et 3 N, THF, rt, 1 h. Ac 2 O, DCM, 40 °C, 90 min. PCC, DCM, rt, 3 h. KMnO 4 , MeCN, 80 °C, 2 h, then MeOH, H 2 SO 4 , reflux 12 h. H 2 O 2 , Na 2 CO 3 , acetone rt, 2 h. H 2 (70 psi), Pd/C, THF, rt, 5 h. H 2 , Raney-Ni, NH 3 aq, rt, 2 h, then i-PrCOCl, Et 3 N, THF, rt, 1 h. Finally, to check whether the isomeric hybrids bearing the 1,2,3-triazole moiety at C(4) of the pyrazole core could be accessed, a plausible C(4)-iodination/lithiation/azide-transfer sequence was investigated. As depicted in Scheme , the representative substrate 3a was converted in fully regioselective fashion into the corresponding iodide 7 using a slight excess of elemental iodine and ceric ammonium nitrate (CAN) as a mild oxidant. To our delight, the subsequent iodine–lithium exchange in 7 (X = I) proceeded smoothly, and after N 3 transfer the expected azide 8 (X = N 3 ) was obtained, albeit in a moderate yield of 42%. Cu-catalyzed (3 + 2)-cycloaddition of 8 with phenylacetylene, furnishing the 4-(1,2,3-triazol-1-yl)­pyrazole derivative 9 (68%), thus demonstrated the feasibility of the designed approach. a I 2 (1.3 equiv), CAN (1.1 equiv), MeCN, reflux, 16 h. b n BuLi, THF, −78 °C, then TsN 3 rt, 4 h. c Ph−C≡CH, CuSO 4 , sodium ascorbate, MeOH/H 2 O (10:1), 55 °C, 5 h. I 2 (1.3 equiv), CAN (1.1 equiv), MeCN, reflux, 16 h. n BuLi, THF, −78 °C, then TsN 3 rt, 4 h. Ph−C≡CH, CuSO 4 , sodium ascorbate, MeOH/H 2 O (10:1), 55 °C, 5 h.

In summary, a method for the rapid assembly of trifluoromethylated pyrazole–triazole hybrids inspired by the structural motifs of numerous biologically relevant 3-CF 3 -pyrazole derivatives of practical significance is presented. By employing a three-step sequence comprising a (3 + 3)-cycloaddition/Eschenmoser-type ring contraction cascade affording 1-arylpyrazoles, followed by C(5)-selective azidation and Cu-catalyzed Huisgen cycloaddition, a broad set of variously functionalized hybrids was obtained in high overall yields. Sterically demanding substrates, including the synthetic steroidal drug gestrinone, as well as more challenging alkynes bearing additional (photo)­labile functionalities or strongly electron-deficient and -rich substituents, were generally well tolerated. Moreover, selected functional group interconversions performed under harsh reductive or oxidative conditions demonstrated the remarkable stability of the bis-heterocyclic core and highlighted its robustness. Thus, the developed protocol can be recommended for the preparation of more advanced analogues. Biological evaluation of selected hybrids is currently underway in our laboratories and will be reported in a separate study.

All commercially available reagents and solvents were used as received. Products were purified by filtration through a short plug of silica gel (FCC) or standard column chromatography (CC) (SiO 2 , 230–400 mesh) by using freshly distilled solvents and recrystallized from appropriate solvents. NMR spectra were taken on a Bruker AVIII instrument ( 1 H at 600 MHz, 13 C at 151 MHz, and 19 F NMR at 565 MHz); chemical shifts are reported relative to the solvent residual peaks [for CDCl 3 : 1 H NMR: δ = 7.26, 13 C NMR: δ = 77.16; for DMSO- d 6 : 1 H NMR: δ = 2.50, 13 C NMR: δ = 39.52] or to CFCl 3 (δ = 0.00) used as an external standard. The IR spectra were taken with an Agilent Cary 630 FTIR spectrometer, in neat. (ESI)-MS was performed with a Varian 500-MS LC ion trap; high-resolution MS (ESI-TOF) measurements were performed with a Waters Synapt G2-Si mass spectrometer. Combustion analyses were obtained with a Vario EL III (Elementar Analysensysteme GmbH) instrument. Optical rotations were determined with a PerkinElmer 241 polarimeter at the temperatures indicated. Melting points were determined in capillaries with a MEL-TEMP apparatus (Laboratory Devices) or with a polarizing optical microscope (POM) (Opta-Tech) and are uncorrected. Single crystals of 1bb were measured on a XtaLAB Synergy, Dualflex, Pilatus 300 K diffractometer using mirror-focused Cu Kα radiation. Crystallographic data have been deposited at the Cambridge Crystallographic Data Center as supplementary publication number CCDC 2425718 . These data can be obtained free of charge from the CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.; fax: + 44 (0) 1223 336 033; email: [email protected] (or via http://www.ccdc.cam.ac.uk/conts/retrieving.html ). A mixture of 5-azidopyrazole 5 (1.00 mmol), acetylene (1.20 mmol), copper­(II) sulfate pentahydrate (38 mg, 0.15 mmol), and sodium l -ascorbate (59.5 mg, 0.30 mmol) in MeOH/H 2 O (10:1, 14 mL) was stirred at 55 °C (oil bath) until the starting pyrazole was fully consumed (typically up to 5 h; TLC monitoring). The solvents were then evaporated, and the crude reaction mixture was dissolved in DCM (20 mL), dried over Na 2 SO 4 , and filtered through a Celite pad, which was washed with additional portions of DCM (2 × 8 mL). After the solvent was removed in vacuo , product 1 was purified by flash column chromatography (FCC) and recrystallized. FCC (SiO 2 , hexane/EtOAc 4:1); colorless solid, 314 mg (85%); mp 120–122 °C (hexane). 1 H NMR (600 MHz, CDCl 3 ) δ: 7.80–7.79 (m, 2H), 7.75 (s, 1H), 7.45–7.43 (m, 2H), 7.39–7.36 (m, 1H), 7.20–7.17 (m, 4H), 7.00 (s, 1H), 2.35 (s, 3H). 13 C­{ 1 H} NMR (151 MHz, CDCl 3 ) δ 148.5, 143.0 (q, 2 J C–F = 39.6 Hz), 140.2, 135.8, 134.5, 130.3, 129.4, 129.2, 129.1, 126.1, 124.3, 121.6, 120.7 (q, 1 J C–F = 269.6 Hz), 103.0 (q, 3 J C–F = 2.2 Hz), 21.3. 19 F NMR (565 MHz, CDCl 3 ) δ −62.76 (s, CF 3 ). IR (neat): ν 3153, 2963, 1580, 1502, 1364, 1238, 1156, 1139, 1014 cm –1 . (+)-ESI-MS ( m / z ): 370.4 (100, [M + H] + ). Anal. Calcd for C 19 H 14 F 3 N 5 (369.4): C, 61.79; H, 3.82; N, 18.96. Found: C, 61.70; H, 3.98; N, 19.07. To a solution of 1-aryl-3-(trifluoromethyl)­pyrazole 3 (1.00 mmol) in anhydrous THF (10 mL), at −78 °C, under argon, was added n -BuLi (2.5 M in hexane, 0.52 mL, 1.30 mmol). After 5 min, a solution of tosyl azide (405 mg, 2.05 mmol) in dry THF (5 mL) was added dropwise. The reaction mixture was allowed to warm to room temperature and stirred for 4 h. The reaction was quenched with 1 M NH 4 Cl­(aq) solution (15 mL) and extracted with DCM (3 × 20 mL). The combined organic layers were washed with water (3 × 10 mL), dried over Na 2 SO 4 , and filtered, and the solvents were removed in vacuo . The crude product 5 was purified by standard column chromatography on silica gel (CC). CC (SiO 2 , hexane/DCM 4:1); red solid, 222 mg (83%); mp 45–47 °C. 1 H NMR (600 MHz, CDCl 3 ) δ: 7.48–7.45 (m, 2H), 7.29–7.27 (m, 2H), 6.45 (s, 1H), 2.41 (s, 3H). 13 C­{ 1 H} NMR (151 MHz, CDCl 3 ) δ 142.9 (q, 2 J C–F = 38.7 Hz), 139.4, 139.0, 134.9, 129.8, 124.0, 120.9 (q, 1 J C–F = 268.7 Hz), 94.1 (q, 3 J C–F = 2.5 Hz), 21.3. 19 F NMR (565 MHz, CDCl 3 ) δ −63.02 (s, CF 3 ). IR (neat): ν 2922, 2136 (N 3 ), 1517, 1472, 1282, 1233, 1162, 1107, 972, 816 cm –1 . (+)-ESI-MS ( m / z ): 268.3 (100, [M + H] + ); Anal. Calcd for C 11 H 8 F 3 N 5 (267.2): C, 49.44; H, 3.02; N, 26.21. Found: C, 49.40; H, 3.02; N, 26.19.

Fluorinated and fluoroalkylated pyrazoles have attracted remarkable attention as privileged scaffolds in the design of pharmaceutical agents, crop protection chemicals, and advanced functional materials. In this context, the 1-aryl-3-trifluoromethylpyrazole core represents a particularly appealing structural motif for the discovery of new bioactive compounds ( Figure ). Introduction of a (hetero)­aryl substituent at C(5) has provided access to a broad spectrum of medicinally relevant molecules, including anti-inflammatory agents (e.g., celecoxib), anticancer candidates (SC-560), antibacterial and antifungal compounds, and veterinary medicines (mavacoxib). Derivatives bearing a furan2-yl unit (e.g., MPA14) exhibit pronounced COX-1 inhibitory activity, whereas analogues incorporating pyridine or thiophene rings have been applied in the treatment of asthma. In addition, more complex architectures featuring an additional azole ring, such as 1,3,4-oxadiazole or pyrazole, located at C(5) have been reported as promising coactivator-associated arginine methyltransferase 1 (CARM1) antagonists, and ATP-binding cassette transporting modulators, respectively. Selected bioactive 1-aryl-3-CF 3 -pyrazoles. On the other hand, installation of an amide functionality at C(5) of the 1-aryl-3-CF 3 -pyrazole framework has furnished a diverse series of compounds with considerable potential in both agrochemical and medicinal chemistry. Accordingly, materials featuring herbicidal, parasiticidal, insecticidal, and nematicidal as well as antiproliferative and CARM1 inhibitory properties have been described. Among these, particular attention should be drawn to two marketed drugs: razaxaban, an orally active inhibitor of coagulation factor Xa, and berotralstat, a human plasma kallikrein inhibitor employed for the prophylactic treatment of hereditary angioedema (HAE). Prompted by the well-documented biological activity of both aforementioned subclasses of 1-aryl-3-trifluoromethylpyrazoles, and taking into account the established significance of the 1,2,3-triazole motif in medicinal chemistry, frequently employed as bioisosteric replacement for amide (peptide) and related groups, we envisaged that the structurally related pyrazole–triazole hybrids of type 1 might represent a valuable platform for the discovery of new bioactive molecules ( Scheme ). A variety of non-fluorinated pyrazole-based compounds bearing either a 1,2,3- or 1,2,4-triazole ring have been extensively investigated in recent years. The reported promising activities encompass anticonvulsant, antimicrobial, antifungal, anticancer, , and tyrosinase inhibitory properties. An elegant approach to cyclooxygenase inhibitors obtained via in situ click chemistry was demonstrated by Wuest, in which the active site of the COX-2 isozyme served as a reaction vessel, enabling the identification of two highly potent and selective products. Because of their high nitrogen content, several pyrazole–triazole hybrids functionalized with amino, azo and/or nitro groups have also been described as thermally stable energetic materials of potential relevance to military, mining, and aerospace applications. In contrast, the corresponding fluoromethylated structural motifs remain only sparsely explored. In 2015, Atmakur et al. reported the synthesis of a library of CF 3 -functionalized pyrazoles linked to a 1,2,3-triazole ring through an acetamide-derived spacer ( Scheme ). The designed products were obtained via a five-step sequence using 4,4,4-trifluoro-1-phenylbutane-1,3-dione as the key fluorinated substrate. Preliminary biological evaluation revealed several promising lead compounds exhibiting an in vitro antimycobacterial activity. Shortly thereafter, the same group disclosed a multistep synthesis of cytotoxic tricyclic diazepine derivatives comprising directly bonded pyrazole and 1,2,3-triazole units, starting with 4,4,4-trifluoro-3-oxobutanoate ( Scheme ). This ester was also employed by the Rajashakar group in the synthesis of fully substituted 5-(1,2,3-triazol-1-yl)-3-CF 3 -pyrazole-4-carbaldehyde derivatives. However, a general protocol for the synthesis of trifluoromethylated pyrazole–triazole hybrids of type 1 , closely resembling the well-established 5-(hetero)­aryl- and 5-aminocarbonyl-functionalized pyrazoles of practical significance, has not been described. Moreover, CF 3 -functionalized hydrazonoyl bromides 2 have emerged as a convenient alternative to the classical 1,3-dicarbonyl substrates typically employed in pyrazole synthesis. Thus, in continuation of our study aimed at development of useful synthetic methodologies for fluorinated azoles, here we report a rapid route to pyrazoles 1 via a three-step sequence comprising the formation of the 1-aryl-3-CF 3 -pyrazole unit, its selective azidation, and a copper-catalyzed azide–alkyne cycloaddition.