New α-Hydroxy-1,2,3-triazoles and 9H-Fluorenes-1,2,3-triazoles: Synthesis and Evaluation as Glycine Transporter 1 Inhibitors

Two series of new compounds containing 1,2,3-triazole moiety were designed as putative GlyT1 inhibitors aiming the discovery of new hits with activity in cognitive disorders. 1,4-Disubstituted α-hydroxy-1,2,3-triazoles were obtained as racemates in moderate to good yields by the copper-catalyzed azide-alkyne cycloaddition (CuAAC) reaction (click chemistry) as the key step between propargyl alcohols and aryl azides, previously prepared from anilines or boronic acids. Benzo[c]chromene-triazoles were planned to be obtained by palladium-catalyzed C−H activation using [bis(trifluoroacetoxy)iodobenzene] (PhI(TFA)2) of some α-hydroxy-1,2,3-triazoles, since benzo[c]chromenes are also privileged groups with several biological activities, including to the central nervous system. Unexpectedly, 9H-fluorenes-1,2,3-triazoles, instead of benzo[c]chromenetriazoles, were obtained by Friedel-Crafts alkylation reaction. The two series of compounds were tested for inhibition of the glycine transporter (rat GlyT1 isoform) but only the α-hydroxy1,2,3-triazole 9b was active (half maximal inhibitory concentration (IC50) = 8.0 μM).


Introduction
Several compounds containing the 1,2,3-triazole moiety have pharmacological properties potentially useful for antifungal and antibacterial, 1 anticancer 2 and antiviral 3 therapy and, in some cases, for the treatment of mental disorders including depression, anxiety, schizophrenia and epilepsy. 4,5 Different reports indicate that triazoles have affinity not only for dopaminergic, adrenergic and serotonergic receptors 4,6 but also inhibit the glycine transporter GlyT1, considered as an important target for the treatment of schizophrenia, cognitive disorders, alcohol dependence and pain. [7][8][9] Figure 1 shows some examples of GlyT1 inhibitors such as ALX 5407 (1), a potent, selective and irreversible compound 10 without clinical use but largely employed as a pharmacological tool.
Bitopertin (2) (Hoffmann-La Roche) was the first of the non-sarcosine derivative GlyT1 inhibitors to be tested in schizophrenia preclinical models and phase II and III clinical trials. 11 This compound is a potent and selective non-competitive inhibitor of the GlyT1 subtype and does not act on other molecular targets. 12 Among the compounds containing a triazole moiety, methyl-1,2,3-triazole 3 stands out since it is a non-competitive inhibitor with very high potency and good metabolic stability in vitro. 9 It is important to notice that among the compounds evaluated by these authors, 9 the 1,2,3-triazoles were more potent than inhibitors. Compound 4 exhibited potent in vivo activity in the cerebral spinal fluid (CSF) glycine model, with excellent drug-like physical properties due to its lipophilicity, whereas compound 5 showed good selectivity and high affinity for the GlyT1 (K i (inhibitor constant) = 1.8 nM). 13,14 New classes of the 4H-1,2,4-triazole derivatives were reported selective GlyT1 inhibitors. 8 Compound 6 was moderately potent (half maximal inhibitory concentration (IC 50 ) = 64 nM) , had a good pharmacokinetics (PK) profile and was the most potent in the induced hyperlocomotion assay. 8 More recently, it was reported 15 another kind of structure based on 1,2,3-triazoles exemplified by compound 7, with excellent inhibitory activity (IC 50 = 1.06 nM). The most common method for the synthesis of 1,2,3-triazole is the copper(I) catalyzed azide-alkyne [3 + 2] cycloaddition reaction (CuAAC), a kind of "click" reaction. 16,17 In a previous work, 18 we described the use of 1,2,3-triazoles as linker to combine different groups in a bifunctional molecules active for cancer human glioblastoma cells (GBM), including highly drugresistant human cell lines GBM-02 and GBM-95. In this work, we have planned new α-hydroxy-1,2,3-triazoles and benzo[c]chromene-triazoles, two other kinds of bifunctional molecules linked by the 1,2,3-triazole moiety. Unexpectedly, we obtained 9H-fluorene linked to the 1,2,3-triazole moiety instead of benzo[c]chromenetriazoles. As 9H-fluorenes are interesting class of polycyclic aromatic hydrocarbons that can be found in many naturally compounds and pharmaceutically active substances as well as organic materials used in the development of photoelectric devices, 19,20 we decided to evaluate them together with the α-hydroxy-1,2,3-triazoles as nonsarcosine based GlyT1 inhibitors ( Figure 2).

Results and Discussion
Synthesis of α-hidroxy-1,2,3 triazoles The propargyl alcohols 13a-13b were obtained in two steps with good yields by cross-coupling reaction between 11a-11b and phenylboronic acid followed by addition of ethynyl trimethylsilane to the aldehydes 12a-12b (Scheme 1). On the other hand, the propargyl alcohols 13c-13g were obtained in one step by addition of ethynyl trimethylsilane to the aldehydes 12c-12g in yields ranging from 67 to 89% (Scheme 1). The lowest yields observed for compounds 13a and 13b were possibly due to the steric hindrance by the phenyl group. The characterization of the compounds was performed by 1 H nuclear magnetic resonance (NMR) and Fourier transform infrared spectroscopy (FTIR). The 1 H NMR spectrum exhibited a singlet signal at d 5.41-5.97 ppm relative to the benzyl alcohol (CH) and another sharp singlet around d 0.2 ppm to the ethynyl trimethylsilane.
The new 1,4-disubstituted α-hydroxy-1,2,3-triazoles 9a-9r were obtained in good yields by the copper-catalyzed azide-alkyne cycloaddition (CuAAC) reaction (Scheme 2). Although the aryl azides can be prepared by different synthetic approaches, the diazotization reaction of aromatic amines is one of the most commonly used methods. Thus, the aryl-azides 15a-15e were prepared from commercial anilines 14a-14d by diazotization reaction with NaNO 2 followed by substitution with sodium azide (NaN 3 ) in yields ranging from 60 to 94%. In order to avoid a reactional step, aryl azide substituted by OCH 3 or F in the position 4 or unsubstituted aryl azide were generated in situ by copper(II)-catalyzed conversion of aryl boron acids, also known as Cham-Lam coupling. 21 In these cases, aryl boronic acids 16a-16c were used in presence of NaN 3 in methanol at 55 °C for 1-2 h. The main advantage of this method is to avoid the isolation of aryl azides, which are sometimes unstable and photosensitive. 22 Thus α-hydroxy- 1,2,3-triazole derivatives 9a-9r were prepared in one step by reaction of propargyl alcohols 13a-13g in the presence of sodium methoxide (NaOMe) as deprotection agent of the trimethylsilyl group (TMS) and sodium ascorbate as reduction agent. 23 The presence of sodium ascorbate favors the in situ formation of Cu I , avoiding the formation of agglomerates, and also avoids atmospheric oxygen interference, that may hinder the reaction process when directly using sources of copper(I). 24 Compounds 9a-9r were characterized by 1 H and 13 C NMR. The 1 H NMR spectrum exhibited a sharp singlet signal in the region d 7.7 ppm confirming the presence of a triazole proton. Although compounds 9f, 9h and 9q were obtained in low yields, there is not any relationship of these yields with the pattern of substitution of aryl azides or alkynes. As described in the literature, 24 the reaction CuAAC is not significantly affected by the steric and electronic properties of the groups attached to the azide and alkyne centers.
Aiming the synthesis of benzo[c]chromene-triazoles, the α-hydroxy-1,2,3-triazoles (9p-9r) were submitted to cyclization conditions using the Pd-catalyzed β-carbon cleavage by the coordination of hydroxy group for obtaining C−O bond. 25 However, instead of benzochromenes-triazoles 14a-14c, new 9H-fluorenes-1,2,3-triazoles 10a-10c were obtained with yields ranging from 25 to 40%, by Friedel-Crafts alkylation (Scheme 3). Similar results have been reported in the literature 26,27 using Lewis acid. The characterization of the 9H-fluorenes-1,2,3-triazoles was performed by IR, 1 H and 13 C NMR. In the IR spectrum, the absence of the broadband in the 3300 cm −1 region referring to the stretching of the O−H bond of alcohols occurs. The 13 C NMR spectrum exhibited signals in the region d 45 and 118 ppm, relative to the presence of benzyl and triazole (CH).

Biological evaluation
The capacity of each compound to inhibit the glycine transporter GlyT1 was evaluated and a maximal concentration of ALX 5407 (0.5 µM) was used as a positive control. At the high concentration tested (10 µM), only the α-hydroxy-1,2,3-triazole 9b presented a statistically significant effect (Table 1).
In order to determine the potency of 9b, we performed a full concentration-response curve ( Figure 3) and measured the mean inhibitory concentration (IC 50 ), i.e., the concentration inhibiting by 50% the specific transport  Table 1). As only compound 9b was active, we were not able to establish some structure-activity relationship. However, as 9b had some activity contrarily to 9a, we can infer that the bromine plays an important role for the pharmacological activity.

General information
For the structural elucidation of the synthesized compounds, 1 H and 13 C NMR spectra were recorded at ambient temperature on a Bruker Advance III spectrometer (operating at 400 MHz for 1 H NMR and 101 MHz for 13 C NMR, with CDCl 3 or dimethyl sulfoxide (DMSO-d 6 ) as solvent). The chemical shifts (d) were given in parts per million (ppm) from internal tetramethylsilane on the d scale, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet). All coupling constants (J values) were given in Hz. Melting points (mp) were determined with an electrothermal, analog model. Infrared spectra were performed using a Varian-3100 spectrometer with ATR (attenuated total reflection). High resolution mass spectra (HRMS) were obtained using a Bruker, MicrOTOF II instrument. Reactions were monitored by thin layer chromatography (TLC) using TLC silica gel 60 F254 (Merck, São Paulo, Brazil). Silica gel column chromatography was performed over Merck silica gel 60 Å (particle size: 0.040-0.063 mm, 230-400 mesh ASTM). All reagents used were commercially obtained (Merck, São Paulo, Brazil and Oakwood Chemical, San Diego, CA, USA) and, where necessary, purified prior to use, such as tetrahydrofuran (THF), that was dried by metallic sodium.
The solvent was evaporated under reduced pressure and the residue was purified by flash chromatography using ethyl acetate/hexane (5:95) as eluent.

Synthesis of propargyl alcohols 13a-13g
The products 13a-13g were prepared following the procedure of Wienhold et al. 28 n-BuLi (2.5 M in hexane, 2.8 mL, 7.0 mmol) was added to a stirred solution of (trimethylsilyl)acetylene (0.99 mL, 7.02 mmol) in anhydrous THF (7 mL) at −5 °C under an N 2 atmosphere. After 15 min, benzaldehyde (5.4 mmol) was added and the reaction mixture was stirred for 3 h at 0 °C. The reaction was quenched with saturated aqueous NaHCO 3 (10 mL) and the layers were separated. The aqueous phase was extracted with Et 2 O (2 × 10 mL) and washed with brine (5 mL). The combined organic layers were dried over Na 2 CO 3 , filtered, and evaporated to dryness under reduced pressure. When necessary the products were purified by flash chromatography using ethyl acetate/hexane (10:90) as eluent.

Preparation of aryl azides 15a-15e
The aryl azides 15a-15e were prepared following the procedure of Wilkening et al. 29 The aniline derivative (7.5 mmol) was dissolved in 5 mL of water and concentrated sulfuric acid (98%, 1.5 mL) and additional water (1.5 mL) was added. The suspension was cooled to 0 °C and a solution of NaNO 2 (0.5 g, 7.6 mmol) in water (1.5 mL) was slowly added under constant stirring. After 15 min, NaN 3 (0.6 g, 9.3 mmol) was added and the mixture was stirred for additional 0.5-1 h. The reaction mixture was extracted with ethyl acetate (3 × 20 mL) and the combined organic fractions were washed with water (50 mL). The organic layer was dried over Na 2 SO 4 and concentrated under reduced pressure. The desired azides 15a-15e were obtained without further purification.   Synthesis of α-hydroxy-1,2,3-triazoles 9a-9r The triazoles were prepared following the procedure of Oikawa et al. 23 Procedure A In a 25 mL round-bottomed flask, aryl azide was prepared using arylboronic acid (0.5 mmol), NaN 3 (0.75 mmol), Cu(OAc) 2 (0.05 mmol) and MeOH (3 mL). Then, the mixture was stirred at 55 °C for 1.5 h under aerobic condition. After cooling to room temperature, sodium ascorbate (0.1 mmol), NaOMe (1 M, 0.5 mmol) and alkyne (0.5 mmol) were added to the resultant mixture, which was stirred at room temperature for 24 h under N 2 atmosphere. At the end of the reaction, the solution was bubbled with air at ambient temperature for 1 h to oxidize the residual organoboron compound. The mixture was extracted with EtOAc (3 × 10 mL), washed with brine (20 mL), dried (Na 2 SO 4 ), and filtered. The solvent was evaporated under reduced pressure and the residue was purified by flash chromatography using ethyl acetate/ hexane (30:70) as eluent. The solvents used to perform TLC plate analysis were ethyl acetate/hexane (40:60).

Biological evaluation
Plasmid corresponding to rat GlyT1b in the pcDNA3 vector was transiently transfected into HEK293 cells using lipofectamine (Invitrogen), as described before. 30 Twentyfour hours after transfection, cells were replated in polylysinecovered 24-well plates. Subconfluent transfected HEK293 cells growing overnight in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum and non-essential amino acids at 37 °C and 5% CO 2 were washed with (4-(2-hydroxyethyl)-1 piperazineethanesulfonic acid (HEPES)-buffered saline (HBS; 10 mM HEPES-Tris buffer containing 150 mM NaCl, 1 mM CaCl 2 , 2.5 mM KCl, 2.5 mM MgSO 4 , and 10 mM (+) D-glucose at pH 7.4). Experiments were performed as previously described, 30 with some modifications. Cells were incubated for 5 min at 22 °C with the test compounds before addition of [ 3 H]-glycine. After 5 min incubation with 10 µM [ 3 H]-glycine (0.01 µM stock solution, 49.9 Ci mmol −1 , PerkinElmer, Boston, USA, diluted with 10 µM unlabeled glycine), the cells were washed twice with HBS and dissolved in 0.2 mL of 0.2 M NaOH. Aliquots of each well were taken for scintillation counting (Packard TriCarb 1600 TR liquid scintillation analyzer, PerkinElmer) and protein concentration determination (BCA TM Protein Assay Kit, Pierce). Results are means ± standard error (S.E.) of two independent experiments performed in duplicate and are expressed as percentage of control uptake. Note that the glycine accumulation in non-transfected HEK293 cells was subtracted from that of transfected cells in order to measure specifically the glycine uptake due to GlyT1. Statistically significant differences were determined by one-way analysis of variance (ANOVA) followed by Dunett's multiple comparisons test (vs. control). To estimate the concentration causing 50% inhibition of glycine uptake (IC 50 ), compound 9b was tested at different concentrations and the concentration-response curve was analyzed by non-linear regression (One site -Fit logIC50) using GraphPad Prism 5.0. 31 This experiment was performed in duplicate and replicated twice (n = 3).

Supplementary Information
Supplementary data ( 1 H NMR and 13 C NMR spectra) are available free of charge at http://jbcs.sbq.org.br as PDF file.