Thiacycloalkynes for Copper-Free Click Chemistry

Bioorthogonal chemistry enables the interrogation of biomolecules and physiological processes that are inaccessible by using conventional research tools.1 A common experimental protocol starts by labeling a target biomolecule in cells or live organisms with a bioorthogonal functional group. Then, a probe molecule bearing complementary functionality is added to the system and the ensuing bioorthogonal chemical reaction delivers the probe specifically to the targets of interest. For many applications, rapid reaction kinetics are essential. This is particularly true for labeling experiments in live animals, in which reagent concentrations are limited (i.e., nm to low μm), or in which the process that is probed occurs on a fast time scale. Consequently, methodologists interested in the development of bioorthogonal reactions are increasingly focused on kinetic optimization.2


General Procedures
All chemical reagents were purchased from Sigma-Aldrich, Acros, or TCI chemicals and used without purification unless noted otherwise. Solvents were purified as described by Pangborn et al. [1] In all cases, magnesium sulfate or sodium sulfate were used as drying agents and solvent was removed by reduced pressure with a Buchi Rotovapor R-114 equipped with a Welch self-cleaning dry vacuum. Non-volatile products were further dried by reduced pressure with an Edwards RV5 high vacuum. Thin layer chromatography was performed with EMD 60 Å silica gel plates. Unless otherwise specified, Rf values are reported in the solvent system the reaction was monitored in. Flash chromatography was performed using Silicycle ® 60 Å 230-400 mesh. All 1 H, 13 C, and 19 F NMR spectra are reported in ppm and referenced to solvent peaks. Spectra were obtained on Bruker AV-300, AVB-400, AVQ-400, DRX-500, or AV-500, AV-600 instruments. High resolution electron ionization (EI) and electrospray ionization (ESI) mass spectra were obtained from the UC Berkeley Mass Spectrometry Facility.

S7
Scheme S1. Synthesis of thiaDIFBO NFSI = N-fluorobenzenesulfonimide S1 was sequentially difluorinated using LDA and N-fluorobenzenesulfonimide (NFSI) to yield 19. This intermediate was homologated using AlMe 3 and TMSCHN 2 to yieldsilyl ketone S3. This step was far less efficient than achieved with the all-carbon cycloheptane analog, perhaps due to Lewis acid/base pairing of the sulfur atom and the trimethylaluminum. Due to the labile nature of the -silyl ketone, crude S3 was immediately converted to vinyl triflate S4, which was then quantitatively converted to thiaDIFBO (9) by treatment with CsF.

4-fluoro-3,4-dihydrobenzo[b]thiepin-5(2H)-one (S2).
A flame-dried flask was charged with 3,4-dihydrobenzo[b]thiepin-5(2H)-one S1 (1.2 g, 6.6 mmol, 1 equiv). The flask was evacuated and backfilled with nitrogen twice. THF (33 mL, anhydrous) was added to the flask and the solution cooled to -78 o C. LDA (4.0 mL of 2M solution in heptane/ THF/ ethylbenzene, 8.0 mmol, 1.2 equiv) was added and the solution warmed to 0 o C and allowed to stir for 1 h. Separately, a dry flask was charged with N-fluorobenzenesulfonimide (NFSI) (2.7 g, 8.6 mmol, 1.3 equiv) and evacuated and backfilled with nitrogen twice. THF (30 mL, anhydrous) was added and the solution cooled to -78 o C. The solution of base was slowly added to the NFSI solution over 12 min via syringe and the mixture was allowed to warm to rt over 30 min, at which point it was quenched with saturated ammonium chloride (50 mL). The organic layer was separated and the aqueous layer extracted with ethyl acetate (2 x 100 mL). The organic layers were combined, dried with sodium sulfate, filtered, and concentrated under reduced pressure. The crude oil was purified using silica gel chromatography (93:7 hexanes/EtOAc) to yield 1.1 g (5.6 mmol, 85%) of the desired product as a yellow oil. R f = 0.48 in 9:1 hexane/EtOAc. 1

4,4-difluoro-3,4-dihydrobenzo[b]thiepin-5(2H)-one (19).
A flame-dried flask was charged with 4-fluoro-3,4-dihydrobenzo[b]thiepin-5(2H)-one S2 (800 mg, 4.1 mmol, 1 equiv). The flask was evacuated and backfilled with nitrogen twice. THF (20 mL, anhydrous) was added to the flask and the solution cooled to -78 o C. LDA (2.5 mL of 2 M solution in heptane/ THF/ ethylbenzene, 4.9 mmol, 1.2 equiv) was added and the solution warmed to 0 o C and allowed to stir for 1 h. Separately, a dry flask was charged with NFSI (1.7 g, 5.3 mmol, 1.3 equiv) and evacuated and backfilled with nitrogen twice. THF (20 mL, anhydrous) was added and the solution cooled to -78 o C. The solution of base was slowly added to the NFSI solution over 10 min via syringe and the reaction then allowed to warm to rt. Saturated ammonium chloride (40 mL) was added to the reaction followed by ethyl acetate (50 mL). The organic layer was separated and the aqueous layer extracted with ethyl acetate (2 x 50 mL). The organic layers were combined, dried over sodium sulfate, filtered, and concentrated under reduced pressure. The crude oil was purified using silica gel chromatography (93:7 hexanes/ EtOAc ) to yield 390 mg (1.8 mmol, 45%) of a light yellow oil that solidified upon cooling. R f = 0.55 in 9:1 hexane/EtOAc. 1

S9
The oil was transferred to a dry flask that was evacuated and backfilled with nitrogen twice. Dry THF (10 mL) was added and the reaction cooled to -78 o C. NaHMDS (600 μL of 2M solution in THF, 1.2 mmol, 1.2 equiv) was added and the reaction stirred for 2 h at -78 o C. Trifluoromethane sulfonic anhydride (200 μL, 1.2 mmol, 1.2 equiv) was then added and the reaction stirred for 1 h at -78 o C. Methanol (1 mL) was added and the reaction was then allowed to warm to rt and concentrated. The oily solid was taken up in dichloromethane and filtered. The filtrate was concentrated and purified via HPLC on a 100 Å C 18 column, (70% to 100% acetonitrile in water over 30 minutes). The desired product eluted at 17 minutes. Concentration of the desired fraction yielded 16 mg (0.037 mmol, 3.8%) of (Z)-4,4-difluoro-6-(trimethylsilyl)-3,4-dihydro-2H-benzo[b]thiocin-5-yl trifluoromethanesulfonate S4 as a clear oil. R f = 0.82 in 9:1 hexane/EtOAc. 1
3,3'-thiobis(2,2-dimethylpropanoic acid) S6 (4.0 g, 17 mmol, 1 equiv) was dissolved in ethanol (45 mL, 770 mmol, 45 equiv) and toluene (45 mL). Concentrated sulfuric acid (300 μL) was added and the flask was equipped with a Dean-Stark apparatus and heated at 90 o C overnight. The next day the reaction was cooled to rt and poured into a separatory funnel containing water (100 mL). The organic layer was separated and washed with saturated sodium bicarbonate (1 x 100 mL). The aqueous layers were combined and back extracted with ethyl acetate (1 x 300 mL). The organic layers were combined, dried over sodium sulfate, filtered, and concentrated. The crude oil was purified via silica gel chromatography (7:1 hexanes/EtOAc) to yield 4.0 g (13 mmol, 81%) of a clear oil. R f = 0.65 in 9:1 hexane/EtOAc. 1
A dry flask was charged with (1Z,1'E)-(3,3,6,6-tetramethylthiepane-4,5diylidene)bis(hydrazine) S10 (100 mg, 0.44 mmol, 1 equiv) and then evacuated and backfilled with nitrogen twice. Dichloromethane (1 mL, anhydrous) was then added, and the solution was cooled to 0 o C. Lead tetraacetate (418 mg, 0.940 mmol, 2.15 equiv) was dissolved in dichloromethane (2.14 mL, anhydrous) and added dropwise over 1.5 h at 0 S13 o C. The reaction was quenched with aqueous sodium bicarbonate (1 mL) and extracted with dichloromethane (3 x 3 mL). The organic layers were combined, dried over sodium sulfate, filtered, and concentrated under reduced pressure. The crude product was used for kinetics experiments. Purification for analytical purposes was performed by fractional distillation using a Kugelrohr apparatus. 10 mg (0.059 mmol, 14 %) of the desired product were collected at 50 o C , 8 mmHg as a light yellow oil. R f = 0.75 in 19:1 hexane/diethyl ether. 1    The reaction of TMTH 10 and benzyl azide was monitored by 1 H NMR for 12 min at rt. TMTH and benzyl azide were separately dissolved in CD 3 CN and mixed together in a 1: 1 ratio at concentration of 5 mM (blue, green, pink) or 2.9 mM (purple, orange, red) . The percent conversion was calculated by the disappearance of TMTH and benzyl azide relative to the formation of product as determined by integration. The second-order rate constant was determined by plotting 1/[10] versus time. The plot was fit to a linear regression and the slope corresponds to the second-order rate constant. Shown are data from three replicate experiments at each conentration. The six lines had an average slope of 4.0 ± 0.4 M -1 s -1 . S15 Figure S4. a) 30 s interval timepoints of NMR kinetics measurements of the reaction between 2.9 mM azide and 2.9 mM 10 in CD 3 CN starting at 1.63 min through 5.63 min.
(Measurements were taken until 15 min from mixing of starting materials.) The singlet at 5.75 ppm corresponds to the triazole benzylic protons, the singlet at 4.39 ppm corresponds to the azide benzylic protons, the singlet at 2.81ppm corresponds to the methylene protons on TMTH, and the doublet at 2.78 ppm corresponds to the methylene protons on the product. b) Direct comparisons of 10 (bottom), the reaction in progress at 2.63 min (middle) and a purified sample of triazole product (top). Peak at 1.9 ppm is CD 3 CN and at 2.15 ppm is H 2 O. At 16.13 min, the NMR spectrum shows 84% conversion to product. To obtain pure triazole sample, the kinetics samples were combined after each had been allowed to react for ~16 min. After concentration, the crude mixture was purified by silica gel chromatography (hexanes to 3:1 hexanes/EtOAc) resulting in pure triazole product in 38% isolated yield.

Scheme S3. Synthesis of 11
As with DIFBO and thiaDIFBO, the synthesis of 11 proceeded through a key ring expansion step (Scheme 3). Compound S11 was dimethylated by treatment with KHMDS and methyl iodide to produce 20. Homologation of 20 was performed using AlMe3 and TMSCHN 2 , producing the unexpected silyl enol ether S12, which was readily converted to ketone S13 upon treatment with acid. Compound S13 was then treated with KHMDS and trifluoromethane sulfonic anhydride to form vinyl triflate S14. Attempts to eliminate the triflate using LDA or hexamethyldisilylamide bases gave no reaction, perhaps due to unfavorable steric interactions between these large bases and the gemdimethyl group. However, treatment of S14 with NaH in the presence of benzyl azide gave triazole cycloadducts S15 and S16, suggesting that 11 was formed in situ.

Western Blot Competition Experiments
Western blot analysis was performed on Jurkat cell lysates. Jurkat cells were incubated in the described media containing 25 μM Ac 4 GalNAz or DMSO vehicle for 3 days. They were then washed in Dulbecco's modified PBS and lysed in 1% NP-40, NaCl (150 mM), Tris pH 7.4 with protease inhibitors. The cell lysate was then sonicated and cleared by centrifugation. Protein concentration was normalized using a BCA assay. Figure 2: TMTH (10) was added at varying concentrations for 1.5 hours. Phosphine-FLAG (500 μM final concentration) was added to the protein solutions and the vials agitated overnight. Figure S8: Phosphine-FLAG (500 μM final concentration) was added to the protein solutions at 0 0 C. TMTH (10) was then added at varying concentrations and the vials agitated at room temperature overnight.
The following day, 4X loading buffer was added and the proteins were separated on a 4-12% Bis-Tris gradient SDS-PAGE gel at 140 V (Bio-Rad, Criterion system). They were then electroblotted onto nitrocellulose, blocked in 5% bovine serum albumin (BSA, Sigma) in Tris-buffered saline with Tween (TBST,10 mM Tris pH 8.0, 150 mM NaCl, 0.1% Tween-20), and treated with anti-FLAG (M2, Sigma, 1:1000 dilution in 5% BSA in TBST from stock) overnight at 4 o C. The blots were washed with TBST three times for ten min then treated with anti-mouse light chain-HRP-conjugated secondary antibody (Southern Biotech, 1:5000 dilution in 5% BSA in TBST). The blots were again washed with TBST three times for ten min and analyzed by standard enhanced chemiluminescence immunoblotting methods (Pierce). The blots were stored at 4 o C. India ink staining was performed by washing the blots copiously with deionized water 3x for ten min then TBST for ten min. The blot was then incubated in 1:1000 dilution of India ink in TBST for 1 h then washed for 30 s with water and ten min with TBS.  by Western blot using an anti-FLAG antibody then a secondary antibody conjugated to horse radish peroxidase.

Figure S9
μM 10 India ink Figure S9. India ink staining of the western blot shown in Figure S8.

Barstar Expression and Purification
A plasmid (pQE30-Barstar) containing the Bacillus amyloliquefaciens protein Barstar as a 6xHis fusion and with two point mutations for improved stability (Cys53Ala and Cys95Ala) in a pQE30 expression vector was obtained from D. Tirrell (California Insititue of Technology). For incorporation of the unnatural amino acid azidohomoalanine (AHA), the E. coli methionine auxotrophic strain M15-MA was also obtained from D. Tirrell.
To generate both the wild-type Barstar protein (Barstar-MET) and the AHA containing Barstar protein (Barstar-AHA) we followed an expression protocol similar to that reported by Beatty et. al. [5] Briefly, the pQE30-Barstar plasmid was transformed into M15-MA cells and individual transformants were used to inoculate 5 mL of M9 minimal media supplemented with 0.04 mg/mL of each of the 20 amino acids, 1mM MgSO 4 , 5 g/mL thiamine, 0.4% glucose, 200 µg/mL ampicillin and 35 µg/mL kanamycin (i.e., M9 complete media with 20 amino acids). After an overnight incubation at 37 °C with shaking, 1 mL was transferred to 50 mL of M9 complete media with 20 amino acids. After reaching an OD 600 of 1.0, the cells were pelleted (6100g for 10 min at 4 °C) and washed three times with a sterile solution of ice-cold 0.9% (w/v) NaCl. The culture was resuspended in M9 complete medium with 19 amino acids (no methionine) and divided in half; one sample was supplemented with 1mM methionine (Barstar-MET) while the other was supplemented with 1mM AHA (Barstar-AHA). After 15 min at 37 °C with S24 shaking, protein expression was induced with 1 mM IPTG. Cultures were clarified by centrifugation 3 h post induction and the Barstar proteins were purified under denaturing conditions using Ni-NTA spin columns according to the manufacturer's specifications (Qiagen

Conjugation Reactions
10 μg of Barstar-MET or Barstar-AHA were acetone precipitated and dissolved in 2.5 μL of PBS. 2.5 μL of 1 mM TMTH in 1% DMSO in PBS were then added to the samples and the samples allowed to incubate at room temperature for 5 d. The proteins were then acetone precipitated and submitted to LC-MS analysis.

LC-MS Analysis
Samples were subjected to RP chromatography with an Agilent 1200 LC system that was connected in-line with an LTQ Orbitrap XL hybrid mass spectrometer. External mass calibration was performed prior to analysis. A binary solvent system consisting of buffer A (0.1% formic acid in water (v/v)) and buffer B (0.1% formic acid in acetonitrile (v/v)) was employed.
For the intact Barstar samples, the mass spectrometer was outfitted with an Ion Max electrospray ionization source. The LC was equipped with a Poroshell 300SB-C8 column and a 100 mL sample loop. For each run, after 100 to 200 picomoles of protein was injected onto the column, analyte trapping was performed for 5 min with 0.5% B, followed by a linear gradient from 30% to 95% B over 19.5 min, and finished with a washing step in 95% B for 5 min. The flow rate was maintained at 90 mL min -1 . Mass spectra were recorded in positive ion mode over the m/z scan range of 500 to 2000 using the Orbitrap mass analyzer in full-scan, profile mode. Raw mass spectra were processed using Xcalibur (version 4.1, Thermo) and measured charge state distributions were deconvoluted using ProMass (version 2.5 SR-1, Novatia), using the default "small protein" parameters and a background subtraction factor of 1.5.

Cell culture procedures
Jurkat (human T-cell lymphoma) cells were grown in RPMI-1640 media containing 10% fetal bovine serum, streptomycin (0.1 mg/ mL), and penicillin (100 units/ mL). Cells were grown in the presence of 5% CO 2 and maintained at densities between 1 x 10 5 and 2 x 10 6 cells/ mL.  ThiaOCT can label azides on cell surfaces. A) Schematic for flow cytometry analysis of live cell labeling with thiaOCT. B) Quantificaton of cell-surface labeling with thiaOCTbiotin (S18). Cells are grown in the presence (red bars) or absence (blue bars) of 25 μM Ac 4 ManNAz for 3 days. The cells were then treated with no reagent, 250 μM thiaOCTbiotin (S18), or 250 μM DIMAC-biotin [7] followed by FITC-avidin. Error bars represent the standard deviation of three replicate experiments. Representative forward-scatter (x-axis) and side-scatter (y-axis) plots for the experiment described in Figure S10 . Jurkat cells were treated with (+ ManNAz) or without ( -ManNAz) 25 μM Ac 4 ManNAz for 3 days and then treated with 250 μM thiaOCT-biotin, DIMAC-biotin, [6] or vehicle followed by FITC-avidin.

Figure S13
Figure S13. Cytotoxicity analysis of thiaOCT-biotin S18. Jurkat cells were treated with (ManNAz, red bars) or without (NoAz, blue bars) 25 μM Ac 4 ManNAz for 3 days and then treated with 250 μM thiaOCT-biotin or vehicle followed by FITC-avidin. Prior to flow cytometry analysis, the cells were treated with 7-amino-actinomycin D (7-AAD) following the provided procedure. [8] The samples were diluted and analyzed by flow cytometry. The error bars represent standard deviations from three replicates.