The discovery of pyridinium 1,2,4-triazines with enhanced performance in bioconjugation reactions

A novel class of pyridinium 1,2,4-triazines with excellent properties for use in bioconjugation reactions was discovered from a systematic kinetic study.


Synthesis of compound 14
To an ice-cold solution of 10 (as PF 6 salt, 15 mg, 0.032 mmol) in dry DMF (1 mL) was added solid disuccinimidyl carbonate (DSC, 16.4 mg, 2 eq.) under argon followed by Et 3 N (18 µL, 4 eq.). The reaction mixture was stirred at room temperature until starting material disappeared (3 h, TLC in DCM/MeOH (9:1) and/or HPLC-MS). Our attempts to isolate the active ester by column chromatography led to partial decomposition of the product back to the starting material. However, the crude reaction mixture could be used directly in the next reaction step. The yield was ca. 80% (based on integrated peak area, Figure S3).

Determination of the second-order rate constants
Second order rate constants of the reactions between triazines and trans-cyclooctenes (TCO) were determined by following the decrease in the concentration of the starting 1,2,4-triazine over time. The concentration decrease was monitored either by HPLC (for slower derivatives) or by UV/VIS spectroscopy (for faster derivatives). The measurements were performed in a mixture of CH 3 CN/H 2 O (1:1) at room temperature under pseudo first-order conditions using an excess of the corresponding TCO. All runs were conducted at least three times. concentration of all triazines was 2 mM using 10 eq. of TCO (1 mM of triazine using 9 eq. of TCO for 6 + TCO-ol and 0.5 mM of triazine using 10 eq. of TCO for 4 + d-TCO). The measurements were performed on a Luna® C18 column (3u, 100A, 100 x 4.6 mm) using a linear gradient of CH 3 CN + 0.05% HCOOH in H 2 O + 0.05% HCOOH (5→95% in 9 min) at a flow rate of 1.0 mL/min. The MS device was disconnected from the HPLC during measurements. The integral of the absorption of the triazine at 254 nm was measured over 105 or 360 min in 15 or 45 min intervals. By using a calibration curve, the measured integrals were converted into the corresponding concentrations of the triazine, which were plotted against time to provide the observed rate constant from the slope of this plot (fitted with single exponential function: y = y 0 + Ae -k/t  1) to a give a final volume of 3 mL and immediately measured on the UV/VIS spectrophotometer. The final concentration of all triazines was either 45 µM or 50 µM using either 5 eq. or 10 eq. of TCO. The decrease in the absorption of the triazine was followed over 5-120 min in intervals of 0.25, 0.5, 1, 2 or 5 min. The time-dependent measurements were performed at the corresponding absorption maxima of the triazine used, which was determined by UV/VIS spectroscopy before the measurement. The measured intensity of the absorption was plotted against time. Fitting the curves with single exponential equation (y = y 0 + Ae -k/t ) provided the observed rate constants. The second order rate constants were calculated by dividing the observed rate constants with initial concentration of the TCO. All data were processed using Origin or Excel software and are summarized in table S2.

Calculated and found low-resolution masses of the click-products from rate studies.
The following experiments were performed to verify the formation of the corresponding click products during kinetic studies. 30 µL of a 10 mM solution of triazine in CH 3  , 100 x 4.6 mm) using a linear gradient of CH 3 CN + 0.05% HCOOH (5→95% in 9 min) in H 2 O + 0.05% HCOOH at a flow rate of 1.0 mL/min. During the kinetic measurement we have observed in some cases also the formation of the corresponding oxidized pyridine products together with the dihydropyridine product. These particular cases are indicated in Table S1. a) for these compounds the mass of the oxidized form of the click-product was found as well, b) for this compound only the mass of the oxidized form of click-product was found.

Stability studies of compound 6
The stability studies for triazine 6 were performed in CH 3    The reaction mixture was incubated at room temperature for 96 h in total. During that time the sample was measured several times by 1 H-NMR ( Figure S7).  Figure S8).

Synthesis and characterization of the click product 17
Scheme S4. Reaction scheme for the preparation of 17.
Solution of d-TCO (5.10 mg, 0.0277 mmol) dissolved in CD 3 CN/D 2 O (1:1) (500 µL) was added to Triazine 6 (7.5 mg, 0.0185 mmol) dissolved in CD 3 CN/D 2 O (1:1) (500 µL). The mixture was stirred under argon at room temperature for 1 h. Formation of the click product 17 was verified by HPLC-MS and the crude reaction mixture was directly used for characterization by NMR ( Figure S11 and S12). 17 is formed as a mixture of diastereomers (4 in total). Two major diastereomers were assigned based on ROESY experiment. The two minor diastereomers are formed from the anti d-TCO isomer (starting d-TCO was used as an inseparable mixture of syn/anti = 12:1).

Formation of the oxidized product 17ox during preparative TLC
Our attempt to purify the click product 17 by preparative TLC (eluting with CH 3 CN/H 2 O (4:1) and washing the silica with CH 3 CN/H 2 O (4:1) + 0.5% HCOOH led to isolation of the corresponding oxidation click product 17ox. The identity of the oxidized product was confirmed by HPLC-MS ( Figure S13) and NMR.

Stability of the click product 17
The stability studies of the click product 17 were performed in CH 3 Figure S13). For stability study, the clickproduct 17 was stirred at room temperature for 97 h in total under exposure to air. During that time the reaction mixture was measured several times by HPLC-MS ( Figure S14).  Figure  S15). This solution was further diluted in different solvents and used as such for absorbance measurements the quantum yield determination measurements (Figure S16-S18).  Quantum yield of the click product 17 in different solvents was measured at 22°C in 1 cm quartz cuvette using 5µM final concentration of 17. The experiment was performed on FluoroMax 4 spectrofluorometer (Jobin Yvon, Horiba) equipped with a 450 W xenon lamp using Rhodamine 6G (solution in EtOH) as reference (ɸ QS = 0.94). The settings were as follows: Excitation wavelength 405 nm, slit 5.0 nm; Emission 500 -750 nm, increment 1.0 nm, slit 5.0 nm and data algebra formula S1c/R1c. The fluorescence quantum yields were calculated using the following equation: Where: ɸ ref is 0,94 (Quantum yield of Rhodamine 6G in EtOH) [13] F are the integrated intensities (areas) of standard and the sample fluorescence spectra (integrals calculated using OriginPro software) abs is the absorbance of standard and sample at the excitation wavelength (405 nm)

Cell labeling experiments
U2OS cells were maintained in high glucose DMEM (Sigma) supplemented with 10% FBS (Biosera) and antibiotics at 37 °C/5% CO 2 . One day before experiment cells were seeded at density 0.3x10 6 at the glass bottom dishes (SPL Life Sciences 3.5 cm diameter). Triazine 6 was dissolved in DMSO/H 2 O = 1/1 (5 mM) and further diluted before experiments. d-TCO stock solution was in DMSO (50 mM). Images were acquired on Zeiss LSM 780 or Leica TCS SP5 confocal scanning microscopes and raw pictures were processed by FIJI software. [14] Excitation: 405nm, emission window: 525-648 nm for click products). The nucleus was stained with DRAQ5 (excitation: 633 nm, emission window: 667-748 nm).

Mitochondria labeling in live U2OS cells
U2OS Cells were incubated with 5 µM (final) concentration of 6 in complete media for 3 h/37 °C. Cells were washed once with the media, and incubated for further 30 min in complete DMEM medium without phenol red containing 500 nM DRAQ5 dye (Thermo Fisher). The cells were then incubated with d-TCO (25 µM final concentration) for 15 min at 37 °C prior to imaging ( Figure S19).

Preparation of concanavalin A d-TCO conjugate (ConA-dTCO)
2.5 mg of Concanavalin A (Sigma #C2010) were dissolved in 1 ml of 150 mM NaCl, 50 mM HEPES pH 8.3 (to obtain a 2.5 mg/ml solution). 250 µL (0.625 mg) of this solution were combined with 3.6 µL of 100 mM d-TCO NHS active ester (10x molar excess dissolved in dry DMSO). The reaction mixture was incubated at room temperature for one hour with constant shaking. After one hour precipitated material was spun at 25000 rpm for 10 min. Clean supernatant was split and 2x 130 µL was loaded onto two Zeba desalting columns preconditioned with 150 mM NaCl, 50 mM TRIS pH 7.4 (in total 2 columns were used). Note: The d-TCO active ester partially precipitated when added to ConA but after one hour the solution became clear again.

ConA-dTCO Experiment 1 on live cells
Live U2OS cells were incubated for 10 min with ConA-dTCO (100x diluted) in Dulbeccos PBS (DPBS, with Ca 2+ and Mg 2+ ). After washing with DPBS cells were incubated with 50 µM 6 in DPBS with 500 nM DRAQ5 for 15 min. Pictures of living cells were taken at S34 405 nm excitation. Emission was collected at 525-648 nm. DRAQ channel: Exc. 633 nm, Em. 667-748 nm ( Figure S21). Figure S21. U2OS live cell labeling experiment using ConA-dTCO conjugate and triazine 6. Only weak signal has been observed when the experiment was performed on live cells.