Mechanism‐Based Fluorogenic trans‐Cyclooctene–Tetrazine Cycloaddition

Abstract The development of fluorogenic reactions which lead to the formation of fluorescent products from two nonfluorescent starting materials is highly desirable, but challenging. Reported herein is a new concept of fluorescent product formation upon the inverse electron‐demand Diels–Alder reaction of 1,2,4,5‐tetrazines with particular trans‐cyclooctene (TCO) isomers. In sharp contrast to known fluorogenic reagents the presented chemistry leads to the rapid formation of unprecedented fluorescent 1,4‐dihydropyridazines so that the fluorophore is built directly upon the chemical reaction. Attachment of an extra fluorophore moiety is therefore not needed. The photochemical properties of the resulting dyes can be easily tuned by changing the substitution pattern of the starting 1,2,4,5‐tetrazine. We support the claim with NMR measurements and rationalize the data by computational study. Cell‐labeling experiments were performed to demonstrate the potential of the fluorogenic reaction for bioimaging.


General information
All reagents were purchased from Sigma-Aldrich, Alfa Aesar, ABCR, Acros or LC Laboratories (Taxol) and were used without further purification unless otherwise noted. All trans-cyclooctenes were synthesized using RPR-200 Rayonet photochemical reactor from Southern New England Ultraviolet Company equipped with 10 or 16 Hg-quartz iodine lamps 254 nm (2537 Å). To enable continuous flow set up for their synthesis a STEPDOS 03 RC membrane-metering pump from KNF was used. The TLC plates for analysis were from Merck silicagel 60 F 254 and for preparative TLC were purchased from Macherey Nagel (SIL G-200, 2 mm silica layer, 200 × 200 mm, PH = 5, MF = 254, glass back). Other chromatographic purifications were conducted using 40-63 μm silicagel from Acros or on Teledyne Isco CombiFlash Rf200 system. The microwave reactions were performed using CEM microwave reactor. All mixtures of solvents are given in v/v ratio. 1 H and 13 C NMR spectroscopy was performed on Bruker Avance III™ HD 400 MHz and a Bruker Avance III™ HD 500 MHz. All 13 C NMR spectra were proton decoupled. Residual solvent peak was used as reference. Mass spectrometry was performed on AB SCIEX TripleTOF™ 5600, Thermo Fisher Scientific LTQ Orbitrap XL, Q-Tof micro (Waters) or SYNAPT G2 mass spectrometers. Fluorescence measurements were performed on Perkin Elmer LD-45 spectrophotometer equipped with a single cuvette reader. Ultraviolet absorption spectra were collected on Agilent Cary 60 spectrophotometer. HPLC experiments were performed on Shimadzu LCMS 2020 system equipped with single-quadrupole ESI-MS and PDA detector. Peptide click labeling experiments were analysed on Leica M205 fluorescent stereomicroscope equipped with pE-300 white LED light source and DFC3000 G grayscale camera. Cell images were acquired on Leica TCS SP5 confocal microscope equipped with HC PL APO CS2 63.0x1.40 OIL UV objective.

Synthesis of 3C
N N N N OMe Following the general procedure 3, 3-phenyl-6-(2-mesyl)ethyl-s-tetrazine and 4-iodoanisole were used and the suspension was irradiated for 40 min at 50 °C. The product was isolated using Hexane:EtOAc

Synthesis of TPP-Tet probe
The NHS active ester 3R (8 mg, 0.018 mmol) and TPP-amine [8] (9 mg, 0.022 mmol, 1.25 equiv) were dissolved in CH 3 CN and cooled on ice-water bath. The flask was purged with argon and DIPEA (11 µL, 3,5 equiv) was added. The reaction mixture was stirred at room temperature until starting tetrazine disappeared as followed by HPLC-MS and/or TLC (DCM/MeOH = 9/1). The crude reaction mixture was concentrated on rotary evaporator, dissolved in minimum amount of DCM and the product was isolated by preparative TLC using DCM/MeOH = 9/1 as eluent. Isolated: 7 mg, yield 54%. The purity and identity of the product was verified by HPLC-MS (shown below) and HRMS. To a 0 °C solution of 7-β-alanyltaxol [9] (10 mg, 0.011 mmol) in anhydrous acetonitrile (1 mL) was added in one portion under flow of argon the active ester 3T (5.4 mg, 0.012 mmol). The mixture was stirred at room temperature and the progress of the reaction was followed by HPLC-MS. After 18h the solvent was evaporated and the residue was purified by preparative TLC using CH 2 Cl 2 :MeOH (10:1) as eluent to get the product as red solid (6 mg

Synthesis of KYHWYGYTPQNVI model peptide
YHWYGYTPQNVI peptide was synthesized on TentaGel S OH resin (particle size 130 µm; Iris Biotech GmbH, Germany) using standard Fmoc chemistry on a PS3 peptide synthesizer (Protein Technologies, Inc., USA). The first amino acid was manually attached to the resin as a symmetrical anhydride (generated in situ using 10 eq amino acid and 5 eq DIC in DCM at 0 °C), further coupling steps used 3.3 eq amino acid and HBTU in DMF containing 0.2 M N-Methylmorpholine (NMM) for 20 min. After removal of the Fmoc protecting group (20 % piperidine (v/v) in DMF, 3x 5 min) the N-terminal lysine was added manually (6.6 eq amino acid and HBTU, 45 min) and the N-terminus was capped with acetic anhydride (7.5 % Ac 2 O and 7.5 % NMM (v/v) in DMF, 1 h).

Fluorogenic click-labeling of Tetrazine-KYHWYGYTPQNVI peptide on Tentagel resin
100 mg of the resin containing the KYHWYGYTPQNVI peptide were incubated with 500 µL of TFA/TIPS/H 2 O= 95/2.5/2.5 mixture for 1.5 h at room temperature to remove the side-chain protecting groups. After washing the resin with DMF (5x 5mL) and DCM (5x 5mL) part of the deprotected peptide (ca. 50mg) was reacted on the resin with Tetrazine NHS ester 3R (5 mg) in DMF (200 µL) in the presence of DIPEA (5 µL) for 2 hours at room temperature. The progress of the reaction was evident since the beads became red as a result of the attached tetrazine moiety. A small portion of the peptide (ca. 5-10 mg) was cleaved from the resin by using 100 mM NaOH solution (50 µL, 1h at room temperature) to verify the success of the synthesis by ESI-MS ( Figure S3). For the fluorogenic labeling experiment a small amount of the beads (in DMSO/ H 2 O) was transferred on a glass slide and were placed under fluorescent stereomicroscope (equipped with grayscale camera: Leica DFC3000 G). A drop of the corresponding TCO isomer (50 mM solution in DMSO/H 2 O = 1/1) was dropped onto the beads to initiate the reaction. The pictures were acquired at indicated time points ( Figure S2). The peptides were then cleaved from the resin using 100 mM NaOH solution (1h at room temperature) and were analyzed by ESI-MS to confirm the formation of the click products. Please note that also in the case of the equatorial TCO isomer we found by HPLC-MS analysis the mass of the expected dihydropyridazine product rather than the water addition product. Presumably the additional treatment of the peptide (cleavage, neutralization etc.) promoted the final isomerization and led to the formation of the final dihydropyridazine product. Figure S2. The resin beads containing the tetrazine modified peptide were incubated with the axial TCO 1 or equatorial TCO 2 and the pictures were captured with grayscale camera at indicated time points using fluorescent stereomicroscope and UV excitation (350 nm). The colour of the beads was adjusted using LAS AF Lite program. S17 Figure S3. Mass spectra (ESI) of the peptides before and after labeling with TCOs 1 and 2.

Fluorogenic click-labeling of TCO-KYHWYGYTPQNVI peptides on Tentagel resin
150 mg of the model peptide on Tentagel OH resin was washed with DMF (5x 5mL) and DCM (5x 5mL) and dried under vacuum. The side-chain protecting groups were removed by shaking the beads in 1 mL of TFA/TIPS/H 2 O= 95/2.5/2.5 mixture for 1.5 h at room temperature. The beads containing the deprotected peptide were filtered using small SPA column (2 mL from GBiosciences), washed with DMF (5x 1mL), DCM (5x 1mL), EtOH (5x 1mL) and finally dried on SpeedVac. Small amount of the resin (ca. 5-10 mg) was weight into PP vial and the peptide was cleaved from the resin using 50 µL 100 mM NaOH (1h at room temperature). After neutralization with 1M HCl (5 µL, 1 equiv.) the solution was analyzed by HPLC-MS showing that the peptide synthesis was successful ( Figure S8).
The rest of the deprotected peptide on the resin was divided into 4 PP vials and to each vial was added different TCO-N-hydroxysuccinimide active ester shown below (each 5 mg dissolved in 200 µL of DMF) followed by 5 µL of DIPEA. The reaction mixture was agitated at room temperature in the dark for 2 h to provide peptides modified with the corresponding TCO as verified by HPLC-MS after cleavage from the solid support by 100 mM NaOH and neutralization with 1M HCl (see below). Please note that the formed carbamate linker between TCO and the peptide is not completely stable toward the cleavage procedure (100 mM NaOH) and we often observed partial cleavage of the TCO moiety giving in HPLC measurements peak corresponding to starting unmodified peptide. Structures of TCO-NHS esters used to modify the peptide Next, the four peptides containing four different TCOs (two axial and two equatorial isomers) were washed with DMF (5x 1mL), DCM (5x 1mL) and finally suspended in DMSO/H 2 O = 1/1. Tetrazine 3I was added to each vial until the reddish color of the tetrazine persisted (excess of tetrazine). The beads were then washed, placed on microscope slide and inspected in time using Fluorescent stereomicroscope (Leica M205 FA). The pictures were captured by grayscale camera (Leica DFC3000 G) and later processed by LAS AF software.
Finally, small portion (ca. 5-10 mg) of the resin containing the modified peptide was treated with 100 mM NaOH (50 µL) for cleavage. After 1h the solution was neutralized with 1M HCl (5 µL, 1 equiv.). Since the peptides partially precipitated after neutralization, 10 µL acetonitrile was added to dissolve them and the peptides were analyzed by HPLC-MS (shown below) to verify the formation of the click products.
Conditions for HPLC: solvent A: H 2 O + 0.05% HCOOH; solvent B: CH 3 CN + 0.05% HCOOH; gradient: 5% B → 95% B in 9 min, then 2 min 95% B and back to 5% B, Column: Luna® C18 column, 3u, 100A, 100 x 4.6 mm, 1 mL/min flow rate. Figure S4. The resin beads containing the axial TCO modified peptide were incubated with 3I and the pictures were captured with grayscale camera at indicated time points using fluorescent stereomicroscope and UV excitation (350 nm). The color of the beads was adjusted using LAS AF Lite program. Figure S5. The resin beads containing the equatorial TCO modified peptide were incubated with 3I and the pictures were captured with grayscale camera at indicated time points using fluorescent stereomicroscope and UV excitation (350 nm). The color of the beads was adjusted using LAS AF Lite program. Figure S6. The fluorescence intensity (from experiments depicted in Figure S4 and S5) was quantified using the image analysis software CellProfiler. [10] Integrated intensities of single beads identified in each time point are plotted in the graph as a mean ± standard error of measurement.
The results show that the structure of the TCO influences the fluorogenic nature of the reaction. The experiment confirmed the superior properties of the axial isomer in this regard. However, the difference between the two axial and equatorial isomers was not so expressive and the fluorescence also developed much slower in this case (compare Figures S4, S5 with Figure S2).
Moreover, we found that by attaching the TCO moiety to the peptide via a carbamate linker almost completely abolishes the fluorogenic properties of the reaction ( Figure S7). This supports conclusions from our computational studies where we found that the free OH group is involved in the tautomerization of the dihydropyridazine product proceeds via 1, 3-hydrogen shift rather than hydration-dehydration steps. It seems that further derivatization of the free OH group leads to alteration of the mechanism. This may also partially explain why this phenomenon remained undetected since the active esters of the transcycloocten-ols are the most commonly employed TCO derivatives for bioconjugations and in this form are also commercially available. Figure S7. The resin beads containing the equatorial or axial TCO modified peptide were incubated with tetrazine 3I and the pictures were captured with grayscale camera at indicated time points using fluorescent stereomicroscope with UV excitation (350 nm). The color of the beads was adjusted using LAS AF Lite program. Figure S8. Mass spectra (ESI) and HPLC chromatogram of the starting peptide KYHWYGYTPQNVI   Table S1 and S2)

Determination of fluorescence quantum yields
Quantum yields of click products were measured at 25°C in CH 3    Copies of absorption and emission spectra of the click products using TCO 1 (please note that peaks in MS spectra with mass of the product + 41 is an artefact arising from the adduct of CH 3

Kinetic measurements
Kinetic measurements were performed under pseudo first-order conditions using an excess of the corresponding axial TCO isomer 1 or the equatorial isomer 2 respectively. Second order rate constants were determined by following the decay in the concentration of the starting 1,2,4,5-tetrazine over time (see  table S3 for the wavelength used for each tetrazine). The concentration decrease was monitored by UV/VIS spectrometry (performed on Agilent Cary 60). The measurements were performed in a mixture of CH 3 CN/H 2 O (4:1) at room temperature.
Conditions: 25 µM solution of the respective tetrazine in CH 3 CN was added to a 250 µM solution of the TCO in CH 3 CN/H 2 O. The final tetrazine concentration was 12,5 µM and the final concentration of TCO 125 µM in CH 3 CN/H 2 O (4:1). The measurement was immediately started after addition of the tetrazine and brief mixing of the reaction mixture in the UV cuvette. The observed decrease of the tetrazine concentration was plotted against time and the data fitted with single exponential equation (y = y 0 + Ae -k/t ) using OriginPro software to provide the observed rate constants k'. The second order rate constants k were calculated by dividing the observed rate constants with the initial concentration of the TCO. All runs were conducted at least three times and the results are summarized in Table S3. a) This wavelength was used to follow the decay in the concentration of starting tetrazine. We next performed a kinetic scanning experiment to follow the progress of the reaction by changes in the absorption spectra over time using tetrazine 3I as an example ( Figure S17). This experiment was performed under conditions identical to the ones used for the above kinetic experiments. Note that the two TCO isomers lead under identical conditions to different spectra profiles indicating a different mechanism S58 of the reaction by using different TCO isomer. This is in agreement with the data observed during NMR studies of the reaction mechanism. Due to limited solubility of most of the tetrazines in water we were unable to perform the kinetic measurements in the presence of more water. Because it is known that the inverse electron-demand Diels-Alder reaction is accelerated in water we performed further experiments using tetrazine 3I which is sufficiently water soluble ( Figure S18). This experiments were performed using a manual stopped flow device connected to the UV spectrophotometer using 12,5 µM final tetrazine concentration and 125 µM final TCO concentration. As evident from this data the reaction rates increase substantially by increased water content what is in good agreement with the excellent reactivity observed during cell labeling experiments ( Figure S26 and S27).

Stability studies in CH 3 CN/PBS buffer:
The stability of the click products with TCO 1 was studied for (E)-3-phenyl-6-styryl-1,2,4,5-tetrazine [6] as well as for the tetrazines 3C and 3G. The stability studies were performed in CH 3 CN/PBS (1:1) at 37 °C and were monitored by HPLC-MS on a Luna® C18 column (3u, 100A, 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. Conditions: 0.3 mL of a 1.25 mM solution of the tetrazine in CH 3 CN were added to 75 µL of a 10 mM solution of TCO 1 in PBS. The mixture was diluted with PBS to a final volume of 0.6 mL in order to get 0.625 mM final concentration of the corresponding tetrazine using 2 eq of TCO 1. The solutions were shaken at room temperature for 1 h and measured by HPLC-MS to verify the formation of the click products (indicated time 0 min in Figure S19-S21). For stability study, the click products were incubated at 37 °C for 168 h in total. During that time the reaction mixtures were measured several times by HPLC-MS ( Figure S19-S21).

Stability studies in fetal bovine serum:
A 20 µL of (E)-3-phenyl-6-styryl-1,2,4,5-tetrazine solution in DMSO (50 mM) was added to 980 µL of serum (Fetal bovine serum from Biosera, 1x diluted with MiliQ water) containing the axial TCO isomer (20 µL of 100 mM in DMSO, corresponding to 2 equivalents). The reaction mixture in serum was incubated at 37 °C and was analyzed by HPLC-MS over time. A sample of the crude reaction mixture in serum was filtered through 0.2 µm syringe filter and was directly used for HPLC-MS. Another part of the sample was mixed with 200 µL of DCM, rigorously vortexed and centrifuged. The upper (aqueous) part was removed with a pipette and the organic phase concentrated using Speedvac. After re-dilution in 100 µL of CH 3 CN/H 2 O (1/1) and centrifugation 50 µL of the sample were placed into HPLC vial and were analyzed by HPLC-MS ( Figure S22). Comment: after first day of incubation under the above conditions we observed a formation of a new product that was assigned (according to the observed mass) to the corresponding oxidation product (pyridazine labeled as 3 in Figure S22) giving a ratio between the pyridazine 3 and the 1,4dihydropyridazines 1 and 2 = ca. 1 / 1 (according to integration of the peak areas). After the second day the ratio increased to: 3 / (1 + 2) = 1.2 / 1 and after third day further to: 3 / (1 + 2) = 1.7 / 1.

Cell labeling experiments
U2OS cells were maintained in high glucose DMEM (Sigma) supplemented with 10% FBS (Biosera) and 0.1mg/ml of penicillin-streptomycin (Sigma) at 37°C/5% CO 2 . One day before the experiment 0.3x10 6 cells were seeded at the 3.5 cm cultivation dishes with a coverglass in the bottom (SPL Life Sciences).

S63
Tetrazine conjugated materials (TPP-Tet or Taxol-Tet) were dissolved in DMSO (5mM). Cells were incubated with the tetrazine conjugated compounds in complete media for indicated time points at 37°C. Cells were then washed once with media, and incubated for further 10-30 minutes in complete DMEM medium without phenol red containing DRAQ 5, Mitotracker deep red or Tubulin tracker (all from Thermo Scientific). In experiments with TPP-Tet and Taxol-Tet probes the cells were incubated prior to imaging for 10 minutes at 37°C with 25 or 50µM (5 or 10 equivalents, final concentration) of TCO 1 (stock 50 mM in DMSO). For real-time click experiment ( Figure S26 and S27), TCO 1 was added to a dish mounted onto the microscope. Pictures of live cells were taken every minute in total time of 20 minutes. Images of live cells were taken using Leica TCS SP5 confocal microscope equipped with HC PL APO CS2 63.0x1.40 OIL UV objective. Excitation for click products was 405 nm. Emission was collected sequentially using Hyd detector in BrightR mode, with AOBS window set to 412-522 nm or 450-550 nm. DRAQ5 for nuclei staining was excited with 633 nm laser and collected in a window 667-748 nm. Mitotracker: excitation 633 nm, emission window 692-734 nm. Tubulin tracker: excitation 561 nm, emission window 600-650 nm. For details of each experiment see the figure captions. Brightness of the raw images was adjusted using FIJI software. [12] Figure S23. U2OS cell labelling experiments using mitochondria selective TPP-Tet probe and microtubule selective Taxol-Tet probe. Conditions: incubation of cells for 3 hours/37°C with the respective tetrazine probe (5 µM), washing and incubation with DMEM medium containing 500 nM DRAQ5 nuclear dye for 30 min and finally addition of TCO 1 (25 µM final). The images were acquired on confocal microscope using 405 nm excitation for click products (emission window 412-522 nm) and 633 nm laser excitation for DRAQ5 and a 667-748 nm emission window. Note: The signal intensity in case of the Taxol-Tet probe was lower than for the TPP-Tet probe and the laser intensity was therefore increased from 10 to 15%.

NMR experiments and computational study
For the NMR monitoring of TCO reaction with diphenyltetrazine, the studied cyclooctenol (ca 5 mg) was dissolved in CD 3 CN (1 mL) and D 2 O (150 μL) solvent mixture. A suspension of diphenyltetrazine in CD 3 CN was added to the solution under vigorous shaking until the violet color of diphenyltetrazine stopped disappearing. 500 μL of the reaction mixture was transferred to an NMR tube and NMR experiments were acquired periodically and the reaction progress was monitored. A combination of 1D ( 1 H and 13 C) experiments with 2D correlation experiments (H,H-COSY, H,C-HSQC, H,C-HMBC and ROESY) was used to determine the structure and conformation of the reactants, intermediates and final products.
The search for preferred conformations of the studied compounds was performed using one hundred simulated annealings to 1000 K followed by slow cooling to 200 K performed with every molecule. Molecular modeling program package Hyperchem 8 (Hypercube) was used. The conjugate gradient method for energy minimizations was used to convergence (less than 0.01 kJ mol -1 RMS force).
Force field MM+ was used for all computations. General protocol for obtaining lowest-energy conformers by simulated annealing: optimized starting structure was subjected to dynamic run -0.5 ps heating from 300 to 1000 °C, 0.7 ps equilibration and 1 ps cooling to 200 °C followed by energy minimization. Every next run started from the previously minimized structure. A set of 100 structures was so obtained for each compound and the calculated structures were then sorted according to the conformation to several structural types. The lowest-energy conformations and other conformations important for the proposed reactions were subjected to geometry optimization at DFT level, using B3LYP functional, [13] standard 6-31+G(d,p) basis set and polarizable continuum model used for implicit acetonitrile solvation.
[14] The Gaussian09 program package was used throughout this study. [15] The QST3 optimization method [16] was applied in the search for the transition state structures of the reaction, that is, the structures of the reactant, product, and estimated transition state were used as input for the TS search. The vibrational frequencies and free energies were calculated for all of the optimized structures, and the stationary-point character (a minimum or a first-order saddle point) was thus confirmed.

Comment
The reaction of TCOs 1 and 2 with diphenyltetrazine The reaction of TCO 2 with diphenyltetrazine leads to a mixture of intermediates and products. The progress of the reaction was followed by NMR spectroscopy and all the components of the mixture were identified. An overview of the reaction is depicted in Scheme S1, illustrative examples of proton NMR spectra monitoring the reaction progress are shown in Figure S28 and the changes of relative concentrations of the mixture components during the hydration-dehydration reaction are shown in Figure  S29. The configuration at the newly arisen asymmetric center (HN-C-OH) in the intermediates 2B was not determined experimentally, but the conformational analysis predicts that the structures with the hydroxy group in pseudoaxial position (i.e. trans to the nearest bridgehead hydrogen atom) are by 1.5-2 kcal/mol more stable. No signals of OH/NH hydrogens were observed during NMR measurements because the reaction was performed in the presence of D 2 O. In agreement with the proposed mechanism, no attachment of deuterium to a carbon atom was observed. The following compounds were identified in the reaction mixture during the hydration-dehydration reaction following the reaction of compound 2 with diphenyltetrazine:  Figure S31. The lowest-energy conformation of compound 1C found by molecular modeling (hydrogen atoms omitted for clarity). This conformation of the product was confirmed experimentally by NMR spectroscopy. For example, ROESY experiment confirmed the spatial proximity of H-1 and one of the hydrogen atoms at C-5, or large coupling constant between H-4 and one of H-3 confirmed that the torsion angle between these hydrogens is close to 180°.

Conformational analysis of compounds 1A, 1B, 2A and 2B
The conformational analysis of intermediate 1A performed with molecular-mechanics simulated annealing shows that the hydroxyl oxygen is very close (2.1 Å) to hydrogen H-8 in a low-energy conformer and we hypothesize that a direct transfer of hydrogen H-8 to the hydroxyl oxygen may be the crucial low-barrier reaction leading to the formation of product 1C. On the other hand, the conformational analysis of compound 2A did not suggest any close contacts between the bridgehead hydrogens and the TCO hydroxy group. Relative energies of the lowest-energy conformers and conformers with the shortest bridgehead hydrogen -TCO hydroxyl oxygen together with the corresponding O•••H distances are summarized in Table S4. Conformers 1-3 of compounds 1A and 2A were re-optimized by DFT method and the corresponding energies and distances are summarized in Table S5.  The calculated reaction mechanism As discussed above, we expect that the first step in the reaction sequence leading to compounds 1C and 2C is the IEDDA followed by nitrogen elimination, which provides 4,5-dihydropyridazine intermediates 1A and 2A. The transition state structures were found for both these reactions starting from compound 1; the reaction barriers were found to be 21.6 and 6.0 kcal/mol, respectively. Similar reaction barriers can also be expected for compound 2.
Short distance was found between hydrogen H8 and TCO hydroxyl oxygen in a low-energy conformer of compound 1A. We hypothesize that this conformer of 1A can be protonated on a nitrogen atom (leading to even shorter O•••H8 distance - Table S5) and hydrogen H8 can migrate to the TCO hydroxyl oxygen. This proton migration can be facilitated by hydration of the hydroxyl group enabling the transfer of the excess hydrogen from the hydroxyl to the solvent water molecules. The transition-state structure and a modest energy barrier (15 kcal/mol) for the hydrogen atom migration to the oxygen were found by DFT calculations.
No similarly short distance between the bridgehead hydrogen atoms and the TCO hydroxyl was found in any conformation of compound 2A and the calculated reaction barrier of the intramolecular proton transfer in the 2A conformer with the shortest O•••H1 distance (2.49 Å - Table S5) is ca 10 kcal/mol higher than in 1A. Figure S32. The calculated relative free energy profile of the reaction of axial TCO (1) with diphenyltetrazine and the subsequent prototropic rearrangement. Relative free energies are shown in kcal/mol; hydrogen atoms not important for the intramolecular proton transfer were omitted in the depicted transition-state structure.

S74
Cartesian coordinates of ground-state and transition-state structures  S75   Table S7. Atomic numbers, Cartesian coordinates (Å), electronic and free energies (a.u.) of the reactant, transition state and product of the nitrogen elimination reaction after IEDDA of compound 1 with dipheyltetrazine.