A double-triggered self-immolative spacer for increased selectivity of molecular release

Abstract

A self-immolative spacer based on dissymmetrical N,N′-bis-carbamate aniline is introduced to liberate a substrate from a precursor after dual activation. The proof of principle of its exclusive selectivity for substrate liberation has been conducted on a profluorophore.

---

Introduction

Prodrug strategies^1,2 are increasingly used in therapy to overcome toxicity issues. In antitumoral chemotherapy, they are considered to circumvent lethal side effects, provided that cytotoxic agents (e.g. auristatins) can be reliably released at the tumor site after activation involving overexpressed enzymes or metabolites.^3,4 The search for selectivity in discriminating cancerous and healthy cells is key. Since the former differs from the latter by a plurality of factors, it has been considered to govern the release of the active compound not by one but by several triggering events as it is performed in OR logic gate strategies.^5,6 In this report, we aim at using two triggering events to liberate a covalently bound substrate from a precursor of low molecular weight.

In relation to the preceding goal, profluorophores requiring dual (e.g. enzymatic and chemical) activation followed by bond breakage to become fluorescent have been interestingly reported.^7–10 However, the reported strategies have relied on directly introducing two orthogonal protecting groups on the fluorophore backbone. This constraint has restricted their scope to specific multifunctional substrates and few prodrug examples have been reported at present with these strategies. To overcome this limitation, an AND logic gate strategy recently developed uses a self-immolative spacer that triggers the drug release after two enzymatic reaction steps.^11–13 However, this sequential strategy intrinsically carries on a hierarchy: a first enzymatic activation has to precede a second one for the liberation to occur. In order to solve this drawback, Dal Corso, Gennari and co-workers¹⁴ developed the first non-sequential dual-responsive self-immolative spacer, considering a protease and a phosphatase sensitive moiety. Both enzymatic steps can occur independently without intrinsic hierarchy but it has to be noted that the protease step is determinant in the self-immolation process. Indeed, the drug release occurs slowly but significantly after the action of the protease even if the phosphatase step did not occur.

In this communication, we report on an alternative AND logic gate strategy, which does not require such a hierarchy for liberating a substrate conditioning it to two trigger. Instead of being sequential, dual activation is parallel so as to confer exactly the same status to both activating triggers.

Results and discussion

Molecular design

Our self-immolative motif is based on an elimination mechanism for self-immolation since this is compatible with liberating alcohols and amines through ether, carbonate, carbamate or ammonium functions.^15,16 In order to avoid any possible steric hindrance, we adopted a heteroatom-para-benzyl core. To enable parallel dual activation, we chose nitrogen as the initiating heteroatom since this atom can be further linked to two activatable groups. To avoid any detrimental self-immolation after the first activation step, we linked the nitrogen to two carbonyl groups so as to deactivate its lone pair up to the second activation step. Further envisioning therapeutical applications, we balanced considerations of stability and reactivity and integrated the nitrogen atom in a di-carbamate. To establish a proof of principle, we eventually retained (i) light and Pd(II) as triggers and respectively introduced 4,5-dimethoxy-2-nitrobenzyl alcohol and allyl alcohol in the di-carbamate and (ii) 7-hydroxy-trifluoromethylcoumarin as a model substrate, which becomes bright when it is liberated from its ether-bound precursor.

We performed a series of experiments on model compound 7 to establish the orthogonality of the triggers and their effectiveness once combined. The validation of our design of a double-triggered self-immolative spacer is shown in Scheme 1.

Scheme 1 Release of the coumarin substrate 11 from its N,N′-biscarbamate precursor 7. Pd(II): Pd(OAc)₂ 2 mol%, TPPTS 4 mol%, and HNEt₂ 2.2 eq.; hν: 365 nm. The last step denoted Δ necessitates H₂O to occur. Solvent: CH₃CN/H₂O (4 : 1 v : v); T = 293 K.

Synthesis of bis-carbamate 7

The synthesis of the bis-carbamate spacer 7 is displayed in Scheme 2. It started from 4-amino benzyl alcohol, which was quantitatively protected with the tert-butyldimethylsilyl group to yield the silyl ether 1. To our knowledge, no dissymmetrical N,N′-biscarbamate has been ever reported in the field of self-immolative spacers. Hence, we had to find conditions for synthesizing this new family of compounds. Among the few related compounds,^17–19 aniline could yield the corresponding bis-carbamate after reaction with anhydride or chloroformate in basic media, which indicated that the first carbamate could react with strong electrophiles under these conditions. Thus, we quantitatively generated the alloc monocarbamate 2 after the reaction of 1 with the corresponding chloroformate. Then we had to introduce the second nitro-veratryl after the reaction of the carbamate moiety. All attempts with a soft base (Et₃N, DIEA, and K₂CO₃) with or without a catalyst (DMAP, KI) were unsuccessful. Although carbamates were often reported as unstable under strongly basic conditions,²⁰ we achieved deprotonation of 2 with several strong bases (LDA, nBuLi, tBuOK, NaH, and KHMDS) to yield the anion that subsequently reacted with the nitro-veratryl chloroformate to afford the dicarbamate 3. After optimisation, we could obtain a 60% yield by using KHMDS as a base, concentration in the 0.3–0.5 M range, and temperature evolving overnight from −78 °C to room temperature. The next step was the deprotection of the silyl ether. Fluoride salts such as TBAF generated significant degradation. In contrast, acidic treatment with Amberlyst®15 cleanly gave the deprotected alcohol (89% yield). The resulting benzylic alcohol 4 was engaged in chlorination with MsCl to give the benzylic chloride 5, which was subsequently converted into benzylic iodide 6 by the Finkelstein reaction. The final compound 7 was eventually obtained by coupling 6 with the coumarin phenol 11 in 49% yield by using Ag₂O in acetonitrile at room temperature. Overall, 7 was obtained in 7 steps with a 20% yield.

Scheme 2 Synthesis of model 7 to establish the proof-of-principle of a double-triggered self-immolative spacer.

Self-immolation experiments

In the first set of preparative experiments (see the ESI†), we first subjected 7 to Pd(II) in a water/acetonitrile mixture in order to remove the alloc carbamate moiety. After 1 h treatment (no more starting material 7 detected by UPLC/MS), we could isolate the nitro-veratryle monocarbamate 8 in 91% yield. We similarly subjected 7 to UV light (365 nm) to remove the nitro-veratryl photolabile protecting group and isolated the alloc monocarbamate 9 in 86% yield after 48 h illumination (disappearance of 7). We further evaluated the stabilities against hydrolysis of the monocarbamates 8 and 9 by UV absorption. Hence, we could evidence the absence of release of coumarin 11 over 24 h, which confirmed that monodeprotection could occur and stop at this stage. In the second series of experiments analyzed by UPLC/MS (see the ESI†), we showed that the exposure of (i) the nitro-veratryl monocarbamate 8 to UV light (365 nm) for 24 h removed the nitro-veratryle photolabile protecting group and liberated coumarin 11 in 86% yield; (ii) the alloc monocarbamate 9 to Pd(II) in a H₂O/CH₃CN mixture removed the alloc carbamate moiety and quantitatively yielded coumarin 11 after 30 min. We eventually demonstrated that coumarin 11 could be quantitatively obtained from 7 upon combining both deprotection conditions (UV illumination + Pd(II)).

To further analyze the liberation of coumarin 11 from the dual activation of 7, we then recorded the temporal evolution of the fluorescence intensity from a 7 solution subjected to both Pd(II) and continuous illumination at 365 nm. Fig. 1 displays the temporal evolution of the fluorescence emission reporting on the release of coumarin 11 upon continuously illuminating a 20 μM solution of precursor 7 containing Pd(II) at 298 K. The complete kinetic analysis is reported in the ESI.† The experiment was reproduced for various incident light intensities to rely on a global fit to extract the kinetic constants associated with the photochemically- and thermally-driven steps. Starting from a poorly fluorescent 7 solution, fluorescence emission increases as anticipated upon releasing 11.

Fig. 1 Temporal evolution of the fluorescence emission (λ_em = 500 nm) upon illuminating at λ_exc = 365 nm a 20 μM 7 solution containing Pd(OAc)₂ 2 mol%, TPPTS 4 mol%, and HNEt₂ 2.2 eq. at various light intensities in the ratios 1 (circles), 0.4 (squares), and 0.2 (triangles). Markers: experimental data; solid line: fits with eqn (8) in the ESI.† Solvent: CH₃CN : H₂O (4 : 1 v : v). T = 298 K.

The time evolution of the fluorescence signal was analyzed by adopting the reduced kinetic scheme shown in Scheme 1S (see the ESI†). In relation to Scheme 2 and after the second series of experiments analyzed by UPLC/MS, the precursor 7 yields the intermediate 10 (rate constant k_low), which subsequently releases coumarin 11 (rate constant k₃). We obtained satisfactory fits by constraining k_low to be proportional to the light intensity (as expected from a photochemical step) and k₃ = 2.2 × 10⁻³ ± 1 × 10⁻³ s⁻¹. The same experiment was reproduced at 310 K (see the ESI†). In addition, we obtained satisfactory fits by constraining k_low to be proportional to the light intensity (as expected from a photochemical step) and k₃ = 8.4 × 10⁻³ ± 1 × 10⁻³ s⁻¹.

Conclusion

Through the synthetic pathway developed in this article, an original asymmetric N,N′-bis-carbamate linker has been obtained with 20% overall yield in 7 steps. These new synthetic conditions open the way to the synthesis of a family of new molecules. The potential use of this structure as a new molecular delivery system, according to the “double gate” strategy, has been established. The structure performed the quantitative release leaving a phenol in a non-sequential activation pathway upon two independent triggers. Moreover, intermediates 8 and 9 generated after single activations were found to be stable under physiological-like conditions, meaning that the second activation was mandatory to initiate the self-immolation process. Despite 6-nitroveratryl and the alloc protecting groups being used in a medicinal chemistry or chemical biology context^21,22 and considering that any carbamate-linked leaving group is supposed to give similar results, these groups could be replaced by biological relevant moieties (enzymatic or metabolic substrates)^23–25 and cytotoxic compounds could replace the coumarin derivative. The use of bis-carbamate was validated on a model and therefore opens the way to applications in medicinal chemistry or chemical biology strategies based on self-immolative spacers.

Experimental

General information

Commercially available chemicals were used without further purification. Anhydrous solvents were freshly distilled before use or obtained from a Pure SolvTM solvent purificator. Low actinic glassware and aluminium film were used for all experiments involving compounds bearing the nitroveratryl moiety. Column chromatography (CC) was led on silica gel 60 (0.040–0.063 mm) (Merck). Analytical and thin layer chromatography (TLC) was performed on Merck silica gel 60 F-254 precoated plates; detection by UV (254 and 365 nm). ¹H NMR spectra were recorded at 300 MHz. ¹³C NMR spectra were recorded at 75 MHz with complete proton decoupling; chemical shifts (δ) in ppm related to protonated solvent as an internal reference (¹H: CHCl₃ in CDCl₃, 7.26 ppm; ¹³C: ¹³CDCl₃ in CDCl₃, 77.16 ppm; ¹H: CH₃OH in CD₃OD, 3.31 ppm; ¹³C: ¹³CD₃OD in CD₃OD, 49.00 ppm; ¹H: CH₃CN in CD₃CN, 1.94 ppm; ¹³C: ¹³CD₃¹³CN in CD₃CN, 1.32, 118.26 ppm); coupling constants J in Hz; current notations are used for multiplicity (s: singlet; bs: broad singlet; d: doublet; dd: double doublet; t: triplet; q: quadruplet; m: multiplet).

Experimental procedures

4-(((tert-Butyldimethylsilyl)oxy)methyl)aniline (1). To a solution of 4-aminobenzyl alcohol (2.00 g, 16 mmol, 1 eq.) in 20 mL of dry methylene chloride (DCM) under argon were added tert-butyldimethylsilyl chloride (2.45 g, 16 mmol, 1 eq.) and imidazole (1.22 g, 18 mmol, 1.1 eq.). The solution was stirred for 1 h at room temperature and then washed with NH₄Cl(aq), distilled water, and brine. The organic phase was dried on MgSO₄, filtered and concentrated under vacuum. The residue was purified by column chromatography (100% DCM) to yield 1 as a dark red oil (3.80 g, 99%). R_f = 0.37 (DCM). δH (CDCl₃) 7.12 (d, 2H, J = 8.3 Hz), 6.66 (d, 2H, J = 8.1 Hz), 4.62 (s, 2H), 3.61 (bs, 2H), 0.93 (s, 9H), 0.08 (s, 6H); δC (CDCl₃) 145.5, 131.7, 127.9, 115.2, 65.2, 26.2, 18.7, −4.9. HRMS [M + H]⁺ calcd for C₁₃H₂₄NOSi 238.1627, found 238.1622.

Allyl (4-(((tert-butyldimethylsilyl)oxy)methyl)phenyl) carbamate (2). To a solution of 1 (500 mg, 2.1 mmol, 1 eq.) in 5 mL of dry THF cooled at 0 °C under argon were injected N,N-diisopropylethylamine (1.8 mL, 10.5 mmol, 5 eq.) and allyl chloroformate (1.1 mL, 10.5 mmol, 5 eq.). The ice bath was removed when the precipitation of salts and discoloration of the red solution were observed. After stirring for 1 h and 30 min, NH₄Cl(aq) and DCM were added. The organic phase was washed with water and brine, dried on MgSO₄, filtered, evaporated and purified on silica column (100% DCM). 2 was isolated as a slightly brown oil (676 mg, quantitative yield). R_f = 0.81 (DCM). δH (CDCl₃) 7.27 (d, 2H, J = 8.1 Hz), 7.15 (d, 2H, J = 8.4 Hz), 6.98 (bs, 1H), 5.85 (m, 1H), 5.24 (dd, 1H, J = 17.2, 1.1 Hz), 5.14 (d, 1H, J = 10.4 Hz), 4.59 (s, 2H), 4.56 (d, 2H, J = 5.6 Hz), 0.85 (s, 9H), 0.01 (s, 6H). δC (CDCl₃) 153.5, 136.8, 136.4, 132.5, 126.9, 118.7, 118.1, 65.8, 64.7, 26.0, 18.4, −5.2. HRMS [M + H]⁺ calcd for C₁₇H₂₈NO₃Si 322.1838, found 322.1831.

N-Allyl-N-4,5-dimethoxy-2-nitrobenzyl (4-(((tert-butyldimethylsilyl)oxy)methyl)phenyl)biscarbamate (3). Compound 2 (564 mg, 1.7 mmol, 1.2 eq.) was dissolved in an inactinic flask containing 5 mL of dry THF cooled at −78 °C. After temperature stabilization, potassium bis(trimethylsilyl)amide (350 mg, 1.7 mmol, 1.2 eq.) was introduced into the solution. After complete dissolution of the base, 4.5-dimethoxy-2-nitrobenzyl chloroformate (403 mg, 1.5 mmol, 1 eq.) was added. The resulting solution was allowed to come back very slowly to room temperature and was further stirred overnight. The organic phase was diluted with DCM, washed with NH₄Cl(aq), distilled water, and brine, dried on MgSO₄ and evaporated. The resulting powder was purified on silica gel (0 to 20% EtOAc/cyclohexane – Cx) to yield 3 as a yellow solid (490 mg, 60%). R_f = 0.47 (30% EtOAc/Cx). δH (CDCl₃) 7.73 (s, 1H), 7.41 (d, 2H, J = 8.2 Hz), 7.25 (d, 2H, J = 7.4 Hz), 6.96 (s, 1H), 5.86 (m, 1H), 5.68 (s, 2H), 5.20 (dd, 2H, J = 12.0, 1.3 Hz), 4.78 (s, 2H), 4.68 (d, 2H, J = 5.3 Hz), 3.95 (s, 3H), 3.80 (s, 3H), 0.97 (s, 9H), 0.14 (s, 6H). δC (CDCl₃) 154.0, 152.6, 152.2, 148.0, 142.1, 139.0, 136.6, 131.3, 128.1, 127.2, 126.7, 118.6, 109.5, 108.1, 67.6, 66.1, 64.4, 56.7, 56.5, 26.1, 18.5, −5.2. HRMS [M + H]⁺ calcd for 561.2268, found: 561.2259 ([M + H]⁺), [M + Na]⁺ calcd for C₂₇H₃₆N₂NaO₉Si 583.2088, found 583.2079 ([M + Na]⁺).

N-Allyl-N-4,5-dimethoxy-2-nitrobenzyl (4-(hydroxymethyl)phenyl)biscarbamate (4). Compound 3 (479 mg, 0.86 mmol) was first dissolved in 4 mL of DCM. Then dry MeOH (4 mL) and a catalytic amount of Amberlyst®15 were added. The suspension was stirred overnight at room temperature. It was filtered and the resin was washed several times with acetone. The resulting solution was concentrated to give a residue, which was purified on a silica column (0 to 100% EtOAc/DCM) to give 4 as a white solid (341 mg, 89%). R_f = 0.11 (40% EtOAc/Cx). δH (DMSO-d6) 7.69 (s, 1H), 7.37 (m, 4H), 6.88 (s, 1H), 5.83 (m, 1H), 5.51 (s, 2H), 5.29 (t, 1H, J = 5.7 Hz), 5.17 (d, 1H, J = 4.8 Hz), 5.12 (s, 1H), 4.62 (d, 2H, J = 4.8 Hz), 4.52 (d, 2H, J = 5.6 Hz), 3.85 (s, 3H), 3.71 (s, 3H). δC (DMSO-d6) 153.4, 151.7, 151.4, 147.6, 142.7, 138.7, 136.4, 132.0, 127.9, 126.9, 126.4, 117.4, 109.5, 108.1, 66.7, 65.1, 62.3, 56.2, 56.1 HRMS [M + H]⁺ calcd for C₂₁H₂₃N₂O₈ 447.1404, found: m/z 447.1400, [M + Na]⁺ calcd for C₂₁H₂₂N₂NaO₈ 469.1223, found 469.1218.

N-Allyl-N-4,5-dimethoxy-2-nitrobenzyl (4-(chloromethyl)phenyl)biscarbamate (5). To a solution of alcohol 4 (229 mg, 0.51 mmol, 1 eq.) in 2 mL of dry DMF was added N,N-diisopropylethylamine (435 μL, 2.6 mmol, 5 eq.) under argon. The solution was then cooled to 0 °C and methanesulfonyl chloride (200 μL, 2.6 mmol, 5 eq.) was injected. The mixture was heated to 85 °C for 2 h before cooling down to room temperature. The organic phase was diluted with DCM and washed with NH₄Cl(aq), distilled water, and brine, dried on MgSO₄ and concentrated. After purification of the residue by column chromatography (0 to 30% EtOAc/Cx), 6 was obtained as a yellow solid (205 mg, 86%). R_f = 0.21 (30% EtOAc/Cx). δH (CDCl₃) 7.64 (s, 1H), 7.39 (d, 2H, J = 8.3 Hz), 7.19 (d, 2H, J = 8.3 Hz), 6.98 (s, 1H), 5.76 (m, 1H), 5.60 (s, 2H), 5.12 (d, 1H, J = 1.2 Hz), 5.07 (dd, 1H, J = 8.8, 1.2 Hz), 4.58 (d, 2H, J = 5.4 Hz), 4.54 (s, 2H), 3.86 (s, 3H), 3.77 (s, 3H). δC (CDCl₃) 153.8, 152.2, 152.0, 148.0, 139.0, 137.8, 131.0, 129.4, 128.5, 126.8, 118.6, 109.6, 108.0, 67.6, 66.1, 56.7, 56.4, 45.3. HRMS [M + Na]⁺ calcd for C₂₁H₂₁ClN₂NaO₈ 487.0884, found 487.0879, 489.0860 ([MCl³⁷ + Na]⁺).

N-Allyl-N-4,5-dimethoxy-2-nitrobenzyl (4-(iodomethyl)phenyl)biscarbamate (6). Compound 5 (175 mg, 0.38 mmol, 1 eq.) was dissolved in 5 mL of distilled acetone and heated to reflux before sodium iodide (564 mg, 3.8 mmol, 10 eq.) was added. After 2 h, the solution was diluted with DCM and washed with water and brine. The organic layer was dried on MgSO₄ and concentrated. The residue was purified on silica gel (100% DCM) to give 6 as a white solid (341 mg, 89%). R_f = 0.21 (30% EtOAc/Cx). δH (CDCl₃) 7.69 (s, 1H), 7.43 (d, 2H, J = 8.2 Hz), 7.18 (d, 2H, J = 8.3 Hz), 7.11 (s, 1H), 5.82 (m, 1H), 5.65 (s, 2H), 5.14 (ddd, 1H, J = 15.1, 10.0, 1.0 Hz), 4.64 (d, 2H, J = 5.3 Hz), 4.45 (s, 2H), 3.92 (s, 3H), 3.88 (s, 3H). δC (CDCl₃) 153.8, 152.2, 152.1, 148.0, 139.7, 139.0, 137.2, 131.0, 129.7, 128.5, 126.9, 118.5, 109.8, 108.0, 67.5, 66.0, 56.7, 56.4, 4.0. HRMS [M + H]⁺ calcd for C₂₁H₂₂IN₂O₈ 557.0421, found 557.0418, [M + H]⁺ calcd for C₂₁H₂₁IN₂NaO₈ 579.0240, found 579.0235.

N-Allyl-N-4,5-dimethoxy-2-nitrobenzyl (4-(((2-oxo-4-(trifluoromethyl)-2H-chromen-7-yl)oxy)methyl)phenyl)biscarbamate (7). To a solution of compound 6 (181 mg, 0.33 mmol, 1 eq.) in 5 mL of dry acetonitrile under argon, 7-hydroxy-4-trifluoromethylcoumarin (188 mg, 0.82 mmol, 2.5 eq.) and silver(I) oxide (378 mg, 1.6 mmol, 5 eq.) were added. After stirring overnight at room temperature, the suspension was filtered on Celite and the solid was washed several times with DCM. The resulting organic solution was washed with aqueous NaHCO₃, water, and brine, dried on MgSO₄, and evaporated. The crude residue was purified on silica gel (0 to 40% EtOAc/Cx) to yield 7 as an off-white solid (97 mg, 45%). Before the physicochemical experiments, another purification was done by HPLC using 80 to 90% gradient of CH₃CN/water/0.1% TFA and the Waters XBridge® Prep C18 5 μm OBD™ 30 × 150 mm column. R_f = 0.29 (40% EtOAc/Cx). δH (CDCl₃) 7.64 (s, 1H), 7.59 (dd, 1H, J = 7.5, 1.4 Hz), 7.43 (d, 2H, J = 8.2 Hz), 7.23 (d, 2H, J = 8.3 Hz), 7.07 (s, 1H), 6.93 (dd, 1H, J = 9.1, 2.3 Hz), 6.88 (d, 1H, J = 2.3 Hz), 6.56 (s, 1H), 5.76 (m, 1H), 5.60 (s, 2H), 5.10 (s, 2H), 5.06 (d, 2H, J = 8.0 Hz), 4.58 (d, 2H, J = 5.3 Hz), 3.86 (s, 3H), 3.80 (s, 3H). δC (CDCl₃) 162.3, 159.2, 156.3, 153.8, 152.2, 148.1, 139.2, 138.0, 135.9, 131.0, 128.6, 128.2, 126.7, 126.6, 118.7, 118.6, 113.8, 112.6, 110.1, 108.1, 107.5, 102.4, 69.9, 67.6, 66.1, 56.6, 56.4. HRMS [M + Na]⁺ calcd for C₃₁H₂₅F₃N₂NaO₁₁ 681.1308, found 681.1304.

4,5-Dimethoxy-2-nitrobenzyl (4-(((2-oxo-4-(trifluoromethyl)-2H-chromen-7-yl)oxy)methyl)phenyl)carbamate (8). Compound 7 (12 mg, 0.02 mmol, 1 eq.) was dissolved in 5 mL of a 4 : 1 (v : v) mixture of acetonitrile and water before palladium(II) acetate (0.1 mg, 0.0004 mmol, 0.02 eq.), triphenylphosphine m-trisulfonate (0.4 mg, 0.0008 mmol, 0.04 eq.) and diethylamine (4 μL, 0.04 mmol, 2.2 eq.) were added. The solution was stirred for 1 h at room temperature, then diluted with DCM and washed with water and brine. After drying on MgSO₄, it was evaporated and the residue was purified on silica gel (0 to 100% DCM/EtOAc). The monocarbamate 8 was obtained as a slightly yellow solid (9.3 mg, 91%). For the physicochemical experiments, a further purification was done by HPLC using 80 to 90% gradient of CH₃CN/water/0.1% TFA and the Waters XBridge® Prep C18 5 μm OBD™ 30 × 150 mm Column. R_f = 0.29 (40% EtOAc/Cx). δH (CDCl₃) 7.65 (s, 1H), 7.56 (dd, 1H, J = 9.0, 1.5 Hz), 7.39 (d, 2H, J = 8.6 Hz), 7.35 (d, 2H, J = 8.5 Hz), 6.97 (s, 3H), 6.91 (dd, 1H, J = 9.0, 2.4 Hz), 6.86 (d, 1H, J = 2.4 Hz), 6.76 (bs, 1H), 6.55 (s, 1H), 5.53 (s, 2H), 5.04 (s, 2H), 3.90 (s, 3H), 3.89 (s, 3H). δC (CDCl₃) 162.4, 159.3, 156.3, 153.4, 152.8, 148.5, 141.8, 141.4, 140.2, 137.8, 130.7, 128.7, 126.4, 119.0, 114.0, 112.4, 110.0, 108.3, 107.2, 102.4, 70.3, 64.1, 56.5, 56.4, 48.4. HRMS [M + H]⁺ calcd for C₂₇H₂₂F₃N₂O₉ 575.1277, found 575.1269, [M + Na]⁺ calcd for C₂₇H₂₁F₃N₂NaO₉ 597.1097, found 597.1095.

Allyl (4-(((2-oxo-4-(trifluoromethyl)-2H-chromen-7-yl)oxy)methyl)phenyl)carbamate (9). Compound 7 (23 mg, 0.035 mmol, 1 eq.) was dissolved in 20 mL of a 4 : 1 (v : v) mixture of acetonitrile and water. It was subsequently stirred and irradiated for 48 h at room temperature by using 3 TLC 6 W lamps emitting at 365 nm. The solution became red/orange over the time. After checking that the starting material disappeared by TLC, the solution was diluted with DCM and then washed with water and brine. After drying on MgSO₄, it was evaporated and the residue was purified on silica gel (0 to 40% EtOAc/Cx). The mono carbamate 9 was obtained as an off-white solid (13 mg, 86%). For the physicochemical experiments, another purification was done by HPLC using 80 to 90% gradient of CH₃CN/water/0.1% TFA and the Waters XBridge® Prep C18 5 μm OBD™ 30 × 150 mm Column. R_f = 0.29 (40% EtOAc/Cx). δH (CD₃CN) 7.63 (bs, 1H), 7.43 (dd, 1H, J = 7.8, 1.7 Hz), 7.27 (d, 2H, J = 8.5 Hz), 7.18 (d, 2H, J = 8.5 Hz), 6.80 (dd, 2H, J = 6.9, 2.4 Hz), 5.77 (m, 1H), 6.44 (s, 1H), 5.14 (dd, 1H, J = 17.3, 1.5 Hz), 5.01 (d, 1H, J = 10.5 Hz), 4.91 (s, 2H), 4.40 (d, 2H, J = 5.3 Hz). δC (CD₃CN) 163.3, 159.9, 157.0, 154.1, 140.6–141.9 (q, CF3), 139.8, 133.9, 131.2, 129.6, 126.9, 124.5, 120.8, 119.3, 117.6, 114.4, 113.9, 113.8, 107.8, 103.3, 71.0, 65.9. HRMS [M + H]⁺ calcd for C₂₁H₁₇F₃NO₅ 420.1059, found 420.1051.

Author contributions

Conceptualization, F. S. and L. J.; methodology, S. H., F. S., T. L. S., and L. J.; software, T. L. S.; formal analysis, S. H., T. L. S, and L. J.; investigation, S. H. and T. L. S.; resources, S. H. and T. L. S.; writing original draft, S. H., T. L. S., L. J., and F. S.; writing-review & editing, S. H., L. J., and F. S.; funding acquisition, L. J., and F. S.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The ComUE Paris Sciences et Lettres (PSL) is acknowledged for financial support to SH. Christine Gaillet is acknowledged for NMR measurements and UPLC experiments.

Notes and references

1. F. Kratz, I. A. Muller, C. Ryppa and A. Warnecke, ChemMedChem, 2008, 3, 20.
2. A. Dal Corso, L. Pignataro, L. Belvisi and C. Gennari, Chem. – Eur. J., 2019, 25, 14740.
3. Y. L. Leu, C. S. Chen, Y. J. Wu and J. W. Chern, J. Med. Chem., 2008, 51, 1740.
4. M. Grinda, B. Renoux, I. Tranoy-Opalinski and S. Papot, MedChemComm, 2012, 3, 68.
5. B. A. Badeau, M. P. Comerford, C. K. Arakawa, J. A. Shadish and C. A. DeForrest, Nat. Chem., 2018, 10, 251.
6. R. J. Amir, M. Popkov, R. A. Lerner, C. F. Barbas III. and D. Shabat, Angew. Chem., Int. Ed., 2005, 44, 4378.
7. S. Debieu and A. Romieu, Org. Biomol. Chem., 2015, 13, 10348.
8. S. Y. Li, L. H. Liu, H. Cheng, B. Li, W. X. Qiu and X. Z. Zhang, Chem. Commun., 2015, 51, 14520.
9. J. B. Grimm, T. D. Gruber, G. Ortiz, T. A. Brown and L. D. Lavis, Bioconjugate Chem., 2016, 27, 474.
10. For selected reviews, see: (a) A. Romieu, Org. Biomol. Chem., 2015, 13, 1294; (b) J. L. Kolanowski, F. Liu and E. J. New, Chem. Soc. Rev., 2018, 47, 195; (c) H.-W. Liu, L. Chen, C. Xu, Z. Li, H. Zhang, X.-B. Zhang and W. Tan, Chem. Soc. Rev., 2018, 47, 7140.
11. G. C. Van de Bittner, C. R. Bertozzi and C. J. Chang, J. Am. Chem. Soc., 2013, 135, 1783.
12. M. Prost and J. Hasserodt, Chem. Commun., 2014, 50, 14896.
13. J. D. Bargh, S. J. Walsh, N. Ashman, A. Isidro-Llobet, J. S. Carroll and D. R. Spring, Chem. Commun., 2021, 57, 3457.
14. A. Dal Corso, S. Arosio, N. Arrighetti, P. Perego, L. Belvisi, L. Pignataro and C. Gennari, Chem. Commun., 2021, 57, 7778.
15. A. Alouane, R. Labruere, T. Le Saux, F. Schmidt and L. Jullien, Angew. Chem., Int. Ed., 2015, 54, 7492.
16. P. J. Burke, J. Z. Hamilton, T. A. Pires, J. R. Setter, J. H. Hunter, J. H. Cochran, A. B. Waight, K. A. Gordon, B. E. Toki, K. K. Emmerton, W. Zeng, I. J. Stone, P. D. Senter, R. P. Lyon and S. C. Jeffrey, Mol. Cancer Ther., 2016, 15, 938.
17. T. J. Thornton and M. Jarman, Synthesis, 1990, 295.
18. M. Nowak, Z. Malinowski, A. Jozwiak, E. Fornal, A. Blaszczyk and R. Kontek, Tetrahedron, 2014, 70, 5153.
19. Y. Basel and A. Hassner, J. Org. Chem., 2000, 65, 6368.
20. J. E. C. Hutchins and T. H. Fife, J. Am. Chem. Soc., 1973, 95, 2282.
21. P. Murat, M. V. Gormally, D. Sanders, M. Di Antonio and S. Balasubramanian, Chem. Commun., 2013, 49, 8453.
22. E. Latocheski, G. M. Dal Forno, T. M. Ferreira, B. L. Oliveira, G. J. L. Bernardes and J. B. Domingos, Chem. Soc. Rev., 2020, 49, 7710.
23. E. W. P. Damen, T. J. Nevalainen, T. J. M. Van der Bergh, F. M. H. de Groot and H. W. Scheeren, Bioorg. Med. Chem., 2002, 10, 71.
24. J. L. M. Jourden and S. M. Cohen, Angew. Chem., Int. Ed., 2010, 49, 6795.
25. B. Huang, W. Chen, Y. Q. Kuang, W. Liu, X. J. Liu, L. J. Tang and J. H. Jiang, Org. Biomol. Chem., 2017, 15, 4383.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ob02124a

---