A Cyanine Photooxidation/β-Elimination Sequence Enables Near-Infrared Uncaging of Aryl Amine Payloads

Abstract

Uncaging strategies that use near-infrared (NIR) wavelengths can enable the highly targeted delivery of biomolecules in complex settings. Many methods, including an approach we developed using cyanine photooxidation, are limited to phenol-containing payloads. Given the critical role of amines in diverse biological processes, we sought to use cyanine photooxidation to initiate the release of aryl amines. Heptamethine cyanines substituted with an aryl amine at the C4’ position undergo only inefficient release, likely due electronic factors. We then pursued the hypothesis that the carbonyl products derived from cyanine photooxidation could undergo efficient β-elimination. After examining both symmetrical and unsymmetrical scaffolds, we identify a merocyanine substituted with indolenine and coumarin heterocycles that undergoes efficient photooxidation and aniline uncaging. In total, these studies provide a new scheme – cyanine photooxidation followed by β-elimination – through which to design photocages with efficient uncaging properties.

Graphical Abstract

Chemical optimization leads to a novel reaction sequence – cyanine photooxidation followed by β-elimination – for near-infrared uncaging.

INTRODUCTION

The photocaged/uncaged paradigm is a useful manifold to modulate biological function with high spatial and temporal control (1, 2). To transition these experiments into increasingly complex physiological environments, photocages activated with single photon flux of far-red and near-infrared wavelengths would be highly enabling. Recent efforts have identified some of the first useful small molecule-based methods in this range. Given the modest photonic energies of these wavelengths, it is not surprising that these approaches employ a range of chemical mechanisms for bond cleavage (3). These include the direct cleavage of certain chromophore-payload bonds, the cleavage of carbon-metal bonds, the use of photoinduced electron transfer (PET)-mediated processes, and the local generation of singlet oxygen (¹O₂) (4–12)(13). Despite progress, significant reaction discovery efforts are needed to uncover methods characterized by broad payload flexibility, activation with wavelengths over 700 nm, and rapid release.

Our group has sought to use cyanine photooxidation as a triggering event to initiate uncaging (14–16). Molecules of the general type shown in Figure 1A undergo ¹O₂-mediated oxidative cleavage between the C2/C1’ and C2’/C3’ positions and then a hydrolysis/cyclization cascade to liberate phenol payloads (17). The key element of this approach is the altered hydrolytic stability of the C4’-N bond following oxidative cleavage. We have applied this approach to create NIR-activated antibody-drug conjugates that release potent anti-tumor agents with in vivo activity (18, 19). Given the critical role that nucleic acids and other amine-containing molecules play in biological processes (20, 21), NIR uncaging strategies that release amine payloads have broad potential impact.

Figure 1.

(A) Key Background and (B) These Studies.

Here we detail efforts to apply cyanine photooxidation as an initiating event for amine uncaging. We find that the direct attachment of aniline payloads to the C4’ position of the heptamethine cyanine leads to only inefficient release under physiological conditions (Figure 1B). We then considered whether the oxidative formation of carbonyl products could be harnessed to initiate a β-elimination event. We prepared derivatives, including a symmetrical heptamethine cyanine and unsymmetrical merocyanines, with suitably positioned carbamate functional groups. Through this approach, we identify a merocyanine that is highly water soluble and undergoes efficient photooxidiation with useful uncaging yields.

MATERIALS AND METHODS

Reagents.

Unless stated otherwise, reactions were conducted in oven-dried glassware under an atmosphere of argon using anhydrous solvents (passed through activated alumina columns). All commercially obtained reagents were used as received. Flash column chromatography was performed using reversed phase (100 Å, 20–40 micron particle size, RediSep® Rf Gold® Reversed-phase C18 or C18Aq) on a CombiFlash® Rf 200i (Teledyne Isco, Inc., Lincoln, NE).

Apparatus.

High-resolution LC/MS analyses were conducted on a Thermo-Fisher LTQ-Orbitrap-XL hybrid mass spectrometer system with an Ion MAX API electrospray ion source in negative ion mode. Analytical LC/MS was performed using a Shimadzu LCMS-2020 Single Quadrupole utilizing a Kinetex 2.6 μm C18 100 Å (2.1 × 50 mm) column obtained from Phenomenex, Inc (Torrance, CA). Runs employed a gradient of 0–90% MeCN/0.1% aqueous formic acid over 4.5 min at a flow rate of 0.2 mL/min. ¹H-NMR and ¹³C-NMR spectra were recorded on Bruker spectrometers (at 400 or 500 MHz or at 100 or 125 MHz) and are reported relative to deuterated solvent signals. Data for ¹H-NMR spectra are reported as follows: chemical shift (δ ppm), multiplicity, coupling constant (Hz), and integration. Data for ¹³C-NMR spectra are reported in terms of chemical shift. Absorbance curves were obtained on a Shimadzu UV-2550 spectrophotometer operated by UVProbe 2.32 software. Samples in 96-well plates (Corning black plate, clear bottom, polystyrene, #3603) were irradiated using a 690 ± 20 nm LED (L690–66-60, Marubeni America Co.) at a light intensity of 20 mW/cm2 as measured using a power meter. Fluorescence traces were recorded on a PTI QuantaMaster steady-state spectrofluorimeter operated by FelixGX 4.2.2 software, with 5 nm excitation and emission slit widths, and a 0.1 s integration rate. Data analysis and curve fitting were performed using MS Excel 2011 and GraphPad Prism 7.

Synthesis of Cy7-C4-Cou.

In a glove-box, IR-783 (100 mg, 0.13 mmol), 7-amino-4-(trifluoromethyl)chromen-2-one, Coumarin 151 (42.8 mg, 0.19 mmol), Pd₂(dba)₃•CHCl₃ (12 mg, 0.013 mmol), BrettPhos (14.5 mg, 0.026 mmol), Cs₂CO₃ (87 mg, 0.26 mmol) and N,N-dimethylformamide (2 mL) were added to an oven dried microwave and the vial was sealed. The resulting heterogenous mixture was heated to 80 ºC and vigorously stirred for 1 h under inert atmosphere, after which time LC/MS analysis of the reaction showed completed consumption of IR-783. The reaction was cooled to room temperature and filtered through celite. The flow-through was diluted with water (3 mL) subjected to reverse-phase column chromatography (C₁₈, 20–60% MeCN/H₂O). The solvent was removed in vacuo to afford Cy7-C4-Cou (71 mg, 57% yield) as a green solid. ¹H NMR (DMSO-d₆, 500 MHz) δ 9.26 (s, 1H), 7.90 (d, J = 14.1 Hz, 2H), 7.57 (d, J = 8.1 Hz, 1H), 7.48 (d, J = 7.3 Hz, 2H), 7.39–7.33 (m, 4H), 7.17 (t, J = 7.7 Hz, 2H), 6.94 (br, 1H), 6.77 (br, 1H), 6.60 (s, 1H), 6.25 (d, J = 14.1 Hz, 2H), 4.13 (br, 4H), 2.71 (br, 4H),1.93 (br, 4H), 1.80–1.67 (m, 8H), 1.32 (s, 12H); ¹³C NMR (DMSO-d₆, 125 MHz) δ 171.63, 159.40, 156.82, 153.04, 152.97, 142.63, 142.35, 141.33, 140.20 (q, J = 30.9 Hz), 128.90, 127.05, 126.67, 124.95, 123.35, 122.77, 121.16, 112.57, 111.62, 110.47, 104.14, 100.82, 100.39, 72.96, 63.54, 51.23, 48.93, 43.97, 27.74, 26.48, 24.51, 23.00, 21.38; HRMS (ESI) calculated for C₄₈H₅₃F₃N₃O₈S₂ (M⁺) 920.3221, observed 920.3200.

Synthesis of 4.

At 0 ºC, phosphorous oxychloride (1.12 g, 7.32 mmol) was added dropwise to anhydrous DMF (1.37 mL, 17.7 mmol). After stirring for 30 min, 4-hydroxy-cyclohexanone, 1, (186 mg, 1.62 mmol) in anhydrous DMF (1 mL, 14.5 mmol) was added dropwise to the reaction at 0 ºC and the mixture was heated at 50 ºC for 1 h in an oil bath. A mixture of aniline/ethanol (50% v/v, 4 mL) was then added dropwise to the mixture at room temperature. The resulting mixture was poured into ice-cold conc. HCl (10:1 v/v, 22 mL). The mixture was allowed to stand for 5 h at –4 ºC and then the formed solid was filtrated off, washed with cold water, ethanol and diethyl ether, and then dried in vacuo to afford chloride 2 as a purple solid (332 mg). The solid was used without further purification. To a 20-mL vial was added indoleine 3 (500 mg, 1.69 mmol),(22) chloride 2 (192 mg, 0.57 mmol), sodium acetate (232 mg, 2.80 mmol), ethanol (2 mL) in succession. Acetic anhydride (286 mg, 2.80 mmol) was added and the reaction was stirred at room temperature for 2 h. The reaction was precipitated into ether (50 mL), the slurry was centrifuged at 3,500 rpm for 5 min and the supernatant was discarded. This procedure was repeated twice. The crude solid was placed under vacuum for 1 h, and the solid was dissolved in H₂O (3 mL). The mixture was subjected to reverse-phase chromatography (C₁₈ column, 0 to 100% MeCN/H₂O). The solvent was removed in vacuo to afford 4 (110 mg, 25% yield) as a green solid. ¹H NMR (Methanol-d₄, 500 MHz) δ 8.51 (d, J = 14.3 Hz, 2H), 7.54 (d, J = 7.4 Hz, 2H), 7.47–7.40 (m, 4H), 7.30 (t, J = 7.43 Hz, 2H), 4.30 (m, 1H), 4.25 (t, J = 7.51 Hz, 4H), 2.98 (dd, J = 15.7, 4.0 Hz, 2H), 2.92 (t, J = 6.9 Hz, 4H), 2.78 (dd, J = 15.7, 6.9 Hz, 2H), 2.0 (m, 8H), 1.76 (s, 12H); ¹³C NMR (Methanol-d₄, 125 MHz) δ 173.00, 148.92, 145.05, 142.21, 141.24, 128.53, 125.13, 123.75, 122.04, 111.06, 101.08, 63.53, 50.33, 49.29, 48.10, 43.66, 33.66, 26.91, 25.89, 22.22. HRMS (ESI) calculated for C₃₈H₄₈ClN₂O₇S₂ (M⁺) 743.2586, observed 743.2570.

Synthesis of Cy7-β-Cou.

To a 25-mL pear-shaped flask was added 7-amino-4-(trifluoromethyl)chromen-2-one (100 mg, 0.44 mmol) and phosgene solution (3 mL, 4.9 mmol; 15%wt in toluene) at room temperature. The mixture was stirred for 20 min at room temperature and then heated to 80 ºC for 13 h. The reaction was cooled and the solvent and excess phosgene was removed in vacuo to afford 6, which was used without further purification. To the resulting solid was added a mixture of 4 (20 mg, 26 µmol), dichloromethane/N-methyl-2-pyrrolidone (50%v/v) and N,N-diisopropylethylamine (45 µL, 0.26 mmol) at 0 ºC. The reaction was stirred for 1 h in an ice bath and then precipitated in ether (50 mL). The slurry was centrifuged at 3,500 rpm for 5 min and the supernatant was discarded. This procedure was repeated twice. The crude solid was placed under vacuum for 1 h, and the solid was dissolved in H₂O/MeCN (3 mL, 50% v/v). The mixture was subjected to reverse-phase chromatography (C₁₈ column, 0 to 100% MeCN/H₂O). The solvent was removed in vacuo to afford Cy7-β-Cou (20 mg, 76% yield) as a green solid. ¹H NMR (Methanol-d₄, 500 MHz) δ 8.53 (d, J = 14.2 Hz, 2H), 7.74 (s, 1H), 7.62 (d, J = 8.6 Hz, 1H), 7.54 (d, J = 7.6 Hz, 2H), 7.46–7.39 (m, 5H), 7.31 (t, J = 7.1 Hz, 2H), 6.72 (s, 1H), 6.38 (d, J = 14.3, 2H), 5.49 (br, 1H), 4.26 (t, J = 7.3 Hz, 4H), 3.24–3.02 (m, 4H), 2.89 (t, J = 7.3 Hz, 4H), 2.08–1.90 (m, 8H), 1.76 (d, J = 7.19 Hz, 12H); ¹³C NMR (Methanol-d₄, 125 MHz) δ 173.34, 159.41, 155.42, 153.27, 148.39, 144.97, 144.06, 142.17, 141.29, 140.76 (q, J = 32.6 Hz), 128.56, 125.33, 125.26, 122.23, 122.05, 115.01, 113.01, 112.96, 111.22, 107.92, 105.36, 101.10, 67.94, 50.37, 49.39, 43.75, 30.78, 26.93, 26.90, 25.96, 22.18. HRMS (ESI) calculated for C₄₉H₅₂ClF₃N₃O₁₀S₂ (M⁺) 998.2729, observed 998.2711.

Synthesis of 9.

To a round-bottom flask was added dichloromethane (30 mL) and pyridine (218 µL, 2.7 mmol). After the mixture was cooled to –78 ºC, glycidol (140 µL, 2.7 mmol) and trifluoromethanesulfonic anhydride (454 µL, 2.7 mmol) were added dropwise in succession to the mixture. The reaction was stirred at –78 ºC for 40 min and was warmed to room temperature. To the mixture was added hexane (30 mL) and the resulting suspension was subjected to short silica gel column chromatography (10 g of silica gel with 30 mL of dichloromethane). The solvent was removed in vacuo and the resulting clear oil was used without further purification. To this oil was added dichloromethane (10 mL) under inert atmosphere. After the mixture was then cooled to –78 ºC, 2,3,3-trimethylindolenine, 7, (216 µL, 1.35 mmol) and N,N-diisopropylethylamine (235 µL, 1.35 mmol) were added to the solution in succession. The reaction was stirred at –78 ºC for 20 min and then warmed up to room temperature over 1 h. The reaction was directly subjected to silica gel column chromatography (0–30% MeOH/DCM). The solvent was removed to afford 9 as a clear oil (215 mg, 75% yield). ¹H NMR (Chloroform-d, 400 MHz) δ 7.82–7.77 (m, 1H), 7.56–7.48 (m, 3H), 5.06 (d, J = 15.1 Hz, 1H) 4.42 (dd, J =15.1, 6.6 Hz, 1H), 3.49 (m, 1H), 2.87 (t, J = 4.3 Hz, 1H), 2.63 (dd, J = 4.4, 2.5 Hz, 1H), 1.53 (d, J = 3.6 Hz, 6H); ¹³C NMR (Chloroform-d, 100 MHz) δ 198.89, 141.26, 141.16, 130.14, 129.55, 123.15, 115.59, 54.88, 49.91, 48.72, 45.72, 22.79, 22.68, 14.65; HRMS (ESI) calculated for C₁₄H₁₈NO⁺ (M⁺) 216.1383, observed 216.1378.

Synthesis of 10.

To a round-bottom flask was added 9 (670 mg, 3.1 mmol) and dichloromethane (30 mL). To the mixture was added trifluoromethanesulfonic acid (603 µL, 6.8 mmol). The reaction was stirred for 20 h at room temperature. The resulting solution was washed with saturated sodium bicarbonate aqueous solution (40 mL) and brine (40 mL). The organic layer was dried over Na₂SO₄, added formic acid (500 µL) and the solvent was removed in vacuo. The resulting oil was subjected to reverse-phase column chromatography (C₁₈, 0–100% MeCN/H₂O). The solvent was removed in vacuo to afford 10 as a clear oil (476 mg, 71% yield). ¹H NMR (Methanol-d₄, 500 MHz) δ 7.80–7.76 (m, 2H), 7.67–7.63 (m, 2H), 4.59 (m, 1H), 4.43–4.34 (m, 2H), 2.23–2.12 (m, 2H), 1.63 (d, J = 21.8 Hz, 6H); ¹³C NMR (Methanol-d₄, 126 MHz) δ 195.51, 141.93, 141.79, 129.86, 128.96, 122.96, 114.14, 60.44, 53.67, 51.78, 23.26, 21.86, 21.41, 19.38; HRMS (ESI) calculated for C₁₄H₁₈NO⁺ (M⁺) 216.1383, observed 216.1380.

Synthesis of 13.

In a round-bottom flask was added N-[5-(phenylamino)-2,4-pentadienylidene]aniline monohydrochloride, 11, (0.50 g, 1.75 mmol), 1,3-dimethyl barbituric acid, 12, (0.27 g, 1.75 mmol), sodium acetate (0.14 g, 1.75 mmol) and acetic anhydride (2.5 mL). The flask was evacuated and flushed with argon. The reaction was stirred at 160 °C for 5 min as the solution color transitioned from dark red to dark brown. The reaction mixture was cooled and concentrated with 2 g of SiO₂ and then purified by normal phase chromatography (80 g silica column, 0–80% EtOAc/Hexane) to afford 13 (0.23 g, 37% yield) as a dark brown solid. ¹H NMR (Chloroform-d₁, 500 MHz) δ 8.16 (d, J = 13.9 Hz, 1H), 8.04 (d, J = 12.5 Hz, 1H), 7.70 (dd, J = 14.4, 12.4 Hz, 1H), 7.59 – 7.52 (m, 3H), 7.23 – 7.18 (m, 3H), 5.40 (dd, J = 13.7, 11.4 Hz, 1H), 3.35 (s, 3H), 3.30 (s, 3H), 1.97 (s, 3H); ¹³C NMR (Chloroform-d, 125 MHz) δ 169.24, 162.60, 161.95, 157.18, 154.95, 151.59, 140.76, 138.23, 130.59, 129.75, 128.29, 126.76, 113.75, 111.60, 28.57, 27.87, 23.37; HRMS (ESI) calculated for C₁₉H₂₀N₃O₄ (MH⁺) 354.1448, observed 354.1444.

Synthesis of BI-β-Cou.

To a microwave vial was added 10 (60 mg, 0.27 mmol), 13 (75.7 mg, 0.21 mmol), sodium acetate (46 mg, 0.56 mmol) and MeOH/DCM (2 mL, 50% v/v) in succession. The vial was sealed and heated with a microwave reactor to 75 °C for 1 h as the solution color transitioned from dark red to blue. The reaction mixture was cooled and concentrated with 0.5 g of silica gel and then purified by normal phase chromatography (12 g silica column, 0–100% ethyl acetate/hexane) to afford the intermediate merocyanine as a dark blue solid, which was immediately used for the next. The intermediate merocyanine in dichloromethane (3 mL) was added to 6 (0.22 mmol, generated as above) at 0 °C. After addition of N,N-diisopropylethyamine (38 µL, 0.22 mmol) at 0 ºC, the reaction was stirred for 30 min in an ice bath. Then the reaction was warmed to room temperature and stirred for 3 h. The reaction mixture was concentrated and purified by normal–phase chromatography (12 g of silica gel, 0–100% EtOAc/hexane) to afford BI-β-Cou (28 mg, 19% yield) as a dark blue solid. ¹H NMR (Chloroform-d, 500 MHz,) δ 8.05 (d, J = 12.7 Hz, 1H), 7.85 (t, J = 13.7 Hz, 1H), 7.67 (s, 1H), 7.62 (d, J = 8.7 Hz, 1H), 7.49–7.45 (m, 2H), 7.32–7.25 (m, 4H), 7.06 (t, J = 7.2 Hz, 1H), 6.80 (d, J = 8.7 Hz, 1H), 6.66 (s, 1H), 6.29 (t, J = 13.0 Hz, 1H), 5.53 (br, 1H), 3.99 (d, J = 13.4 Hz, 1H), 3.88 (d, J = 13.0 Hz, 1H), 3.33 (d, J = 17.4 Hz, 6H), 2.78 (s, 2H), 1.66 (s, 6H); ¹³C NMR (Chloroform-d, 125 MHz) δ 163.27, 162.44, 160.36, 159.15, 156.81, 155.42, 151.99, 151.94, 144.53, 142.68, 142.20, 141.28 (q, J = 33.1 Hz, CF₃), 139.52, 128.25, 126.12, 124.38, 122.68, 122.56, 122.00, 120.37, 120.08, 115.21, 113.66, 109.04, 107.49, 107.36, 106.31, 102.31, 65.96, 47.30, 44.90, 28.75, 28.58, 28.42, 27.74; HRMS (ESI) calculated for C₃₆H₃₂F₃N₄O₇ (MH⁺) 689.2218, observed 689.2216.

Synthesis of 15.

To a round-bottom flask was added N,N-dimethylformamide (2 mL, 25.8 mmol). After cooled to 0 ºC, phosphoryl chloride (298 µL, 3.20 mmol) was added dropwise over 5 min. The reaction was then warmed to room temperature and stirred for 30 min. The reaction was cooled to 0 ºC again and added 2-[6-(diethylamino)-2,3-dihydro-1H-xanthen-9-yl]-benzoic acid 14 (400 mg, 1.07 mmol)(23) in N,N-dimethylformamide (2 mL) dropwise under 0 ºC. The reaction was stirred for 10 min at 0 ºC and for 1 h at room temperature. The reaction was cooled to 0 ºC and quenched with H₂O (2 mL) and stirred at room temperature for 3 d. The resulting solution was subjected to reverse-phase chromatography (C₁₈, 5–100% MeCN/H₂O). The solvent was removed in vacuo to afford aldehyde 15 (356 mg, 82% yield). ¹H NMR (DMSO-d₆, 400 MHz) δ 10.29 (s, 1H), 8.05 (dd, J = 1.3, 7.9 Hz, 1H), 7.70 (dt, J = 1.4, 7.5 Hz, 1H), 7.58 (dt, J = 1.4 and 7.6 Hz, 1H), 7.24 (dd, J = 7.6, 1.1 Hz, 1H), 6.57 (d, J = 2.4 Hz, 1H), 6.39 (dd, J =9.0, 2.4 Hz, 1H), 6.32 (d, J = 9.0 Hz, 1H), 3.36 (q, J = 7.0 Hz, 4H), 2.27 (t, J = 6.0 Hz, 2H), 2.17–2.02 (m, 2H), 1.58–1.44 (m, 2H), 1.09 (t, J = 7.1 Hz, 6H);¹³C NMR (DMSO-d₆, 100 MHz) δ 186.07, 167.65, 161.66, 153.46, 149.59, 140.87, 136.37, 132.81, 131.70, 130.81, 130.40, 128.95, 126.79, 118.61, 111.21, 110.93, 108.25, 97.11, 44.31, 27.39, 21.65, 20.50, 12.92, 12.73. HRMS (ESI) calculated for C₂₅H₂₆NO₄ (MH⁺) 404.1856, observed 404.1850.

Synthesis of CI-β-Cou.

To a 20-mL vial was added 15 (292 mg, 0.68 mmol), 10 (220 mg, 1.02 mmol), sodium acetate (342 mg, 4.07 mmol), ethanol (5 mL) and acetic anhydride (451 µL, 4.07 mmol) in succession. The reaction was heated at 100 ºC for 2 h. After cooled to room temperature, ethyl acetate (50 mL) was added to the reaction and the organic layer was washed with saturated sodium bicarbonate solution (30 mL × 2) and brine (30 mL × 2). The organic layer was dried over Na₂SO₄ and concentrated in vacuo. To the resulting solid was added sodium iodide (4g) in H₂O/MeOH/DCM (4 mL/1 mL/3 mL) and the solution was stirred for 8 h. To the mixture was added dichloromethane (30 mL) and the solution was washed with H₂O (30 mL). The resulting organic layer was dried over Na₂SO₄ and concentrated in vacuo. The crude mixture was purified by silica gel chromatography (40 g of silica gel, 0–100% MeOH/dichloromethane) to afford the intermediate secondary alcohol as a green solid (170 mg, 34% yield). ¹H NMR (Methanol-d₄, 400 MHz) δ 8.63 (s, 1H), 8.01–7.97 (m, 1H), 7.57–7.51 (m, 3H), 7.43 (t, J = 7.7 Hz, 1H), 7.29 (t, J = 7.3 Hz, 1H), 7.24 (d, J = 7.9 Hz, 1H), 7.15–7.11 (m, 1H), 6.94 (dd, J = 9.1, 3.7 Hz, 1H), 6.80–6.76 (m, 1H), 6.63 (br, 1H), 4.49–4.42 (m, 1H), 4.04–3.93 (m, 2H), 3.57 (q, J = 7.1 Hz, 4H), 3.12 (d, J = 3.3 Hz, 2H), 2.99–2.89 (m, 2H), 2.55–2.32 (m, 2H), 1.88 (dd, J = 14.8, 2.4 Hz, 6H), 1.83 (br, 2H), 1.28 (t, J = 7.3 Hz, 6H); ¹³C NMR (Methanol-d₄, 100 MHz) δ 173.07, 170.63, 164.09, 155.89, 154.29, 151.67, 142.54, 142.44, 140.94, 139.72, 133.74, 129.26, 129.02, 128.85, 128.32, 128.16, 124.64, 121.96, 120.93, 114.39, 113.66, 111.69, 109.54, 108.86, 94.89, 61.13, 49.09, 44.68, 32.53, 27.90, 27.38, 26.93, 26.36, 21.30, 11.27; HRMS (ESI) calculated for C₃₉H₄₁N₂O₄⁺ (M⁺) 601.3061, observed 601.3034. To 6 (0.17 mmol, generated as above) was added the intermediate above (20 mg, 0.033 mmol), dichloromethane and N,N-diisopropylethylamine (45 µL, 0.26 mmol) in succession at 0 ºC. After the reaction was stirred for 1 h in an ice bath, the bath was removed and for 12 h. To the reaction was added dichloromethane (50 mL) and washed with saturated aqueous NaHCO₃ solution (50 mL × 2) and brine (50 mL × 2). The organic layer was dried over Na₂SO₄ and the solvent was removed in vacuo. The mixture was subjected to quick silica gel chromatography (24 g of silica gel, 30–100% hexane/ethyl acetate, and then 0–100% MeOH/DCM). The solvent was removed in vacuo to afford a green solid. The solid was dissolved in H₂O/MeOH (50% v/v, 8 mL) and then sodium iodide (3 g) was added. The colored mixture was stirred for 16 h under a light–shielded condition. To the solution was added dichloromethane (50 mL) and washed with H₂O (50 mL). The organic layer was dried over Na₂SO₄ and the solvent was removed in vacuo. Purification by silica gel chromatography was performed (24 g of silica gel, 30–100% hexane/ethyl acetate, and then 0–100% MeOH/DCM). The solvent was removed in vacuo to obtain CI-β-Cou (20 mg, 62% yield) as a green solid. ¹H NMR (Methanol-d₄, 500 MHz) δ 8.62 (s, 1H), 8.05 (d, J = 7.8 Hz, 1H), 7.71 (br, 1H), 7.62–7.54 (m, 4H), 7.42 (t, J = 7.6 Hz, 1H), 7.38–7.23 (m, 3H), 7.15 (d, J = 7.1 Hz, 1H), 6.91 (d, J = 9.1 Hz, 1H), 6.81 (d, J = 8.8 Hz, 1H), 6.70 (s, 1H), 6.63 (s, 1H), 5.58 (br, 1H), 4.26 (dd, J = 14.7 and 70.4 Hz, 2H), 3.58 (q, J = 7.0 Hz, 4H), 3.45 (d, J = 16.0 Hz, 1H), 3.32 (quint, J = 1.7 Hz, 4H), 3.24 (d, J = 16.0 Hz, 1H), 3.30–2.84 (m, 2H), 2.46–2.33 (m, 2H), 1.91 (d, J = 14.8 Hz, 6H), 1.77–1.65 (br, 2H), 1.28 (t, J = 7.1 Hz, 6H); ¹³C NMR (Methanol-d₄, 125 MHz) δ 169.72, 164.24, 159.28, 156.07, 155.36, 155.04, 152.77, 152.01, 143.72, 142.25, 141.88, 140.83, 140.62, 140.51, 134.28, 129.99, 129.84, 128.80, 128.60, 128.38, 128.33, 125.33, 124.59, 123.12, 122.04, 121.01, 120.39, 114.97, 114.81, 113.76, 113.21, 113.15, 112.17, 109.35, 108.09, 107.25, 105.49, 94.85, 65.34, 49.10, 45.85, 44.76, 29.54, 28.06, 27.38, 26.93, 26.04, 21.20, 11.27. HRMS (ESI) calculated for C₅₀H₄₅F₃N₃O₇⁺ (M⁺) 856.3204, observed 856.3160.

Procedure for Determining Photooxidation Efficiency and Uncaging Kinetics.

Stock solutions of the coumarin 151 conjugates Cy7-C4-Cou, Cy7-β-Cou, BI-β-Cou, and CI-β-Cou were diluted into 5% DMSO/pH 7.4 PBS (50 mM) to afford 2 μM solutions. Samples were dispensed into 96-well plates (Corning black plate, clear bottom, polystyrene, #3603) and sealed with ThermalSeal RT film (Sigma, Z707465). Samples in the 96-well plate were irradiated using a 780 LED ± 20 nm (L780–66-60, Marubeni America Co.) for Cy7-C4-Cou and Cy7-β-Cou, 690 nm LED (L690–66-60, Marubeni America Co.) for BI-β-Cou and 740 nm LED (L740–66-60, Marubeni America Co.) for CI-β-Cou at a light intensity of 20 mW/cm² as measured using a power meter. Single point absorption measurements at the λ_max and Coumarin 151 fluorescence (ex 372 nm, em 490 nm) were recorded at regular intervals (1–30 minutes for 12–150 minutes depending on photobleaching rate) until ablation of the absorbance signal. Experiments were run in quadruplicate and plotted with error bars derived from the standard deviation (<5% in all cases). Half-lives were obtained by fitting the curves with one phase decay parameters in all cases except for CI-β-Cou, for which two phase decay parameters were used. Percent yield of Coumarin 151 release was obtained from the input of the experimental fluorescence values into the equation derived from the standard calibration curve, with 2 μM defined as 100% yield (Figure S1).

Procedure for LC/MS Relative Ion Analysis of Cy7-β-Cou Photolysis.

A 10 mM stock of Cy7-β-Cou in DMSO was diluted into water to yield a 10 uM solution. Phenylalanine (40 uM) was used as an internal standard. The photolysis was run for 1 h at 23 ºC with 20 mW/cm² 780 ± 20 nm light in a 2 mL vial without a cap. After 1 h, the reaction was capped and keep in the dark. Aliquots were taken at 0 (prior to irradiation), 1 h, 2 h, 6 h, and 14 h and immediately analyzed by a direct loop injection method with a Shimadzu LCMS-2020 Single Quadrupole instrument (normal resolution) The relative ion counts in Figure 3C were calculated by integrating the extracted ion chromatogram (EIC) of the m/z of Cy7-β-Cou, Int-1 + Int-1-H₂O, Int-2, and C151 and dividing by the ion count of phenylalanine internal standard.

Figure 3.

(A) General scheme for uncaging of Cy7-C4-Cou. (B) Cyanine relative absorbance (780 nm) (left) and fluorescence (ex. 372 nm, em. 490 nm, shown as coumarin 151 concentration [μM], right) traces of a solution of Cy7-C4-Cou (2 μM, pH 7.4 PBS w/ 5% DMSO) that was irradiated (red boxes, 780 nm ± 20 nm) or not and (black boxes).

RESULTS

Based on our prior studies (Figure 1A), a straightforward approach to this problem entails the direct attachment of the aniline-containing substrate onto the C4’ position of the heptamethine scaffold. The preparation of the corresponding aniline-modified cyanines proceeds readily through a Pd(0)-mediated arylation reaction to provide Cy7-C4-Cou directly from the corresponding C4’-chloro IR-783 (Figure 2A). The optical properties of Cy7-C4-Cou are consistent with previously prepared C4’ N-arylated cyanines (Table 1) (24).

Figure 2.

Synthesis of Cy7-C4-Cou and Cy7-β-Cou.

Table 1.

Key parameters.

Cmp.	λabs (nm)a	λem (nm)a	εabs (M−1cm-1)a	Cyanine t1/2 (min)b	Uncaging yield (%)b
Cy-C4-Cou	779	803	2.38 × 105	5.7	0.6 ± 0.04c 8.0 ± 0.1d
Cy-β-Cou	781	805	2.99 × 105	4.9	16 ± 0.2c
BI-β-Cou	664	693	1.61 × 105	20.2	22 ± 1.0c
CI-β-Cou	713	729	1.21 × 105	3.4	33 ± 0.4c

^aMeasured in EtOH.

^bAs determined in Figures 3, 4, and 6.

^cYield after 1 h at room temperature.

^dYield after 24 h at 37 °C.

Photooxidation of Cy7-C4-Cou using a 780 nm LED (20 mW/cm²) in pH 7.4 PBS w/5% DMSO proceeds efficiently, with similar kinetics to our prior efforts (t_1/2 = 5.7 min, Figure 3B). However, irradiation leads to only barely detectable release of the fluorescent coumarin 151 at rt over 1 h (<1 %, Figure S2). When Cy7-C4-Cou is subjected to complete photooxidation at rt and then incubated at 37 °C, modest release occurs over a 24 h time frame (8% vs <1% without irradiation, Figure 3B, see Figure S3 for background release curves). These results suggest that C4’ aniline-substituted cyanines will only be useful for uncaging applications requiring slow release of the payload.

We then set out to identify an approach characterized by rapid kinetics. Our prior efforts were based on the premise that cyanine photooxidation alters the electronic structure of the cyanine scaffold, thereby enhancing the rate of C4’-N hydrolysis. Another feature of photooxidation is the formation of a carbonyl product. Release strategies using a carbonyl formation/β-elimination manifold have been reported in several contexts, leading us to consider if similar logic could be applied here. (25–28) As an initial entry, we generated Cy7-β-Cou through carbamate formation from the corresponding alcohol 4 (Figure 2), which is accessed in two steps from compound 2 (29). Photooxidation of Cy7-β-Cou proceeds efficiently (t_1/2 = 4.9 min, Figure 4). The release of coumarin 151 occurs nearly concurrently with photooxidation at early time points, however, the yield of release only reached 14% after 20 min, and no significant additional release was observed thereafter (Figure 4B). To examine the product mixture being formed in this process, detailed mass spectrometry studies were carried out (Figure 4C). The expected intermediates Int-1 and Int-2 that derive from C2/C1’ and C2’/C3’ photooxidation, respectively, were observed immediately following irradiation. In following the mixture over time, we observed that Int-1 and its hydrate persists for several hours. This suggests that these species are a major and persistent component of the oxidative products derived from Cy7-β-Cou. By contrast, Int-2, disappears with apparent concurrent formation of the coumarin 151 payload. Thus, it appears the β-elimination reaction from Int-2 proceeds efficiently, but the corresponding δ-elimination from Int-1 is inefficient. In separate work, we have determined that cleavage at C2/C1’ occurs preferentially to C2’/C3’ cleavage in roughly a 4:1 ratio,(17) which may explain the modest yield of release.

Figure 4:

(A) General scheme for uncaging from Cy7-β-Cou. (B) Cyanine absorbance (780 nm) (left) and coumarin fluorescence readings (ex. 372 nm, em. 490 nm, shown as concentration [μM], right) of a solution of Cy7-β-Cou (2 μM, pH 7.4 PBS w/ 5% DMSO) that was irradiated (red boxes, 780 nm ± 20 nm) or not and (black boxes). (C) Mass spectrometry analysis following irradiation of solution Cy7-β-Cou (2 μM in H₂O).

These observations suggested that it might be advantageous to pursue strategies where cyanine photooxidation occurs with high regioselectivity. To control the site of photooxidation, we considered using unsymmetrical merocyanines containing an electron-donating indolenine and an electron-deficient heterocycle partner. This approach was prompted in part by prior studies by Hahn that found that certain merocyanines undergo regioselective cleavage at the position adjacent to the more electron-rich heterocycle. (30) We designed molecules that installed a carbamate within a pendent tetrahydropyridine ring system on the electron-rich indolenine heterocycle. We anticipated that (Figure 1C) photooxidation, perhaps occurring selectively adjacent to the electron-rich indolenine heterocycle, would produce a carbonyl intermediate poised for uncaging. For the electron deficient unit, we decided to test merocyanines containing barbituric acid and substituted-coumarin heterocycles. Merocyanines containing various barbituric acid derivatives have been used by Hahn and others as solvent-sensitive fluorophores.(31) The more recently discovered mixed courmain-indoleine merocyanines have been used in several in vivo biological contexts and were also chosen for their improved water solubility.(23, 32)

The synthesis of our target molecules is detailed in Figure 5. The indolenine subunit was prepared through the alkylation of indolenine 7 with glycidol triflate 8 to access epoxide 9.(33) Epoxide 9 undergoes cyclization with triflic acid at r.t. to provide alcohol 10.(34) The regiochemistry of the ring opening reaction to provide 10 was confirmed by COSY analysis (Figure S4). Compound 10 could be converted to the merocyanine BI-β-Cou by addition to polymethine 13, accessed from commercial 11 and barbituic acid 12, and carbamate formation with isocyanate 6. Similarly, the coumarin indocyanine hybrid structure CI-β-Cou could be prepared through formylation of known 14(23) to provide 15, which was converted to the merocyanine through the addition of indoleine 10, followed by carbamate formation with 6.

Figure 5.

Synthesis of BI-β-Cou and CI-β-Cou.

With access to BI-β-Cou and CI-β-Cou, we investigated their optical and uncaging properties. The absorbance/emission maxima of these two molecules, 664/693 nm and 713/729 nm (Table 1), respectively, is in line with prior molecules in these class. Photooxidation of CI-β-Cou proceeds readily with a t_1/2 of 3.4 min, with BI-β-Cou requiring longer irradiation times (t_1/2 = 20.2 min). These highly disparate photobleaching kinetics may reflect differences in aggregation state, with CI-β-Cou being mostly monomeric and BI-β-Cou being partially aggregated in the pH 7.4 PBS buffer used here (Figure S5A,B). Uncaging kinetics and yield were reasonable in both cases, with CI-β-Cou releasing 33% of coumarin 151 after 1 h and BI-β-Cou releasing 22% after 1 h (Figure 6 and Table 1). Some additional release (4–5%) is observed over the next two hours in both cases. Of note, similar yields are obtained with incubation at high temperature (e.g. 37 °C). The analysis of these reactions by mass spectrometry proved ambiguous, and, in particular, it was difficult to assess the site(s) of oxidative cleavage. We speculate that incomplete release in this case may reflect either fluorophore oxidation at some position away from desired site(s) or ¹O₂-mediated oxidation of the uncaged amino coumarin product.(35) In total, these studies have identified scaffolds, particularly the water-soluble CI-β-Cou, that provide a versatile starting point for the development of small molecule photocages capable of uncaging a range of amine containing payloads.

Figure 6:

Uncaging of BI-β-Cou (blue) and CI-β-Cou (red). Absorbance (at the cyanine λ_max) and fluorescence traces (ex. 372 nm, em. 490 nm, shown as concentration [μM] of coumarin 151) for solutions of BI-β-Cou (blue) and CI-β-Cou (red) in pH 7.4 PBS w/ 5% DMSO following irradiation (solid boxes, 660 nm ± 20 nm for BI-β-Cou and 690 nm ± 20 nm for CI-β-Cou) or not (empty boxes).

Conclusion

In conclusion, we report efforts to use cyanine photooxidation as an initiating event for the release of aniline payloads. Our prior studies had shown that certain secondary amines could be released following photooxidative cleavage of heptamethine cyanines with useful yields and reasonable kinetic parameters. These studies illustrate that the release of amine payloads is highly subject to the electronic properties of the amine. Specifically, placing an aryl amine payload at the C4’ position does not appear to interfere with cyanine photooxidation, but leads to only inefficient release (perhaps due to reduced basicity). To develop a more general approach, we have implemented a photooxidation/β-elimination reaction sequence. Embedding a carbamate functional group within the central carbocyclic ring of symmetric heptamethine cyanines improves the kinetics of the uncaging process but the release yield was only modest. Mass spectrometry studies suggest this is likely due to the selective release from only the minor photooxidation product (C2’/C3’ cleavage) and inefficient release from the major product (C2/C1’ cleavage). By contrast merocyanines, and especially the coumarin containing compound, CI-β-Cou, undergoes efficient photooxidation and useful yields of uncaging. It is likely that the improved efficiency of release results from regioselective photooxidation cleavage reaction(s) to provide carbonyl intermediates that then undergo β-elimination. Additional structural optimization to improve the overall yield of release would be enabling for future efforts. In total, these studies suggest that the photooxidation/β-elimination framework provides a promising approach through which to develop NIR-activated photocages.

Supplementary Material

Supp info

Figure S1. Coumarin 151 standard release curve.

Figure S2. Photooxidation and release of Cy7-C4-Cou.

Figure S3. Background Release Curves.

Figure S4A. Absorbance and normalized emission curves of 2 μM Cy7-C4-Cou (A), Cy7-β-Cou (B), BI-β-Cou (C), CI-β-Cou (D) in ethanol.

Figure S4B. Absorbance and normalized emission curves of 2 μM Cy7-C4-Cou (A), Cy7-β-Cou (B), BI-β-Cou (C), CI-β-Cou (D) in 5%DMSO/PBS (pH 7.

Figure S5. NMR Analysis of 10.

¹H and ¹³C NMR Spectra