Multi-Photon-Sensitive Chromophore for the Photorelease of Biologically Active Phenols

Phenols confer bioactivity to a plethora of organic compounds. Protecting the phenolic functionality with photoremovable protecting groups (PPGs) sensitive to two-photon excitation (2PE) can block the bioactivity and provide controlled release of these compounds in a spatially and temporally restricted manner by photoactivation with IR light. To develop an efficient 2PE-sensitive PPG for releasing phenols, the (8-cyano-7-hydroxyquinolin-2-yl)methyl (CyHQ) chromophore was functionalized at the C4 position with methyl, morpholine, methoxy, para-tolyl, and 3,4,5-trimethoxyphenyl groups to provide 4-methyl-CyHQ (Me-CyHQ), 4-morpholino-CyHQ (Mor-CyHQ), 4-methoxy-CyHQ (MeO-CyHQ), 4-(p-tolyl)-CyHQ (pTol-CyHQ), and 4-(3,4,5-trimethoxyphenyl)-CyHQ (TMP-CyHQ) PPGs. The probes possess attributes useful for biological use, including high quantum yield (Φu), hydrolytic stability, and good aqueous solubility in physiological conditions. The MeO-CyHQ PPG enhanced the two-photon uncaging action cross section (δu) of dopamine 3.5-fold (0.85 GM) compared to CyHQ (0.24 GM) at 740 nm and 1.49 GM at 720 nm. MeO-CyHQ was used to mediate photoactivation via 2PE of serotonin, rotigotine, N-vanillyl-nonanoylamide (VNA) (a capsaicin analogue), and eugenol. The constructs except rotigotine showed excellent efficiency in 2PE with δu ranging from 0.75 to 1.01 GM at 740 nm and from 1.31 to 1.36 GM at 720 nm high yielding release of the payloads. These probes also performed well by using conventional single photon excitation (1PE). The spatially and temporally controlled release of dopamine from CyHQ-DA and MeO-CyHQ-DA and serotonin (5-HT) from MeO-CyHQ-5HT was quantified in cell culture by using genetically encoded sensors for dopamine and serotonin, respectively. Calcium imaging was employed to quantify the release of VNA and eugenol (EG) from MeO-CyHQ-VNA and MeO-CyHQ-EG, respectively. These tools will enable experiments to understand the intricate mechanisms involved in neurological signaling and the roles played by neurotransmitters, such as dopamine and serotonin, in the activation of their respective receptors.


■ INTRODUCTION
Phenols are a ubiquitous functional group occurring in a number of biologically active natural and synthetic compounds, such as neurotransmitters and neuromodulators, synthetically prepared agonists and antagonists of receptors and enzymes, steroids and other hormones, and amino acids.Arguably, the two most important endogenous neurotransmitters possessing phenolic functionality are dopamine (DA) and serotonin (5-HT) (Figure 1).Dopamine is the primary agonist of the dopamine receptors, 1,2 which are divided into two subclasses, the D1-and D2-like receptors.Dopamine signaling plays an important role in the regulation of movement as well as reward-driven learning, 3,4 and abnormal dopaminergic function has been associated with diseases such as Parkinson's, 5 Alzheimer's, 6 schizophrenia, 7 bipolar disorder, 8 and attention-deficit hyperactivity disorder (ADHD). 9During the last 20 years, a number of synthetic agonists of dopamine receptors have been developed and used as probes to investigate dopaminergic pathways or as drugs for neurological conditions.A potent synthetic agonist possessing a phenol functionality is rotigotine (RTG), an approved medication for the treatment of Parkinson's disease and restless legs syndrome. 10,11Additionally, rotigotine has been shown to possess antidepressant activity, a common property among dopamine agonists. 12erotonin is an aminophenol neurotransmitter that acts as a hormone in diverse tissues in the central and peripheral nervous systems. 13Its function is regulated by a large number of different receptors that are grouped into seven major families of receptors designated as 5-HT1−7. 14−17 Furthermore, it plays a complex role in mediating pain in the CNS and periphery. 18,19wo other neurologically relevant phenols are N-vanillylnonanoylamide (VNA) and eugenol (EG), natural products derived from chili pepper 20 and clove oil, 21 respectively.They both activate ion channels belonging to the TRP family, a class of nociceptors highly expressed in the peripheral nervous system and involved in the regulation of pain. 22VNA is an equipotent analogue of capsaicin. 23ubstantial efforts have been directed toward understanding the intricate mechanisms involved in neurological signaling and the roles played by neurotransmitters like dopamine and serotonin, as well as other agonists, in the activation of their respective receptors. 24−26 A challenge in these studies is that neurotransmitters simultaneously evoke the activation of a large number of receptors such that it becomes impossible to study the outcome of these agonists on a particular receptor.A key advantage of photoactivation is that photoremovable protecting groups (PPGs) can spatially target different populations of receptors, as well as receptors with different subcellular localizations.In this regard, developing a mechanism of delivery in which neurotransmitters can be made inactive during delivery and then activating them in a spatially and temporally controlled manner when required is highly desirable.Protecting catecholamines on the phenol, as opposed to on the amine, prevents oxidation and yields faster/ more efficient photorelease.The existing coumarin PPGs work poorly for phenols, so reliable options for 2PE-mediated release of this functional group are needed. 27,28PPGs for phenols can also be used to protect and release tyrosine on peptides, thereby substantially expanding the scope beyond drugs and catecholamines.This would eliminate off-target effects and help to develop a deeper understanding of the mechanisms involved in the functioning of these complex neurological systems.−31 This approach masks the activity of the neurotransmitter with a covalently bound photoremovable protecting group (PPG) that can be easily removed, restoring the biological effect, by light irradiation, an exogenous, noninvasive, and traceless stimulus.−31 Nevertheless, most of these PPGs are photolyzed only by UV illumination, and just a handful of them can be cleaved with infrared (IR) light.Radiation in the IR region is particularly useful due to its longer wavelength; it can penetrate deeper into the tissue compared to UV light, and its lower energy minimizes damage to the cells.Developing biologically useful PPGs that absorb in the near-IR region, however, requires overcoming some barriers, including inefficient photolysis, lower excited state energy, and limited solubility. 24,33wo-photon excitation (2PE) is an excellent alternative for releasing PPGs in the near-IR region.This photophysical phenomenon relies on the absorption of two photons simultaneously by the chromophore, thus promoting the molecule to an excited state and triggering a photolysis reaction. 34The efficiency of a PPG toward 2PE-mediated photolysis is defined by the two-photon uncaging action cross section (δ u ), expressed in Goppert-Mayer (GM), which depends upon the two-photon absorbance cross section (δ a ) and the quantum yield of the photochemical reaction (Φ u ).Even though a number of biologically useful PPGs have been designed and synthesized, there is still a significant void in the development of PPGs that can efficiently release neurotransmitters with 2PE.−40 Coumarin-based PPGs have been among the most successful probes, possessing high δ u values (>1 GM) and fast release kinetics, but their bright fluorescence upon excitation can limit their applicability in conjunction with fluorescent indicators. 41,42−49 Previously, we applied this strategy to the direct release of biologically relevant phenols, although low δ u values (<0.36 GM) limited the utility of these probes. 46,50In this work, we carried out a functionalization of the CyHQ core aimed at producing a PPG to release phenols with 2PE cross section values adequate for complex biological studies.Installing a methoxy group at position 4 of the PPG (MeO-CyHQ), significantly improved the photolysis efficiency, resulting in higher δ u values (1.31−1.61GM) than the parent CyHQ (<0.36 GM).Using this strategy, we were able to "cage" a series of phenolic bioeffectors (dopamine, serotonin, rotigotine, N-vanillyl-nonanoylamide, and eugenol) (Figure 1) and successfully release them with 2PE.In addition, proof-ofconcept experiments in biological environments demonstrated the effectiveness of these probes for the study of receptor activation with spatiotemporal control using IR light.

■ RESULTS AND DISCUSSION
There are two critical requirements that a two-photon PPG must meet in order to be useful in a biological setting: (1) it must have a large δ u (to be efficiently activated under 2PE) and (2) release kinetics faster than the rate of diffusion (to avoid diffusion of the neurotransmitter outside the small volume of irradiation).The CyHQ PPG has extremely fast release kinetics on the nanosecond time-scale, 44 but possesses modest δ u values, making it inadequate for applications that demand 2PE-mediated photolysis.We recently discovered that the C4 functionalization with electron-donating groups (EDGs) significantly enhances the photochemical properties of CyHQ, including the δ u . 45,47Using dopamine as a model substrate, we employed some of the best C4-substituted PPGs previously identified in our research to protect its phenolic function, with the aim to identify an efficient system for the photoactivation of phenols under 2PE.Five PPGs were selected: 4-methyl-CyHQ (Me-CyHQ), (TMP-CyHQ).The synthetic pathway for the preparation of PPG-dopamine conjugates 3a−e commenced with the activation of the hydroxyl group of 1a−e (prepared as previously described) 45,47 as a methanesulfonate, furnishing 2a−e.These latter were coupled with N-Boc-dopamine using cesium carbonate as base and subsequently deprotected from the acid-labile groups using trifluoroacetic acid (TFA), affording the protected dopamine analogues 3a−e (Scheme 1).All derivatives were isolated as trifluoroacetic acid salts and a mixture of regioisomers.
The photochemical and photophysical properties were investigated in a simulated physiological buffer (KMOPS, pH 7.2) and compared with the literature data for CyHQ-O-DA (Table 1).A 0.1 mM solution was easily prepared for most compounds with the exception of the pTol-CyHQ-based conjugate which showed slightly lower solubility in aqueous buffer.For this derivative, 10% CH 3 CN in the buffer was used to ensure complete solubilization for the photochemistry experiments.The spontaneous hydrolysis in the dark was monitored over a period of 7 days and no significant hydrolysis was detected for any of the constructs.The molar absorptivity at the wavelength used for 1PE photolysis (ε 365 ) ranged between 4800 and 5670 M −1 cm −1 , in line with the values of most CyHQ-based PPGs.The UV−vis spectra showed a tail of absorption that extends above 400 nm (Supporting Information, UV−vis spectra), enabling the photolysis reaction at 405 nm (a standard wavelength on modern laser confocal microscopes), and a slight blue shift of the maximum absorption band (λ max ) was observed for the MeO-CyHQand Mor-CyHQ-based constructs, compared to CyHQ.
Photolysis reactions driven by 1PE were carried out using 365 nm light, and the resulting time courses are given in Figure 2A.All compounds successfully released their payload with moderate to good yields.The reaction mechanism of the photolysis follows the well-known pathway of hydroxyquinoline PPGs: 49 after light absorption, the anionic species of 3a−e is promoted to the excited state where it undergoes bond cleavage and hydrolysis, leading to the release of the phenol and formation of the benzylic alcohol remnant 4a−e.Intriguingly, major differences were observed in the photolysis efficiency based on the PPG employed.With C4 aromatic substituents, the rate of the photolysis reaction was significantly slower and the quantum yields decreased by 3to 6-fold in comparison to CyHQ-O-DA (Φ u = 0.07 and 0.03 for pTol-CyHQ-O-DA and TMP-CyHQ-O-DA, respectively).This observation contradicts what we previously observed for the photorelease of acetates, 47 where the pTol and TMP cages were among the most efficient PPGs, suggesting that the nature of the cleaved bond (phenol vs acetate) is a critical factor during the photochemical process.We speculate that in the case of phenols, the lower efficiency is due to a photoinduced electron transfer process between the catechol ring and the aromatic C4 substituent, which competes with the photochemical reaction in the excited state.Conversely, the introduction of heteroatom-based C4 substituents (methoxy and morpholino) had a positive effect on the photochemical properties.In this case, the reaction was completed in less than 1 min of irradiation with a 2-fold increase of quantum yield and sensitivity (Φ u = 0.36 and 0.35, sensitivity (ε Φ u ) = 2045 and 1728 for MeO-CyHQ-O-DA and Mor-CyHQ-O-DA, respectively).No significant improvement in the photochemical properties was noted with the introduction of a 4methyl group (Me-CyHQ-O-DA).
We next investigated the ability of the photolysis reactions to be driven by 2PE using a Ti:sapphire laser as a light source.Most of the PPG-dopamine conjugates were efficiently photolyzed by 2PE (Figure 2B), releasing their payload with superior δ u values than CyHQ-O-DA (Table 1).The only exception was TMP-CyHQ-O-DA (3e), whose photolysis could not be executed through 2PE, consistent with what was observed during the 1PE experiments.Intriguingly, MeO-CyHQ-O-DA (3c) emerged as the best photoactivatable dopamine derivative (δ u = 1.49GM), which was 6-fold higher than that of CyHQ-O-DA (0.24 GM).The photolysis of MeO-CyHQ-O-DA was carried out at 720 nm since its λ max is blueshifted (354 nm) compared to the rest of the tested molecules.Under these conditions, up to 28% photolysis was achieved within 30 min of irradiation (Figure 2B) and a significant amount of released dopamine was detected in solution.
From the proof-of-concept investigation using dopamine as a model substrate, MeO-CyHQ emerged as the most efficient PPG.We used this PPG for the photoactivation of other biologically relevant phenols: serotonin, rotigotine, N-vanillylnonanoylamide, and eugenol.For the preparation of these probes, we followed the same procedure used for the PPGdopamine conjugates, starting from mesylate 2c.The two-step sequence, involving coupling with the appropriate phenol in basic media followed by deprotection with trifluoroacetic acid, afforded MeO-CyHQ-protected phenols 4−7c (Scheme 2).with a 365 nm (5 mW power) light emitting diode (LED).Percent remaining was determined by high-performance liquid chromatography (HPLC) analysis and reported as an average of three runs (error bars represent the standard deviation).Lines are least-squares fits of a simple exponential decay (solid lines) and an exponential rise to max (dotted lines).(B) Time courses for the photolysis reactions upon 2PE (720 or 740 nm, 400−500 mW average power).
The photochemical behavior of constructs 4−7c was investigated in KMOPS buffer at the concentration of 0.1 mM.Under these conditions, all compounds were soluble except for the VNA analogue 6c, which, because of the lipophilicity of the vanilloid, required 20% CH 3 CN for full solubilization.The stability toward hydrolysis in the dark was monitored over a week, and no significant decomposition was detected for any probe.All constructs were successfully photolyzed after light exposure, as shown in the time courses of the photolysis reactions (Figure 3).MeO-CyHQ constructs 4c, 6c, and 7c were photoactivated with excellent quantum efficiency and sensitivities, and good chemical yields were obtained by monitoring the appearance of the released bioeffector over the time course of the photoreaction (Table 1).The rotigotine probe 5c displayed low quantum yield and sensitivity (0.12 and 278, respectively) compared to the rest of the series (Φ u = 0.27−0.51and ε Φ u = 1731−2168, respectively).Additionally, MeO-CyHQ-O-RTG did not undergo photorelease through 2PE (vide infra), confirming its poor photochemical efficiency.We believe this effect is due to the presence of the thiophene ring, a well-known photoactive group, 51 which could act as an energy sink in the excited state and quench the photochemical reaction.
The photolysis reaction on 4c, 6c, and 7c could be successfully performed through 2PE.For these experiments, we employed two different wavelengths (720 and 740 nm) and, as in the case of dopamine derivative 3c, the photochemical reactions were more efficient with 720 nm light (Table 1).Under these conditions, high values of two-photon uncaging action cross section were obtained (δ u = 1.31−1.61GM), which is a 4-to 6-fold increase in comparison to previously reported CyHQ-protected phenols. 46,50onsidering the excellent dark stability, good solubility, and larger 2PE cross section, the 4-MeO-CyHQ conjugates with dopamine and serotonin were tested by exploring the activation of the respective genetically encoded sensors dLight 1.2 52,53 and GRAB5-HT 54 expressed on the surface of HCT 116 cells.The efficacy of the MeO-CyHQ constructs of VNA and eugenol was evaluated by activating the TRPV1 receptor expressed on the membrane of HEK 293 cells through calcium imaging.
To establish the cellular assay for photoactivatable dopamine and serotonin, we tested CyHQ-O-DA for its ability to release dopamine and activate dLight because it was previously shown to activate the dopamine-1 receptor on MDA-MB-231 cells in culture and the dopamine-2 receptor in mouse brain slice. 46CT 116 cells expressing dLight 1.2 were treated with CyHQ-Scheme 2. Preparation of Caged Phenolic Bioeffectors 4−7c  The same GRAB5-HT and MeO-CyHQ-O-5HT system was used for the 2PE-mediated release of serotonin.A 200 ms pulse of 720 nm light from a Ti:sapphire laser directed near the membrane of the cell resulted in an increase in serotonin concentration to activate the GRAB5-HT receptor (Figure 5 and Video S2).A robust response with a subsequent decay of the fluorescence signal was observed.No response was observed only in the presence of a 300 ms burst of 720 nm light (Figure S1F, Video S8) in the absence of MeO-CyHQ-O-5HT showing that GRAB5-HT does not respond to light alone.
Calcium imaging of HEK 293 cells expressing TRPV1 receptors on the cell surface was carried out to study the release of VNA and EG from MeO-CyHQ-VNA and MeO-CyHQ-EG, respectively.The conditions were optimized for both 1PE-and 2PE-mediated photoactivation.MeO-CyHQ-VNA (2.5 μM) released VNA in sufficient amounts to elicit a robust response from the calcium-sensitive dye Fluo-4 using a 50 ms pulse of 405 nm wavelength light (Figure S11, Video S21) and EG was released from 5 μM MeO-CyHQ-EG with a 500 ms pulse of 405 nm light (Figure S12, Video S22).These results demonstrate that VNA is a more potent activator of TRPV1 channels.
Using the same TRPV1-expressing HEK 293 cell line, MeO-CyHQ-VNA (5 μM) released a sufficient amount of VNA to generate a robust response from Fluo-4 through 2PE by a pulse of 250 ms, 740 nm light (Figure 6 and Video S19).Owing to its lower potency, a 500 ms light pulse was required to obtain a sufficient response when MeO-CyHQ-EG was used to activate the TRPV1 receptors on the cell surface (Figure 7 and Video S20).Notably, CyHQ-VNA and CyHQ-EG did not induce any response from Fluo-4 upon exposure to 740 nm light, indicating no 2PE-mediated activation of TRPV1 receptors (data not shown).In the absence of MeO-CyHQ-VNA or MeO-CyHQ-EG, the Fluo-4 indicator did not respond to bursts of 400-(1PE) or 740 nm (2PE) light (data not shown).
The value of δ u was higher at 720 nm (1.49 vs 0.85 GM).The MeO-CyHQ PPG was tested for its ability to mediate the release of serotonin, rotigotine, VNA, and eugenol via 1PE and 2PE.With the exception of rotigotine, all of the bioactive phenols were efficiently released through 1PE and 2PE (Φ u = 0.27−0.51and δ u = 0.75−0.85GM at 740 nm and δ u = 1.31− 1.61 GM at 720 nm).Using genetically encoded fluorescent sensors of dopamine (dLight 1.2) and serotonin (GRAB5-HT) and the calcium sensor Fluo-4, we showed that the biological effectors could be released in sufficient quantity through 1PE and 2PE to generate robust signals from the sensors in cell culture.Further optimization efforts will focus on optimizing the concentration of the photoactivatable compounds to use a shorter pulse duration and a lower average laser power than the 24 and 15 mW used here.For comparison, a 0.2 ms pulse of 720 nm light (unreported average power) released a sufficient amount of glutamate from 10 mM MNI-Glu to activate a fluorescent probe of glutamate (iGluSnFR). 55The probes are excellent for use on biological preparations due to sufficient solubility in aqueous media at physiological pH, excellent hydrolytic stability in the dark, fast kinetics, and the production  of benign byproducts after photocleavage.Overall, we created an efficient PPG with sensitivity toward 2PE to release phenols in biologically relevant systems.The tools developed in this study can be used to study the action of these biologically active molecules in tissues.

Synthesis.
General.Commercially available reagents and solvents were employed without further purification. 1H and 13 C NMR spectra were recorded on a Bruker Avance III HD 500 and 600 MHz NMR spectrometer.A Lambda 25 UV−vis−NIR spectrophotometer (PerkinElmer) recorded the UV spectra.The uHPLC analysis and preparative HPLC purifications were performed on an Agilent Infinity series system equipped with an autosampler and diode array detector using Zorbax Eclipse C-18 reversed-phase columns, having a mobile phase composed of water with 0.1% TFA and acetonitrile.An Agilent 6540 HD Accurate Mass QTOF/LC/MS with electrospray ionization (ESI) measured the high-resolution mass spectrometry (HRMS).Purification of the compounds was carried out by flash chromatography on an Isolera Spektra 4 with Biotage SNAP cartridges packed with KPSIL silica.An aqueous solution of 100 mM KCl and 10 mM MOPS (3-(N-morpholino)propanesulfonic acid) titrated to pH 7.2 with 0.1 N NaOH afforded the KMOPS buffer.
General Procedure for the Preparation of 4-R-CyHQ-OMs 2a−e.The alcohols 4-R-CyHQ-OH (0.68 mmol) and triethylamine (0.38 mL, 2.72 mmol) were dissolved in CH 2 Cl 2 (20 mL) in a roundbottom flask and cooled to 0 °C in an ice bath.Methanesulfonyl chloride (0.16 mL, 2.05 mmol) was added to this solution in a dropwise manner.The ice was removed and the reaction was stirred at room temperature for 12 h.After the completion of the reaction, it was diluted with CH 2 Cl 2 (150 mL), washed with H 2 O (3 × 100 mL) and brine (3 × 100 mL), dried over MgSO 4 , and concentrated to dryness.The resulting residue was purified by column chromatography (0−70% EtOAc in n-hexane, gradient elution), yielding the corresponding 4-R-CyHQ-mesylates 2a−e.

Photochemistry. Measurement of UV Spectra and the Molar Extinction Coefficient (ε).
A 100 μM solution of each protected phenol was prepared in KMOPS buffer (20% CH 3 CN was used as a cosolvent for compounds 3d and 6c).The UV−vis spectra of these solutions were recorded at between 250 and 500 nm.All of the measurements were repeated in triplicate, and the absorbance values were averaged.The Beer−Lambert law, ε = A(cl) −1 , was used for the calculation of ε values at λ = 365 nm, where A is the absorbance value measured at 365 nm, c is the concentration of the sample, and l is the cuvette length (1 cm).See the Supporting Information for the spectra.
Spontaneous Hydrolysis in the Dark.A 0.1 mM solution of each compound was prepared in KMOPS buffer.For compounds 3d and 6c, 20% CH 3 CN was used as a cosolvent.The solutions were kept in the dark for 7 days, and the percentage of the starting materials was determined by periodic HPLC analysis (Supporting Information, HPLC Data).Negligible to no hydrolysis was observed for a period of 7 days for all of the compounds.
One-Photon Excitation (1PE).A 0.1 mM solution (3 mL) of each protected phenol in KMOPS buffer (10 or 20% CH 3 CN was used as a cosolvent for 3d and 6c) was placed in a 3 mL quartz cuvette along with a stirring bar.The solution was irradiated with stirring using a 365 nm LED lamp (Cairn OptoLED Lite).At different time intervals, 80 μL aliquots of the solution were taken out and analyzed by uHPLC (Agilent 1290 Infinity series), monitoring the AUC at 320 nm (Supporting Information, HPLC Data).An external standard calibration method of quantification was used for the calibration of the samples.The experiments were repeated in sets of three.A gradient elution with a flux rate of 0.3 mL/min using a mobile phase comprising A = 0.1% trifluoroacetic acid in water and B = acetonitrile (starting from 5% B to 100% over 10 min and re-equilibrating to 5% B before the next run) was used for the separations.A comparison of the AUC with the calibration curves of the substrates (obtained from known concentration) determined the percentages of the remaining starting materials.The percentages of the remaining starting materials were plotted versus time.The time to 90% of decomposition of the substrate (t 90% values) was obtained by fitting a single exponential decay curve to the data using DeltaGraph (Red Rock Software).The following equation was used to calculate the quantum yield (Φ u ) of each photolysis reaction: 42,47,56,57 Φ u = (I σ t 90% ) −1 , where I is the irradiation intensity (determined by potassium ferrioxalate actinometry and measured in Einstein cm −2 s −1 ), 58 σ is the decadic extinction coefficient (1000 × ε) at 365 nm, and t 90% is the time required to consume 90% of the starting material.See Supporting Information Table S1 for the I, σ, and t 90% data.
Two-Photon Excitation (2PE).A 50−100 μM solution (25 μL) of each protected phenol was placed in a microcuvette (26.10F-Q-10,Starna, 10 × 1 × 1 mm illuminated dimensions).The solution was irradiated for different time intervals (typically 5, 10, and 30 min) with 740 nm light from a femtosecond-pulsed and mode-locked Ti:sapphire laser (Mai Tai HP DeepSee, Spectra-Physics) focused on the center of the cuvette chamber.Average power of 450 to 650 mW of the laser was used for the photolysis.The quantification of the samples was carried out the same way as in 1PE experiments (Supporting Information, HPLC Data).DeltaGraph (Red Rock Software) was used to plot the data to fit to a single exponential decay.The following equation was used to calculate the 2-photon uncaging action cross section (δ u ) in Goeppert Mayer (GM, 10 −50 cm 4 •s/photon) using fluorescein as the external standard: 42−44  N p = number of photolyzed per second determined by HPLC analysis.Q f2 = fluorescence quantum yield of the external standard fluorescein (0.9). 59,60 δ aF = 2-photon absorption cross section of fluorescein (30 GM at 740 nm). 61 F = concentration of fluorescein.<F(t)> = time-averaged fluorescence photon flux of the fluorescein standard measured by a radiometer positioned at a right angle to the excitation laser beam (photons/s).
C s = concentration of the sample being photolyzed.φ = estimated collection efficiency of the fluorescence detector, calculated according to the following equation: Technology Platform resources at New York University Abu Dhabi.

Figure 1 .
Figure 1.Structures of the probes developed in this work.

Figure 2 .
Figure 2. Photolysis reaction time courses of PPG-dopamine conjugates.(A) Time courses of the photolysis reactions upon 1PEwith a 365 nm (5 mW power) light emitting diode (LED).Percent remaining was determined by high-performance liquid chromatography (HPLC) analysis and reported as an average of three runs (error bars represent the standard deviation).Lines are least-squares fits of a simple exponential decay (solid lines) and an exponential rise to max (dotted lines).(B) Time courses for the photolysis reactions upon 2PE (720 or 740 nm, 400−500 mW average power).

Figure 3 .
Figure 3. Photolysis reaction time courses of MeO-CyHQ conjugates 4−7c.(A)Time courses of the photolysis reactions upon 1PE with a 365 nm (5 mW power) LED.Percent remaining was determined by HPLC analysis and reported as an average of three runs (error bars represent the standard deviations).Lines are least-squares fits of a simple exponential decay (solid lines) and an exponential rise to max (dotted lines).(B) Time courses for the photolysis reactions on 2PE (720 nm, 400−500 mW average power).

Figure 4 .
Figure 4. Dopamine released in a controlled spatial and temporal manner from MeO-CyHQ-O-DA via 2PE (720 nm) as depicted by the activation of dLight 1.2 expressed on the membrane of HCT 116 cells.(A) Response of dLight to a single, 200 ms pulse of 720 nm light to activate MeO-CyHQ-O-DA (100 μM).See Video S1. (B) Plot of ΔF/F 0 vs time (s) for MeO-CyHQ-O-5HT (100 μM) measured in nine experiments.(C) Plot of average ΔF/F 0 vs time (s) for MeO-CyHQ-O-DA (100 μM) from nine experiments.Gray areas show the standard deviation of the measurement.Arrow marks the time point of the 200 ms light pulse.

Figure 5 .
Figure 5. Serotonin released in a controlled spatial and temporal manner from MeO-CyHQ-O-5HT via 2PE (720 nm) as depicted by the activation of GRAB5-HT expressed on the membrane of HCT 116 cells.(A) Response of GRAB5-HT to a single, 200 ms pulse of 720 nm light to activate MeO-CyHQ-O-DA (200 μM) (see Video S2).(B) Plot of ΔF/F 0 vs time (s) for MeO-CyHQ-O-5HT (200 μM) measured in six experiments.(C) Plot of average ΔF/F 0 vs time (s) for MeO-CyHQ-O-5HT (100 μM) from each of six experiments.Gray areas show the standard deviation of the measurement.The arrow marks the time point of the 200 ms light pulse.

Figure 6 .
Figure 6.Calcium influx measured in HEK 293 cells stably expressing TRPV1 channels after activation of MeO-CyHQ-VNA with a 250 ms flash of 740 nm light (20% power).(A) Plot of ΔF/F 0 vs time (s) for MeO-CyHQ-VNA (5 μM).The graph is an average of three different experiments.Gray areas show the standard deviation of the measurement.The black arrow indicates the timing of the light flash at the 30 s time point.(B) Image of cells before light exposure.(C) Image of cells at peak response of the fluorescent Ca 2+ indicator (Fluo-4).See Video S19.

Figure 7 .
Figure 7. Calcium influx measured in HEK 293 cells stably expressing TRPV1 channels after activation of MeO-CyHQ-EG with a 500 ms flash of 740 nm light (20% power).(A) Plot of ΔF/F 0 vs time (s) for MeO-CyHQ-EG (5 μM).The graph is an average of three different experiments.Gray areas show the standard deviation of the measurement.The black arrow indicates the timing of the light flash (43 s).(B) Image of the cell before light exposure.(C) Image of cells at peak response of the fluorescent Ca 2+ indicator (Fluo-4).See Video S20.

Table 1 .
46otophysical and Photochemical Data for Photoactivatable Phenols a Chemical yield of released compounds.cGM= 10 −50 cm 4 s/photon.dTimeconstant of spontaneous hydrolysis in buffer in the dark at room temperature.eTakenfrom the literature.46fNot determined.g No hydrolysis (<5% detected after 7 days).
b h 10