Research Methodology. The proposed work involves development of coumarin-and BODIPY-photocages:
Synthesis of the PM-targeting caged-signaling molecules. The proposed research aims at the development of PM-targeting Memcaged-glutamate and Memcaged-dopamine probes, and their respective caged-antagonists to achieve selective labelling of neuronal PM, spatiotemporal mono/bidirectional photocontrol on neuronal signaling.
Task 1.1. Synthesis of Memcaged-Glutamate. We adopted a modular approach that involves the caging of glutamate with Coumarin-based fluorescent photocages bearing the lipid membrane linkers (Scheme 1). As the primary choice, we used zwitterionic membrane linker with sulfonate group and C-12 alkyl chain.
As shown in Scheme 1, I synthesized the clickable version of coumarin (1) and zwitterionic membrane linker using the reported methods. Next, I incorporated N-Boc-L-glutamic acid-1-tert-butylester in coumarin 1 by EDC-mediated esterification reaction. In the following step, I tethered the zwitterionic membrane linker with coumarin 2 via copper mediated click reaction. Finally, I obtained the target compound 3 by simultaneous Boc-deprotection and tert-butyl ester hydrolysis with the treatment of 50%TFA in dichloromethane (Scheme 1). In addition, I also synthesized a DEAC-glutamate without membrane linker to highlight that the membrane localized caged-glutamate photorelease a gradient of free glutamate in the proximity of the glutamatergic receptors, and hence need really minimal concentration than caged-glutamate present in the bulk.
Task 1.2. Photophysical and photochemical characterization of Memcaged-Glutamate. I started with the photophysical investigations of memcaged-Glutamate. I measured the absorption/emission wavelengths, fluorescence quantum yield of the probe in the buffer and solvents of different polarity (Table S1). The probe exhibited moderate solvatochromic behavior (Fig.2 b). It exhibited a blue shifted emission spectra with strong enhancement in its emission intensity (Ф =~0.5) in apolar solvents (see Table1).
Further to check the affinity of the probes to biomembranes, DOPC/Cholesterol-liposomes (7:3) was used as the artificial model neural membrane. Mem-DEAC-glutamate displayed strong fluorescence enhancement (Фliposomes = 0.67) upon treating with the liposomes than its buffered solution (Фbuffer = 0.05). In a control experiment, the quenched emission of mem-DEAC-glutamate in buffer was significantly recovered upon addition of 10%TritonX-100 (Ф = 0.67) as shown in Fig 3a. Based on the findings, we speculated that mem-DEAC-glutamate aggregates itself in a micellar form in the buffer, which cause quenching of its fluorescence. Moreover, the micelles disintegrate upon interaction with the lipid membrane, and the probe recovers its emission by diffusion of individual molecules into the lipid membrane (Fig. 3b).
To validate it further, the emission spectra of mem-DEAC-glutamate was recorded in the THF/water binary mixture with varying water proportion. The probe exhibited significant enhancement in its emission with increasing THF content which again confirm the involvement of aggregation caused quenching of emission in buffer (Fig. 4a). Moreover, the emission spectra showed a blue shift in probe’s emission maxima with decreasing water quantity, validating again the moderate solvatochromic nature of probe.
Next, we tested the fate of our control compound, DEAC-glutamate to the lipid membranes. The probe displayed weak emission behavior both in liposomes and buffer (Fig. 4b), revealing the importance of membrane linker for achieving high affinity to the lipid membranes. Further, the photochemical behavior of mem-DEAC-glutamate was checked in water and water/acetonitrile mixture. The solution of probe 70%water/acetonitrile with OD<2 was illuminated with 405 nm laser (Power = 40 mW) at regular interval of time and its uncaging kinetics was monitored by absorption spectra and HPLC. The probe exhibited fast and clean photolysis (Fig. 5). In addition, the probe exhibited decent stability in dark under physiological conditions (t1/2 = ~4 days).
Task 1.3. Application of Memcaged-Glutamate in cellular system. After demonstrating the photophysical/photochemical properties of the probe, I co-applied the probe with Membright-Cy5 to HEK-293 cells to confirm its localization. Like Membright-Cy5, the probe quickly and selectively localized at the cell PM and allowed the wash free imaging (Fig. 6a-c). Next, we checked the fate of probe in cultured neurons (obtained from the cerebellum of 1 week old mice). The probe displayed the selective labeling of neuronal PM similar to HEK-293 cells (Fig. 6d-f). The neuronal cells are quite delicate, and need an immense care and handling. Therefore, to avoid any possible issues, the optimization of conditions (maturity time of neurons to express the glutamatergic receptors, dose of the probe and power of illumination source) for neuronal signaling studies upon photoactivation are currently under progress. We will use Fluo4 stain to visualize the increase in Ca2+ levels upon photoactivation of glutamatergic receptors.
Development of memcaged dopamine:
Initially, we planned to develop Memcaged-dopamine using BODIPY-based photocage. However, the solvatochromic behavior of Memcaged-glutamate invigorated us to consider incorporation of environment sensitivity in the photocages by integrating the push-pull systems in their structural framework. With the decrease in polarity, the solvatochromic probes exhibit high energy excited state, and therefore we hypothesized that it could be highly favorable for the photorelease of a cargo (Fig. 7). Therefore, by assuming that emission and the photorelease occurs in the same excited state, we expect that synergy between microenvironment and photocontrol over the release of signaling molecule could make the photocages as modular photochemical tools for drug/bioactive molecule delivery specifically in targeted cellular compartments, such as biomembranes of target cells or their organelles.
Design and synthesis of solvatochromic probes.
To incorporate this idea, we chose a versatile meso-methyl BODIPY as the chromophore which is easy to functionalize as needed, and stands out for its high extinction coefficient and quantum yield, and. Unlike commonly used coumarins- or nitrobenzyl-, BODIPY-photocages are activated by visible light, reducing phototoxicity and enabling deep tissue penetration. Therefore, we planned to incorporate a donor unit at the 4’-position of BODIPY through vinyl linkage to construct a push-pull framework, and to achieve red wavelength of activation (λabs ~650 nm).First, we designed four probes BPC1- BPC4 by changing the donor substituents (with different electron donating tendency) in the styryl arm (Scheme2). The key BODIPY precursor was prepared using the reported protocol, and the target compounds were obtained by its condensation with respective aldehydes (Scheme2).
The preliminary photophysical investigations of these probes were started in parallel of their synthesis. Solvents of different polarity and viscosity were used to monitor the absorption and emission properties of the probes (Table S2). BPC1/BPC2 did not exhibit any significant change in their emission maxima (λem = ~595 and ~605 nm, respectively) and their respective emission quantum yields (Ф = 0.11 and 0.31) with the decrease in the solvent polarity (Fig.6). Possibly, the negligible or weak electron donating effect of benzene and methoxy benzene substituents in the styryl arm of the probes was not capable enough to construct a strong push-pull system. In contrast, probes BPC3/BPC4 were non-emissive in high polarity aprotic and protic solvents. They displayed the blue shifted (λem = 795-682 nm) and enhanced emission (Ф = 0-0.17) with decrease in solvent polarity (Fig.8 and Fig.S2). This strong solvatochromic behavior of the probes was attributed to their efficient push-pull systems triggered by the presence of strong electron donating N,N-dimethyl amino group in their styryl arms.
First, we investigated the influence of microenvironment on the photochemical response of BPC3. The meso-methyl BODIPY photocages are known to release the cargo via photo-solvolysis. Therefore, we started the photochemical investigations in methanol. Water was avoided due to its immiscibility in apolar solvents like toluene and cyclohexane. The stock solution of BPC3 was diluted in methanol (optical density, OD~0.5). The sample was illuminated with 638 nm laser (60mW, 20 min), and the photoconversion was monitored by TLC. The sample of same concentration was also illuminated in apolar solvents (toluene, dioxane, THF) and different solvent mixtures (10% methanol/toluene, 10% methanol/dioxane, 10% methanol/THF), but no photochemical change was detected.
Literature reports reveal that the BODIPY with fluorinated boron suffer from QYPC, and the replacement of fluorine with methyl group can bring a significant improvement in its QYPC. Therefore, to check it we made two small changes in the structural framework of BPC3 to get a new probe BPC5 (Scheme2). We replaced fluorides with methyl groups, and incorporated p-nitrobenzoic acid (easy to track after release) as the cargo instead of acetate. BPC5 was prepared by one pot hydrolysis of acetate and boron methylation of BPC3 followed by EDC mediated esterification with p-nitrobenzoic acid.
Under same set of conditions, BPC5 exhibited similar photophysical properties and strong solvatochromism like BPC3. Next, we checked the effect of microenvironment on its photochemical behavior. We illuminated the solution of BPC5 in 10% Methanol/toluene with 638 nm laser (60mW, 20 min). After confirming the photoconversion by TLC, the extent of photolysis was estimated by HPLC. The traces clearly displayed the appearance of two new polar species, and significant decrease in the peak of BPC5. Next, to identify the photoproducts, the same illuminated sample was subjected to LC-MS. A careful analysis of LC-MS data revealed the presence of BP1, BP2 and p-nitrobenzoic acid as the major photoproducts (Fig.S3).
As a control, the solution of BPC5 was also illuminated in methanol under similar conditions. The photochemical response of BPC5 was three times lesser in methanol (~10% disappearance) compared to 10% methanol/toluene(~30% disappearance). The photochemical response was not completely silent as the presence of polar protic solvent favors the photosolvolysis. The sample was further illuminated in 10%methanol/acetonitrile and 10% acetonitrile/toluene. In the case of 10% methanol/acetonitrile, BPC5 showed negligible photoconversion, while ~6% disappearance of BPC5 was observed in case of 10% acetonitrile/toluene (Fig.9d) that may be possibly due to the slight moisture in acetonitrile. It indicates clearly that a small proportion of polar protic solvent in apolar media is required for the photoreactivity of BPC5, whereas polar aprotic solvent does not exert much impact on it. Altogether, the findings clearly revealed a tight microenvironment-control over the photoreactivity of solvatochromic probe BPC5.
Further, we synthesized BPC6 (Scheme2) to validate whether boron methylation or presence of nitrobenzoic acid facilitate the efficient photorelease from BPC5. Under similar conditions, HPLC traces of illuminated BPC6 in methanol and 10% Methanol/toluene did not show any photorelease (Fig.10). It reconfirmed the importance of boron methylation to QYPC of BODIPY-photocages instead of nature of cargo. We also synthesized two more probes BPC7 and BPC8 (Scheme1) to demonstrate that the environment-sensitivity only regulates the photoreactivity of a solvatochromic photocage. Non-solvatochromic probe BPC7 showed ~20% photoconversion both in methanol and 10%methanol/toluene upon irradiation with 532 nm laser (60mw, 20 min), (Fig.10). On the contrary, BPC8 exhibited much higher extent of photoconversion (~98% disappearance) in methanol compared to 10%methanol/toluene (~52% disappearance) upon irradiation with 488 nm laser (60mW, 20 min). However, the photolysis of BPC8 was not clean like BPC5 which can be attributed to the poor stability of its photoproducts (Fig.10).
The BODIPY compounds are known photosensitizers that generate singlet oxygen (1O2) upon irradiation. To estimate the 1O2 quantum yield (ФΔ) of BPC5, we used methylene blue (MB,ФΔ = 0.52) as a reference, and AMDA sereved as a 1O2 scavenger which is readily undergoes photobleaching induced by 1O2. The extent of AMDA photobleaching was monitored by measuring its change in absorption at 400 nm in the presence of MB and BPC5, respectively under 638 nm laser irradiation (Fig.S3b). Within one minue, AMDA displayed significant photobleaching in the presence of MB, whereas a minimal photobleaching was observed with BPC5 (ФΔ = 0.0023) even after ~10 min of irradiation (Fig.S3b). It clearly demonstrated that owing to its poor photosensitization efficientcy, BPC5 produces a negligible amount of 1O2, and thus undergoes minimal 1O2 induced oxidative cleavage during uncaging. This observation is consistent with the LC-MS analysis of irradiated BPC5 sample, which revealed a negligible quantity of oxidative cleavage-induced aldehyde photoproduct of BPC5.
Development of membrane probes proton photo-uncaging:
As an additional research directly, which was not originally planned but fully in line with major goal of the project, we work on development of photo-cages for protons (photoacids) for targeting to biomembranes. To this end I took already reported Liao photoacid and functionalized it with membrane anchor. Then, we also synthesized pH-sensitive membrane probe and tested both membrane bound photo-acid and pH probe together on the surface of model lipid membranes. Some, light induced proton liberation was monitored revealing complex time-dependent pH profile. Then, to test these proton photocage on cells, I made a visit to a laboratory of Stephan Kellenberger (Laboratory of ion channel research, University of Lausanne), the well-known expert in acid-sensitive ion channels (ASIC). There, I investigated the photo-induced effect of new membrane bound photoacid on the response of ASIC receptor, measured by electrophysiology method. Even though some photo-stimulation effects were observed, overall the results were not sufficiently conclusive. More research and collaborative efforts will be needed to finalize this research direction. However, those studies were an important preliminary data for larger project, ERC advanced grant (coordinated by A. Klymchenko), which will explore photoacids for communication with cells. Thus, my research had an direct strong impact on further research progress on the host laboratory.
