This is the accepted version of the journal article: Cabré, Gisela; Garrido Charles, Aida; González-Lafont, Àngels; [et al.]. «Synthetic Photoswitchable Neurotransmitters Based on Bridged Azoben- zenes». Organic Letters, Vol. 21, Issue 10 (May 2019), p. 3780-3784. DOI 10.1021/acs.orglett.9b01222 This version is available at https://ddd.uab.cat/record/266063 under the terms of the license Synthetic photoswitchable neurotransmitters based on bridged azobenzenes Gisela Cabréa,‡, Aida Garrido-Charlesb,‡, Àngels González-Lafonta,c, Widukind Moormannd, Daniel Langbehnd, David Egeaa, José M. Llucha,c, Rainer Hergesd, Ramon Alibésa, Félix Busquéa, Pau Gorostizab,e,f,* and Jordi Hernandoa,* a Departament de Química, Universitat Autònoma de Barcelona (UAB), Cerdanyola del Vallès, Spain. b Institut de Bioenginyeria de Catalunya (IBEC), Barcelona Institute of Science and Technology (BIST), Barcelona, Spain. c Institut de Biotecnologia i de Biomedicina (IBB), UAB, Cerdanyola del Vallès, Spain. d Otto Diels-Institute of Organic Chemistry, Christian Albrechts University Kiel, Kiel, Germany. e Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain. f Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina, Zaragoza, Spain. Supporting Information Placeholder ABSTRACT: Photoswitchable neurotransmitters of ionotropic kainate receptors were synthesized by tethering a glutamate moiety to disubstituted C2-bridged azobenzenes, which were prepared through a novel methodology that allows access to diazocines with higher yields and versatility. Because of the singular properties of these photochromes, photoisomerizable compounds were obtained with larger thermal stability for their inert cis isomer than for their biologically activity trans state. This enabled selective neuronal firing violet light upon irradiation without background activity in the dark. green light By enabling remote manipulation of neuronal signaling with light, optogenetics1 and photopharmacology2 have revolution- in the absence of illumination (e.g. when acting as receptor ized neuroscience and neurobiology. Neural receptors re- agonists), which drastically hampers their use. This is the case sponding to glutamate (GluRs), the primary excitatory neuro- of GluAzo (Figure 1a), which is a trans-active, trans-stable transmitter, are one of the major targets in these fields, since partial agonist of ionotropic kainate receptors GluK1 and they regulate several key processes in the nervous system and GluK2,5,14,15 two of the principal GluRs mediating excitatory are related to numerous diseases.3 As such, a plethora of pho- neurotransmission in the central nervous system.3 topharmacological tools have been developed for light-gating To overcome this obstacle while preserving the main design GluRs,4 among which photochromic ligands (PCLs)5-12 are principles behind azo-based PCLs, bridged azobenzenes often preferred because they combine (a) the advantages of (brAzo) such as diazocines (C2 bridge, Figure 1b) could be small-molecule, freely diffusing drugs with (b) the capacity used as photochromes, since they (a) should also favor trans- for reversibly photoswitching their activity without by-product active behavior by switching between extended trans and bent generation nor modification of native receptors.2,4 cis configurations, but (b) exhibit cis thermal stability.16-23 This combination of properties should therefore allow direct The major strategy employed to derive PCLs relies on intro- administration of the inert cis form of the PCL, which could ducing an azoaromatic photoswitch into the structure of well- then be selectively photoactivated. In addition, diazocines known, biologically-active ligands.2,4 Upon trans-cis photo- isomerize with visible light,16-23 which is a further advantage isomerization,13 a geometrical change is induced in these com- with respect to common UV-responding azoaromatic pounds that alters their interaction with the receptor. Typical- compounds.13 However, the limited synthetic accessibility and ly, the more extended configuration of the trans isomer favors versatility of these photochromes has so far limited their such interaction, while affinity is reduced for the folded cis application to the photocontrol of biological systems.22,23b state due to steric effects (i.e. trans-active PCLs).2,4 Because of Actually, their use in photopharmacology to modulate the the inherent photochemical properties of azo-aromatic com- activity of N-methyl-D-aspartate (NMDA) receptors and pounds,13 this imposes a severe limitation to most PCLs de- potassium ion channels has only been reported very recently,24 veloped to date for GluRs5-11 (and other receptors2): they are for which a low-yield 4-mono-substituted diazocine previously active in the dark, where they lie in the more stable trans state. described was employed.23 Consequently, these compounds elicit strong tonic responses complexation energies (in Chemscore units27): 41.2, 37.0, 48.0, 37.3 for trans-Glu_brAzo1, cis-Glu_brAzo1, trans- Glu_brAzo2 and cis-Glu_brAzo2 respectively. Importantly, these figures indicate larger affinity of the trans isomers of Glu_brAzo1-2 to GluK2, thus preserving the targeted trans- active behavior of GluAzo. By contrast, they do not support our hypothesis that the introduction of an ionic, bulky group in Glu_brAzo2 should decrease the binding efficiency of the cis isomer with respect to less hindered Glu_brAzo1. It must be noted, however, that this is compensated by the higher com- plexation energy calculated for trans-Glu_brAzo2 relative to trans-Glu_brAzo1, which arises from the additional attractive hydrogen bonding and lipophilic interactions formed between its bulky group and the receptor. Therefore, an enhanced con- trast in biological activity is indeed to be expected between the two states of Glu_brAzo2, as originally designed. Based on our experience in the synthesis of C2-bridged azo- benzenes16-21 and light-responsive glutamate ligands,28-30 we devised a linear sequence to prepare Glu_brAzo1-2 by Heck coupling reaction between their constituting units: a 3,3’- disubstituted diazocine and previously reported, protected glutamate derivative 129 (Scheme 1). The preparation of the Figure 1. (a) Structure of GluAzo.5 (b) Strategy proposed to photochromic unit started from commercially available 4- prepare trans-active, cis-stable PCLs based on C2-bridged azo- bromo-2-nitrotoluene, 3, which was subjected to deprotonation benzenes for the light-control of neuronal membrane receptors with potassium tert-butoxide and further oxidation with bro- GluK1 and GluK2. (b) Photoisomerization of Glu_brAzo1-2. mine to afford dinitro derivative 4. When attempting common reductive ring-closing conditions on this intermediate that had In this work we aim to apply this strategy to photocontrol been previously reported for the synthesis of diazocines (e.g. GluK1 and GluK2, while broadening its scope by developing Zn, Ba(OH)2),16-21 formation of the desired bridged azoben- diazocine-based PCLs with larger yields and multiple func- zene was only observed with very low yields (< 14%). This tionalization sites to favor chemical versatility. To reach this prompted us to develop a new methodology for the azocy- goal our attention focused on 3,3’-disubstituted diazo- clization process, for which we explored Mills coupling reac- cines,17,18,21 because very inefficient syntheses have been re- tion. With this aim, we first reduced the nitro groups of 4 ported to date for 4,4’-disubstituted analogues (< 1% yield22). using sodium sulphide,3 1 which delivered diamine 5 in 90% Based on these photochromes and the structure of GluAzo,5 yield. Then, oxidative coupling of 5 was undertaken in the we designed Glu_brAzo2 as the first trans-active, cis-stable presence of Oxone®3 2 in glacial acetic acid3 3 at room temper- agonist of GluK1 and GluK2 by tethering a diazocine unit to a ature for 1 h, which nicely furnished dibromosubstituted di- biologically active glutamate moiety through a vinyl linker azocine 6 in good yield (40%). According to this result, Ox- (Figure 1c). In addition, we took advantage of the disubstitu- one®-mediated oxidation of one of the aniline moieties to tion pattern of the photochrome to introduce a lateral ionic, provide the corresponding nitrosoarene followed by condensa- bulky group for (a) enhancing solubility in aqueous media, and tion with the other, unaltered aniline emerges as a novel one- (b) further hindering the interaction of the cis isomer with the pot synthesis of diazocines with reproducible yields. receptor by increasing the steric congestion around the gluta- Azobenzene 6 was next monocyanated under standard mate moiety. The latter should boost the difference in activity Rosenmund-von Braun conditions3 4 to obtain common between the two states of the PCL, a required feature given the asymmetric diazocine 7 in 78% yield (based on recovered 6). moderate photoconversions of most functionalized diazo- This cines.17,18,20-23 To evaluate this effect, Glu_brAzo1 lacking the additional bulky substituent was also synthesized (Figure 1c). To validate our design principles, we conducted molecular docking simulations for the two isomers of Glu_brAzo1-2 on kainate receptors. Our attention particularly focused on GluK2, since (a) it presents a narrower binding cavity that imposes larger steric constraints to ligands,25 and (b) the crys- tallographic structure of its ligand-binding domain after com- plexation with trans-GluAzo is available.15 Dockings were computed on this structure keeping the protein rigid, while the initial geometries of the PCLs were optimized at the B3LYP/6-31G(d) level. In all the cases, the best docking solu- tions placed the glutamate moiety of the ligands in a very Figure 2. Best docking solutions for trans-Glu_brAzo2 (orange) similar position as with trans-GluAzo,26 thus suggesting anal- and cis-Glu_brAzo2 (green) in GluK2. The protein residues ogous interaction with the receptor via hydrogen bonds. How- interacting with the glutamate group of both ligands are also ever, clear differences were observed for the binding arrange- indicated. Oxygen, nitrogen and hydrogen atoms are depicted in ment of each ligand (Figure 2 and S1), which led to different red, blue and white, respectively. Scheme 1. Synthesis of Glu_brAzo1 and Glu_brAzo2 cine-based PCLs,24 thus paving the way for the general appli- cation of this strategy in photopharmacology. By comparison with previous data reported for diazocines,16-21 NMR and UV-vis absorption analysis confirmed obtainment of the cis isomer of Glu_brAzo1-2 due to their larger stability. In addition, upon irradiation of their n-π* absorption band with violet light (abs,max  395 nm, exc = 405 nm), spectral changes were observed indicative of cis-trans isomerization (Figures 3a and S2-S4). This is the case of the new red-shifted absorp- tion band found at abs,max  480 nm, which is distinctive of the trans isomer of diazocines16-18,20-23 and allowed reverting the isomerization process by green light illumination (exc = 532 nm, Figure 3a and S2-S3 and S5). Further characterization of Glu_brAzo1-2 revealed that they undergo cis-trans photo- isomerization with moderate quantum yields (cis-trans  0.1) and efficiencies in aqueous media (Table S1), which resulted in photostationary state mixtures (PSScis-trans) containing 47% and 60% of their trans isomer, respectively. By contrast, trans-cis photoconversion proceeded quantitatively with high quantum yields (trans-cis  0.9), thus ensuring fast light-induced recovery of cis-Glu_brAzo1-2 (Table S1) and repetitive photo- switching without degradation (Figure S6). Back-isomerization of the trans state of these compounds was also found to occur in the dark, though at a very much longer time scale (t1/2  4 h at 298 K in aqueous media, Figure S7 and Table S1). To assess the capacity of Glu_brAzo1-2 to light-gate GluRs, whole-cell voltage clamp recordings were conducted on HEK293 cells expressing GluK1 or GluK2, whose activation causes channel opening and ion flux across the membrane (Figure 1b).3,5 After perfusion of Glu_brAzo1-2, clear chang- es in the currents evoked in these cells were observed upon irradiation, which were consistent with the different absorption properties of their cis and trans isomers (Figure 3b and S8- S10). In particular, maximal inward currents arising from compound was directly coupled to glutamate 1 under Palladi- um catalysis, for which proper selection of the base and the solvent was found to be fundamental. After several tests, the reaction was performed with K3PO4 and N,N- dimethylacetamide3 5 at 140 °C under argon in the presence of 0.1 mol% Pd(OAc)2, which delivered 8 in 61% yield. Finally, acid removal of the protecting groups gave the target com- pound Glu_brAzo1 as its monotrifluoroacetate salt in good yield. For Glu_brAzo2, intermediate 7 was hydrolyzed in basic medium to give the corresponding acid, which was then tethered to previously prepared amine 2 using carbodiimide coupling reagent EDCl·HCl along with HOBt to afford 9 in Figure 3. (a) Absorption spectra of both isomers of Glu_brAzo1 74% yield for the two steps. Heck reaction of this compound and Glu_brAzo2 in PBS:DMSO 1:1, as well as of the PSS mix- with 1 under the aforementioned conditions and subsequent tures obtained under cis-trans (exc = 405 nm) and trans-cis (exc = removal of the protecting groups finally delivered the target 532 nm) photoisomerization. (b) Activation and deactivation PCL. It must be noted that, despite its larger structural com- spectra of GluK2 in HEK293 cells after perfusion of Glu_brAzo1 plexity, Glu_brAzo2 was obtained with an overall yield that is or Glu_brAzo2 and irradiation to induce cis-trans (for activation) ca. 2- to 10-fold larger than those recently reported for diazo- and trans-cis (for deactivation) photoisomerization. Glu_brAzo1-2 were finally tested as photoswitchable neu- rotransmitters in hippocampal neurons where GluK2 is highly expressed.36 Because of their trans-active, cis-stable behavior, physiological behavior was not altered upon perfusion of Glu_brAzo1-2 in the dark, a clear advantage over most PCLs reported for GluRs to date such as GluAzo.5-11 Neuronal firing was instead selectively induced upon cis-trans photoisomeri- zation of Glu_brAzo2 at relatively low concentrations (30 μM, Figure 4a-c) and weak light intensities (22.0 and 47.4 μW mm−2 at exc = 390 and 530 nm, respectively). As a result, sequential and sustained trains of action potentials could be triggered with Glu_brAzo2 by consecutively switching be- tween violet and green light illumination, even at low irradia- tion powers and high excitation frequencies (up to 1 Hz, Fig- ure 4c). Interestingly, this photoinduced behavior was inhibit- ed by addition of DNQX,37 a well-known antagonist of kainate and AMPA GluRs (Figure 4d), which further demonstrates that the efficient photoactivation of hippocampal neurons Figure 4. Whole-cell current clamp recordings of rat hippocampal accomplished with Glu_brAzo2 arose from its light- neurons in culture after perfusion of Glu_brAzo2 (30 μM) and dependent interaction with GluK2. Noticeably, this clean irradiation at exc = 390 (purple) and 530 nm (green). Successful physiological effect was achieved despite the limited activity neuron activation was observed with 390 nm light pulses of (a) 5 modulation measured upon Glu_brAzo2 photoisomerization s, (b) 1s, and (c) 0.5 s (1 Hz stimulation), while it was inhibited in GluK2-expressing HEK293 cells (ca. 20% at 30 M con- upon perfusion of DNQX (10 μM). Neurons were current clamped at -60 mV and, to truly report on their GluK2-mediated centration), while it could not be reproduced with the slightly gating and connectivity, no Concanavalin A was used to prevent less efficient Glu_brAzo1 agonist (Figure S12). We ascribe receptor desensitization after activation. this situation to the well-known nonlinear behavior of neu- ronal signaling, since ligand interaction with a minimum frac- tion of glutamate receptors is needed to overcome the depolar- receptor activation were measured at exc = 390-400 nm (i.e. ization threshold required to initiate an action potential.5 In when inducing cis-trans photoisomerization of the PCLs). our case, such threshold was only surpassed with trans- Minimal signals due to receptor deactivation were instead Glu_brAzo2, probably due to its larger affinity for GluK2, as detected in the dark (i.e. for cis-Glu_brAzo1-2) and at exc > suggested by our measurements in cultured cells. 450 nm (i.e. upon trans-cis photoisomerization). As predicted In conclusion, we demonstrated in this work the potential of by our molecular docking calculations, preferential interaction C2-bridged azobenzenes for the preparation of azo-based between the glutamate agonist ligand and GluK1/GluK2 was photochromic ligands capable of light-controlling neural re- therefore observed for the trans state of Glu_brAzo1-2, which ceptors, which preserve the larger activity typically observed allowed light-induced manipulation of these receptors. for their trans state while ensuring larger stability for the inert The trans-active behavior of Glu_brAzo1-2 was further cis isomer. By introducing a novel synthetic methodology for demonstrated when measuring their dose-response curves with the preparation of 3,3’-disubstituted diazocines based on an GluK1 and GluK2 in the dark and at exc = 390 (10-300 μM, Oxone®-mediated intramolecular azocyclization reaction, the Figure S11). In all the cases, larger inward current signals desired ligands could be obtained with higher yields and larger were retrieved for the PSScis-trans mixture generated upon irra- versatility, thus enabling fine-tuning of their light-dependent diation of the initial cis isomer, though with (a) low-to- biological response. As a proof of concept, this strategy was moderate photoinduced modulation (ca. 10-35% current in- applied to the preparation of new photoswitchable neuro- crease with respect to the dark) and (b) maximal values that transmitters for the light-induced operation of kainate gluta- were only about 20-40% of those evoked by free glutamate at mate receptors, with which we achieved selective neural firing 300 μM. Three main factors should account for these results: upon irradiation without background activity in the dark. incomplete photoconversion into the active trans isomer, as observed in solution (< 60%, Table S1); limited modification ASSOCIATED CONTENT of the glutamate-binding site affinity upon cis-trans isomeriza- Supporting Information tion; and the steric effects imparted by the appended diazocine group with respect to free glutamate even in its less-hindered The Supporting Information is available free of charge on the trans configuration, which are of especial importance for ACS Publications website. GluK2 interaction owing to its narrower ligand binding cavi- General methods and materials, synthetic procedures, and addi- tional computational, photochemical and electrophysiological data ty.25 Actually, the latter accounts for the lower inward current (PDF) signals with higher light-induced selectivity obtained for this receptor. However, it did not prevent Glu_brAzo2 bearing a AUTHOR INFORMATION bulkier terminal group to (a) show higher activity (ca. 2-fold increase with respect to Glu_brAzo1 for GluK2), as anticipat- Corresponding Author ed by our molecular docking simulations, which (b) is further * pau@icrea.cat (P.G.), jordi.hernando@uab.cat (J.H.) modulated upon photoisomerization, as pursued in our initial design. Author Contributions ‡These authors contributed equally. (13) Bandara, H. M. D.; Burdette, S. C. Chem. Soc. Rev. 2012, 41, Notes 1809-1825. The authors declare no competing financial interests. (14) Abrams, Z. R.; Warrier, A.; Trauner, D.; Zhang, X. Front. Neural Circuits 2010, 4, 13. (15) Reiter, A.; Skerra, A.; Trauner, D.; Schiefner, A. Biochemistry ACKNOWLEDGMENT 2013, 52, 8972-8974. We acknowledge financial support from MINECO/FEDER (pro- (16) Siewertsen, R.; Neumann, H.; Buchheim-Stehn, B.; Herges, R.; Nather, C.; Renth, F.; Temps, F. J. Am. Chem. Soc. 2009, 131, jects CTQ2015-65439-R, CTQ2016-80066-R, CTQ2016-75363-R 15594−15595. and CTQ2017-83745-P), AGAUR/Generalitat de Catalunya (17) Sell, H.; Nather, C.; Herges, R. Beilstein J. Org. Chem. 2013, 9, (CERCA Programme and projects 2017-SGR-00465 and 2017- 1−7. SGR-1442), Human Brain Project WAVESCALES projects, (18) Tellkamp, T.; Shen, J.; Okamoto, Y.; Herges, R. Eur. J. Org. Fundaluce foundation, and Deutsche Forschungsgemeinschaft Chem. 2014, 2014, 5456−5461. (SFB677). G. C. acknowledges the Generalitat de Catalunya for (19) Hammerich, M.; Schütt, C.; Stahler, C.; Lentes, P.; Rohricht, F.; her pre-doctoral FI grant. A. G.-C. was supported by fellowship Höppner, R.; Herges, R. J. Am. Chem. Soc. 2016, 138, 13111−13114. BES-2014-068169. (20) Moormann, W.; Langbehn, D.; Herges, R. Synthesis 2017, 49, 3471-3475. (21) Moormann, W.; Langbehn, D.; Herges, R. Beilstein J. Org. Chem. 2019, 16, 727-732. REFERENCES (22) Samanta, S.; Qin, C. G.; Lough, A. J.; Woolley, G. A. Angew. Chem. Int. Ed. 2012, 51, 6452−6455. (1) (a) Deisseroth, K. Nat. Methods 2011, 8, 26-29. (b) Deisseroth, K. (23) (a) Joshi, D. K.; Mitchell, M. J.; Bruce, D.; Lough, A. J.; Yan, H. Nat. Neurosci. 2015, 18, 1213-1225. Tetrahedron 2012, 68, 8670-8676. (b) Eljabu, F.; Dhruval, J.; Yan, H. (2) (a) Gorostiza, P; Isacoff, E. Y. Science 2008, 322, 395-399. (b) Feh- Bioorg. Med. Chem. Lett. 2015, 25, 5594-5596; (c) Jun, M.; Joshi, D. K.; rentz, T.; Schonberger, M.; Trauner, D. Angew. Chem., Int. Ed. 2011, 50, Yalagala, R. S.; Vanloon, J.; Simionescu, R.; Lough, A. J.; Gordon, H. L.; 12156−12182. (c) Kramer, R. H.; Mourot, A.; Adesnik, H. Nat. Neurosci. Yan, H. ChemistrySelect 2018, 3, 2697−2701. 2013, 16, 816−823. (d) Velema, W. A.; Szymanski, W.; Feringa, B. L. J. (24) Thapaliya, E. R.; Zhao, J.; Ellis-Davies, G. C. R. ACS Chem. Neu- Am. Chem. Soc. 2014, 136, 2178−2191. (e) Berlin, S.; Isacoff, E. Y. rosci. 2019, 10.1021/acschemneuro.8b00734. EMBO Rep. 2017, 18, 677-692. (f) Hüll, K.; Morstein, J.; Trauner, D. (25) Mayer, M. L. Neuron, 2005, 45, 539-552. Chem. Rev. 2018, 118, 10710-10747. (26) Guo, Y.; Wolter, T.; Kubar, T.; Sumser, M.; Trauner, D.; Elstner, (3) (a) Traynelis, S. F.; Wollmuth, L. P.; McBain, C. J.; Menniti, F. S.; M. PLoS One 2015, 10, e0135399. Vance, K. M.; Ogden, K. K.; Hansen, K. B.; Yuan, H.; Myers, S. J.; (27) Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G. V.; Mee, Dingledine, R. Pharmacol. Rev. 2010, 62, 405–496. (b) Niswender, C. M.; R. P. J. Comput.-Aided Mol. Des. 1997, 11, 425–445.. Conn, P. J. Annu. Rev. Pharmacol. Toxicol. 2010, 50, 295–322. (c) Bowie, (28) Izquierdo-Serra, M.; Gascón-Moya, M.; Hirtz, J. J.; Pittolo, S.; D. CNS Neurol Disord Drug Targets 2008, 7, 129–143. Poskanzer, K. E.; Ferrer, È.; Alibés, R.; Busqué, F.; Yuste, R.; Hernando, (4) Reiner, A.; Levitz, J.; Isacoff, E. Y. Curr. Opin. Pharmacol. 2015, J.; Gorostiza, P. J. Am. Chem. Soc. 2014, 136, 8693-8701. 20, 135-143. (29) Gascón-Moya, M.; Pejoan, A.; Izquierdo-Serra, M.; Pittolo, S.; (5) Volgraf, M.; Gorostiza, P.; Szobota, S.; Helix, M. R.; Isacoff, E. Y.; Cabré, G.; Hernando, J.; Alibés, R.; Gorostiza, P.; Busqué, F. J. Org. Trauner, D. J. Am. Chem. Soc. 2007, 129, 260-261. Chem. 2015, 80, 9915-9925. (6) Stawski, P.; Sumser, M.; Trauner, D. Angew. Chem. Int. Ed. 2012, (30) Cabré, G.; Garrido-Charles, A.; Moreno, M.; Bosch, M.; Porta-de- 51, 5748-5751. la-Riva, M.; Krieg, M.; Gascón-Moya, M.; Camarero, N.; Gelabert, R.; (7) Pittolo, S.; Gómez-Santacana, X.; Eckelt, K.; Rovira, X.; Dalton, J.; Lluch, J. M.; Busqué, F.; Hernando, J.; Gorostiza, P.; Alibés, R. Nat. Goudet, C.; Pin, J. P.; Llobet, A.; Giraldo, J.; Llebaria, A.; Gorostiza, P. Commun. 2019, 10, 907. Nat. Chem. Biol. 2014, 10, 813-815. (31) Mourot, A.; Kienzler, M. A.; Banghart, M. R.; Fehrentz, T.; (8) Rovira, X.; Trapero, A.; Pittolo, S.; Zussy, C.; Faucherre, A.; Jo- Huber, F. M. E.; Stein, M.; Kramer, R. H.; Trauner, D. ACS Chem. Neu- pling, C.; Giraldo, J.; Pin, J.-P.; Gorostiza, P.; Goudet, C.; Llebaria, A. rosci. 2011, 2, 536–543. Cell Chem. Biol. 2016, 23, 929−934. (32) Yu, B. C.; Shirai Y.; Tour, J. M. Tetrahedron, 2006, 62, 10303- (9) Barber, D. M.; Liu, S. A.; Gottschling, K.; Sumser, M.; Hollmann, 10310. M.; Trauner, D. Chem. Sci. 2017, 8, 611–615. (33) Davey, H. H.; Lee R. D.; Marks, T. J. J. Org. Chem. 1999, 64, (10) Gómez-Santacana, X.; Pittolo, S.; Rovira, X.; Lopez, M.; Zussy, 4976-4979. C.; Dalton, J. A. R.; Faucherre, A.; Jopling, C.; Pin, J.-P.; Ciruela, F.; (34) Friedman, L.; Shechter, H. J. Org. Chem. 1961, 26, 2522-2524. Goudet, C.; Giraldo, J.; Gorostiza, P.; Llebaria, A. ACS Cent. Sci. 2017, 3, (35) Yao, Q.; Kinney, E. P.; Yang, Z. J. Org. Chem. 2003, 68, 7528- 81-91. 7531. (11) Hartrampf, F. W. W.; Barber, D. M.; Gottshcling, K.; Leippe, P.; (36) Carta, M.; Fièvre, S.; Gorlewicz, A.; Mulle, C. Eur. J. Neurosci. Hollmann, M.; Trauner, D. Tetrahedron 2017, 73, 4905-4912. 2014, 39, 1835-1844. (12) Laprell, L.; Repak, E.; Franckevicius, V.; Hartrampf, F.; Terhag, (37) Honore, T.; Davies, S. N.; Drejer, J.; Fletcher, E. J.; Jacobsen, P.; J.; Hollmann, M.; Sumser, M.; Rebola, N.; DiGregorio, D. A.; Trauner, D. Lodge, D.; Nielsen, F. E. Science 1988, 241, 701-703. Nat. Commun. 2015, 6, 8076. Supporting Information Synthetic photoswitchable neurotransmitters based on bridged azobenzenes Gisela Cabréa,‡, Aida Garrido-Charlesb,‡, Àngels González-Lafonta,c, Widukind Moormannd, Daniel Langbehnd, David Egea,a José M. Llucha,c, Rainer Hergesd, Ramon Alibésa, Félix Busquéa, Pau Gorostizab,e,f,* and Jordi Hernandoa,* a Departament de Química, Universitat Autònoma de Barcelona (UAB), Cerdanyola del Vallès, Spain. b Institut de Bioenginyeria de Catalunya (IBEC), Barcelona Institute of Science and Technology (BIST), Barcelona, Spain. c Institut de Biotecnologia i de Biomedicina (IBB), UAB, Cerdanyola del Vallès, Spain. d Otto Diels-Institute of Organic Chemistry, Christian Albrechts University Kiel, Kiel, Germany. e Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain. f Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Zaragoza, Spain. ‡ Equal contribution * Corresponding authors: pau@icrea.cat (P. G.), jordi.hernando@uab.cat (J. H.) S1 Contents Experimental procedure for the synthesis of Glu_brAzo1 and Glu_brAzo2 S3 Computational, photochemical and biological supplementary methods S10 Figure S1. Molecular docking simulations S14 Figures S2-S7 and Table S1. Photoisomerization of Glu_brAzo1 and S15 Glu_brAzo2 Figure S8-S11. Photomodulation of GluK1 and GluK2 in HEK293 cells using S20 Glu_brAzo1 and Glu_brAzo2 Figure S12. Whole-cell voltage clamp recording of rat hippocampal neurons S22 in culture perfused with Glu_brAzo1 and Glu_brAzo2. 1H and 13C NMR spectra of new compounds S23 References S33 S2 Experimental procedure for the synthesis of Glu_brAzo1 and Glu_brAzo2 General procedure: The preparation of ligands Glu_brAzo1 and Glu_brAzo2 was achieved via a multistep synthetic sequence (see Scheme 1 in the article). In both cases, we took commercially available 4-bromo-2-nitrotoluene (3) as a starting material to form the bridged azobenzene core 6, to which fully protected glutamate derivative 1 and, in the case of Glu_brAzo2, amine 2 were introduced. These fragments were obtained from commercially available products as described in refs. 1 (1) and 2 (2). Materials and methods: Commercially available reagents were used as received. Solvents were dried by distillation over the appropriate drying agents. All reactions were monitored by analytical thin-layer chromatography (TLC) using silica gel 60 precoated aluminum plates (0.20 mm thickness). Flash column chromatography was performed using silica gel (230-400 mesh). 1H NMR and 13C NMR spectra were recorded at 250, 360 400 or 500 MHz and 90.5, 100.6 or 125.8 MHz, respectively. Proton chemical shifts are reported in ppm (δ) (CDCl3, δ 7.26 or DMSO-d6, δ 2.50). Carbon chemical shifts are reported in ppm (δ) (CDCl3, δ 77.16 or DMSO-d6, δ 39.5). NMR signals were assigned with the help of COSY, HSQC, HMBC and DEPT135. Infrared peaks are reported in cm–1. Melting points were determined on hot stage and are uncorrected. HRMS were recorded using electron ionization (EI) or electrospray ionization (ESI). Optical rotations were measured at 20 ± 3 °C. Synthesis of Glu_brAzo1 and Glu_brAzo2: The synthesis of Glu_brAzo1 and Glu_brAzo2 proceeded as follows. 1,2-bis(4-bromo-2-nitrophenyl)ethane, 4 To an ice-cooled solution of 4-bromo-2-nitrotoluene, 3, (4.00 g, 18.5 mmol) in anhydrous THF (130 mL), tBuOK (3.10 g, 27.7 mmol) was added and the solution was stirred for 45 s. Then, Br2 (1.4 mL, 27.3 mmol) was added and the mixture was stirred for 1 min. The reaction mixture was let warm up to rt and then was poured into ice. The resulting suspension was filtrated and the solid was washed with acetone to furnish 4 (2.84 g, 6.60 mmol, 71%) as an orange solid: mp 219 ºC (from acetone); 1H NMR (500 MHz, CDCl3) δ 8.16 (d, J3,5 = 2.1 Hz, 2H, H-3), 7.87 (dd, J5,6 = 8.3 Hz, J5,3 = 2.1 Hz, 2H, H-5), 7.41 (d, J6,5 = 8.3 Hz, 2H, H-6), 3.10 (s, 4H, 2xCH2); 13C NMR (125.8 MHz, S3 CDCl3) δ 149.3 (C-2), 134.0 (C-5), 133.9 (C-6), 133.7 (C-1), 126.8 (C-3), 119.6 (C-4), 31.9 (CH2); IR (ATR) 3088, 1557, 1519, 1473, 1390, 1335 cm-1; MS (EI, 70 eV) m/z (%) 213.97 (100) [M2+], 155.97 (66) [M-C8H7BrN2O4]; Anal calcd. for [C14H10Br2N2O4]: C, 39.10; H, 2.35; N, 6.51; found: C, 39.17; H, 2.33; N, 6.29. COSY and 1H/13C correlation were recorded. 6,6'-(ethane-1,2-diyl)bis(3-bromoaniline), 5 Compound 4 (100 mg, 0.23 mmol) was dissolved in a 7:3 degassed mixture of 1,4- dioxane and water (2.6 ml) under N2 atmosphere. The reaction mixture was heated to 90 ºC and sodium sulfide (168 mg, 2.15 mmol) was added portion-wise over 2 h. The mixture was stirred overnight. Then, the reaction was quenched with a saturated aqueous solution of NaHCO3 (4 mL) and the crude product was extracted with CH2Cl2 (3 x 5 ml). The combined organic layers were washed with H2O, dried over anhydrous MgSO4 and the solvent was evaporated under vacuum to furnish a crude solid identified as 5 (77 mg, 0.21 mmol, 90% yield): mp 158.8 ºC (from CH2Cl2); 1H NMR (500 MHz, CDCl3) δ 6.84 (m, 4H, H-4, H-5), 6.81 (m, 2H, H-2), 3.63 (br s, 4H, 2x-NH2), 2.71 (s, 4H, 2xCH2); 13C NMR (125.8 MHz, CDCl3) δ 145.8 (C-1), 131.1 (C-5), 124.7 (C-6), 121.9 (C-4), 120.7 (C-3), 118.5 (C-2), 30.4 (CH2); IR (ATR): 3430, 3354, 1619, 1491, 1409, 1179 cm-1; HRMS (HR-EI) calcd. for [C14H1479Br2N2]: 367.9524; found: 367.9525. COSY and 1H/13C correlation were recorded. (Z)-3,8-dibromo-11,12-dihydrodibenzo[c,g][1,2]diazocine, 6 To a solution of 5 (204 mg, 0.55 mmol) in 150 mL glacial acetic acid was added Oxone® (126 mg, 0.83 mmol) in portions. The mixture was stirred at room temperature for 1 h. The solvent was evaporated under reduced pressure. The residue was dissolved in CH2Cl2 (25 mL) and washed with saturated NaHCO3 solution (30 mL) and saturated NaCl solution (30 mL). The organic layer was dried over anhydrous MgSO4 and the solvent was evaporated in vacuo. The crude product was purified by column S4 chromatography (from cyclohexane/ethyl acetate 10% to ethyl acetate 100%) to give 6 as a yellow solid (80 mg, 0.22 mmol, 40% yield).: mp 179 ºC (from EtOAc); 1H NMR (500 MHz, CDCl3) δ 7.16 (dd, J4,5 = 8.2 Hz, J4,2 = 2.0 Hz, 2H, H-4), 7.00 (d, J2,4 = 2.0 Hz, 2H, H-2), 6.86 (d, J5,4 = 8.2 Hz, 2H, H-5), 2.82 (m, 4H, 2xCH2); 13C NMR (125.8 MHz, CDCl3) δ 155.9 (C-1), 131.2 (C-5), 130.4 (C-4), 126.8 (C-6), 121.3 (C-2), 120.3 (C-3), 31.0 (CH2); IR (ATR) 3042, 2162, 2046, 1743, 1585, 1559, 1465, 1382 cm-1; MS (EI, 70 eV) m/z (%) = 365.92 (6) [M+], 257.99 (14) [M-BrN2], 205.08 (3) [M-Br2], 178.03 (100) [M-Br2N2]; HRMS (EI) calcd. for [C14H1079Br2N2]+: 363.9211; found: 363.9201; Anal calcd. for [C14H10Br2N2]: C, 45.94; H, 2.75; N, 7.65; found: C, 45.83; H, 2.80; N, 7.50. COSY and 1H/13C correlation were recorded. (Z)-8-bromo-11,12-dihydrodibenzo[c,g][1,2]diazocine-3-carbonitrile, 7 To a solution of 6 (600 mg, 1.64 mmol) in dry DMF (4 mL), CuCN (147 mg, 1.64 mmol) was added. The reaction mixture was stirred for 2 h under reflux conditions. Then, it was let cool down to rt and a 10% aqueous solution of ethylendiamine (45 mL) was added. The product was extracted with CH2Cl2 (3 x 50 mL). The combined organic extracts were washed with H2O (2 x 50 mL), dried over anhydrous MgSO4 and concentrated under vacuum. The resulting residue was purified by column chromatography (cyclohexane/EtOAc, 3:1) to furnish unreacted 6 (210 mg, 0.57 mmol, 35%) and compound 7 (260 mg, 0.83 mmol, 51% yield, 78% yield considering recovered 6) as a yellow solid. Compound 7: Mp 179.3 ºC (from EtOAc); 1H NMR (500 MHz, CDCl3) δ 7.34 (dd, J4,5 = 7.9 Hz, J4,2 = 1.6 Hz, 1H, H-4), 7.18 (dd, J4ii,5ii = 8.2 Hz, J4ii,2ii = 2.0 Hz, 1H, H-4ii), 7.14 (d, J2,4 = 1.6 Hz, 1H, H-2), 7.12 (d, J5,4 = 7.9 Hz, 1H, H- 5), 7.02 (d, J2ii,4ii = 2.0 Hz, 1H, H-2ii), 6.87 (d, J5ii,4ii = 8.2 Hz, 1H, H-5ii), 2.90 (m, 4H, H-1i, H-2i); 13C NMR (125.8 MHz, CDCl3) δ 155.8 (C-1ii), 155.1 (C-1), 133.6 (C-6), 131.3 (C- 5ii), 130.9 (C-4), 130.7 (C-5), 130.7 (C-4ii), 126.2 (C-6ii), 122.5 (C-2), 121.7 (C-2ii), 120.6 (C-3ii), 117.9 (CN), 111.2 (C-3), 31.7 (C-1i), 30.8 (C-2i); IR (ATR) 3045, 2922, 2231, 1587, 1471 cm-1; HRMS (HR-EI) calcd. for [C15H1079BrN3]: 311.0058; found: 311.0056; calcd. for [C15H1081BrN3]: 313.00361; found: 313.00361. COSY and 1H/13C correlation were recorded. di-tert-butyl (2S,4R)-2-((tert-butoxycarbonyl)amino)-4-((E)-3-((Z)-8-cyano-11,12- dihydrodibenzo[c,g][1,2]diazocin-3-yl)allyl)pentanedioate, 8 S5 A mixture of diazocine 7 (30 mg, 96 µmol), glutamate derivative 1 (46 mg, 0.12 mmol), Pd(OAc)2 (3 mg, 10 mol) and K3PO4 (30 mg, 0.14 mmol) in DMAc (0.5 mL) under argon atmosphere was heated at 140 ºC overnight. Then, the solvent was evaporated and the resulting residue was purified by column chromatography (hexane/EtOAc, 3:1) to give 8 (37 mg, 59 mol, 61% yield) as a yellow solid: mp 66.5 ºC (from EtOAc); [α]D20 14.4 (c 1.0, CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.30 (dd, J4iv,5iv = 7.9 Hz, J4iv,2iv = 1.7 Hz, 1H, H-4iv), 7.11 (d, J5iv,4iv = 7.9 Hz, 1H, H-5iv), 7.09 (d, J2iv,4iv = 1.7 Hz, 1H, H-2iv), 7.00 (d, J4ii,5ii = 8.0 Hz, 1H, H-4ii), 6.89 (d, J5ii,4ii = 8.0 Hz, 1H, H-5ii), 6.87 (s, 1H, H-2ii), 6.29 (d, J3i,2i = 15.8 Hz, 1H, H-3i), 6.07 (m, 1H, H-2i), 4.99 (d, JNHBoc,2 = 8.5 Hz, 1H, NHBoc), 4.23 (m, 1H, H-2), 2.90 (m, 4H, H-1iii, H-2iii), 2.44 (m, 2H, H-1i, H-4), 2.35 (m, 1H, H-1i), 2.15 (m, 1H, H-3), 1.68 (m, 1H, H-3), 1.44 (s, 9H, C(CH3)3), 1.43 (s, 9H, C(CH3)3), 1.40 (s, 9H, C(CH3)3); 13C NMR (125.8 MHz, CDCl3) δ 173.9 (C-5), 171.7 (C- 1), 155.6 (C-1ii), 155.1 (C-1iv), 136.7 (C-3ii), 134.3 (C-6iv), 130.8 (C-3i, C-5iv), 130.7 (C- 4iv), 130.2 (C-5ii), 128.2 (C-2i), 126.0 (C-6ii), 125.4 (C-4ii), 122.5 (C-2iv), 118.1 (CN), 116.3 (C-2ii), 111.0 (C-3iv), 82.1 (C(CH3)3), 81.1 (C(CH3)3), 79.8 (C(CH3)3), 52.7 (C-2), 42.7 (C-4), 36.2 (C-1i), 34.4 (C-3), 32.0 (C-1i), 31.2 (C-2i), 28.4 (C(CH3)3), 28.1 (2xC(CH3)3); IR (ATR): 2978, 2932, 2232, 1714, 1367, 1148 cm-1; HRMS (HR-ESI) calcd. for [C36H46N4O6+Na]: 653.3310; found: 653.3296. COSY and 1H/13C correlation were recorded. (1S,3R,E)-1,3-dicarboxy-6-((Z)-8-cyano-11,12-dihydrodibenzo[c,g][1,2] diazocin-3- yl)hex-5-en-1-aminium chloride, Glu_brAzo1 To a solution of 8 (45 mg, 0.07 mmol) in CH2Cl2 (11 mL), trifluoroacetic acid (4 mL, 52.2 mmol) was added. After stirring 2 h at rt, the solution was concentrated under vacuum. The resulting solid was washed with Et2O, hexane and EtOAc to furnish Glu_brAzo1 (30 mg, 0.06 mol, 81% yield) as a brown solid: mp 134-139 ºC (from EtOAc); [α]D20 20.8 (c 0.18, DMSO-d6); 1H NMR (360 MHz, DMSO-d6) δ 7.55 (d, J4iii,5iv = 7.8 Hz, 1H, H-4iii), 7.44 (s, 1H, H-2iii), 7.34 (d, J5iii,4iii = 7.8 Hz, 1H, H-5iii), 7.12 (d, J4i,5i = S6 7.9 Hz, 1H, H-4i), 7.04 (d, J5i,4i = 7.9 Hz, 1H, H-5i), 6.88 (s, 1H, H-2i), 6.37 (d, J6,5 = 15.6 Hz, 1H, H-6), 6.15 (m, 1H, H-5), 3.80 (m, 1H, H-1), 2.87 (m, 4H, H-1ii, H-2ii), 2.71 (m, 1H, H-3), 2.50 (m, 2H, H-4), 2.13 (m, 1H, H-2), 1.77 (m, 1H, H-2); 13C NMR (90.5 MHz, DMSO-d6) δ 175.0/170.9 (2xCOOH), 155.1 (C-1iii), 155.0 (C-1i), 135.9 (C-3i), 134.6 (C- 6iii), 131.2 (C-5iii), 131.0 (C-4iii), 130.8 (C-6), 130.4 (C-5i), 127.7 (C-5), 126.3 (C-6i), 125.0 (C-4i), 122.2 (C-2iii), 118.1 (CN), 115.8 (C-2i), 109.6 (C-3iii), 50.9 (C-1), 40.4 (C- 3), 35.0 (C-4), 31.3 (C-2), 30.9 (C-1ii), 30.3 (C-2ii); IR (ATR) 2934, 2231, 1675, 1606, 1243, 1200, 1134 cm-1; HRMS (HR-ESI) calcd. for [C23H23N4O4]+: 419.1714; found: 419.1718. COSY and 1H/13C correlation were recorded. (Z)-8-bromo-11,12-dihydrodibenzo[c,g][1,2]diazocine-3-carboxylic acid, 7a To a solution of nitrile 7 (260 mg, 0.84 mmol) in EtOH (36 mL), a solution of KOH (5.20 g, 92.7 mmol) in H2O (11 mL) was added dropwise. The mixture was heated at reflux temperature for 4 h. Then, the EtOH was removed under vacuum and the resulting aqueous solution was acidified to pH 4. The product was extracted with EtOAc (3 x 50 mL) and the combined organic extracts were dried over anhydrous Na2SO4 and concentrated under vacuum to give 7a (265 mg, 0.80 mmol, 95% yield) as a black solid: mp > 230 ºC (from EtOAc); 1H NMR (400 MHz, acetone-d6) δ 7.72 (dd, J4,5 = 7.9 Hz, J4,2 = 1.7 Hz, 1H, H-4), 7.48 (d, J2,4 = 1.7 Hz, 1H, H-2), 7.24 (d, J5,4 = 7.9 Hz, 1H, H- 5), 7.22 (dd, J4ii,5ii = 8.2 Hz, J4ii,2ii = 2.0 Hz, 1H, H-4ii), 7.11 (d, J2ii,4ii = 2.0 Hz, 1H, H-2ii), 7.06 (d, J5ii,4ii = 8.2 Hz, 1H, H-5ii), 2.94 (m, 4H, H-1i, H-2i); 13C NMR (100.6 MHz, acetone-d6) δ 166.5 (-COOH), 157.5 (C-1ii), 156.2 (C-1), 134.4 (C-6), 132.7 (C-5ii), 131.3 (C-4ii), 130.9 (C-5), 130.3 (C-3), 129.2 (C-4), 128.4 (C-6ii), 122.1 (C-2ii), 120.5 (C-2), 120.5 (C-3ii), 31.9 (C-1i), 31.3 (C-2i); IR (ATR) 2922, 2852, 2538, 1683, 1430, 1296, 1226 cm-1; HRMS (HR-ESI) calcd. for [C15H10BrN2O2+H]+: 331.0077; found: 331.0082. COSY and 1H/13C correlation were recorded. tert-butyl (Z)-4-(8-bromo-11,12-dihydrodibenzo[c,g][1,2]diazocine-3-carboxamido) benzoate, 9 S7 To a solution of acid 7a (50 mg, 0.15 mmol) in dry THF (7 mL), a solution of amine 2 (35 mg, 0.18 mmol), EDCI·HCl (40 mg, 0.21 mmol), HOBt (32 mg, 0.24 mmol) and DIPEA (0.11 mL, 0.63 mmol) in dry THF (3.6 mL) were added. The reaction mixture was stirred at rt overnight. Then, H2O (20 mL) was added and the resulting mixture was extracted with CH2Cl2 (3 x 20 mL). The combined organic extracts were dried over anhydrous Na2SO4, filtered and concentrated under vacuum. The resulting residue was purified by column chromatography (hexane/EtOAc, 3:1) to furnish 9 (60 mg, 0.12 mmol, 78% yield) as a pale yellow solid: mp 226-230 ºC (from EtOAc); 1H NMR (400 MHz, CDCl3) δ 7.98 (d, J2,3 = 8.7 Hz, 2H, H-2), 7.90 (br s, 1H, CONH), 7.67 (d, J3,2 = 8.7 Hz, 2H, H-3), 7.57 (dd, J4i,5i = 8.0 Hz, J4i,2i = 1.9 Hz, 1H, H-4i), 7.37 (d, J2i,4i = 1.9 Hz, 1H, H-2i), 7.16 (dd, J4iii,5iii = 8.2 Hz, J4ii,2ii = 2.0 Hz, 1H, H-4iii), 7.13 (d, J5i,4i = 8.0 Hz, 1H, H-5i), 7.01 (d, J2iii,4iii = 2.0 Hz, 1H, H-2iii), 6.88 (d, J5ii,4ii = 8.2 Hz, 1H, H-5iii), 2.90 (m, 4H, H-1ii, H-2ii), 1.59 (s, 9H, C(CH3)3); 13C NMR (100.6 MHz, CDCl3) δ 165.3 (C, ester), 164.5 (C, amide), 156.1 (C-1iii), 155.4 (C-1i), 141.5 (C-4), 133.6 (C-3i), 132.5 (C-6i), 131.4 (C-5iii), 130.8 (C-2, C-4iii), 130.6 (C-5i), 128.2 (C-1), 126.8 (C-6iii), 126.2 (C-4i), 122.0 (C-2iii), 120.6 (C-3iii), 119.3 (C-3), 118.0 (C-2i), 81.1 (C(CH3)3), 31.7 (C-1ii), 31.1 (C-2ii), 28.4 (C(CH3)3); IR (ATR) 3415, 2965, 1705, 1673, 1521, 1290, 1161, 1103 cm-1; HRMS (HR-ESI) calcd. for [C26H24BrN3O3+H]+: 506.1074; found: 506.1085. COSY and 1H/13C correlation were recorded. di-tert-butyl (2S,4R)-2-((tert-butoxycarbonyl)amino)-4-((E)-3-((Z)-8-((4-(tert-butoxy carbonyl)phenyl)carbamoyl)-11,12-dihydrodibenzo[c,g][1,2]diazocin-3- yl)allyl)pentanedioate, 9a A mixture of diazocine 9 (60 mg, 0.12 mmol), glutamate derivative 1 (57 mg, 0.14 mmol), Pd(OAc)2 (3 mg, 13 mol) and K3PO4 (35 mg, 0.16 mmol) in DMAc (1.0 mL) under argon atmosphere was heated at 140 ºC overnight. Then, the solvent was evaporated and the resulting residue was purified by column chromatography (hexane/EtOAc, 5:1) to give 9a (50 mg, 60 mol, 51% yield) as a yellow foam: 1H NMR (360 MHz, CDCl3) δ 8.10 (s, 1 H, -NHCO), 7.95 (d, J3v,2v = 8.2 Hz, 2H, H-3v), 7.66 (d, J2v,3v = 8.2 Hz, 2H, H-2v), 7.55 (d, J4iv,5iv = 8.0 Hz, 1H, H-4iv), 7.36 (s, 1H, H-2iv), 7.11 (d, J5iv,4iv = 8.0 Hz, 1H, H-5iv), 6.98 (d, J4ii,5ii = 8.0 Hz, 1H, H-4ii), 6.90 (d, J5ii,4ii = 8.0 Hz, 1H, H-5ii), 6.79 (s, 1H, H-2ii), 6.26 (d, J3i,2i = 15.7 Hz, 1H, H-3i), 6.02 (m, 1H, H-2i), 4.99 (d, S8 JNHBoc,2 = 8.5 Hz, 1H, NHBoc), 4.21 (m, 1H, H-2), 2.97 (m, 2H, H-1iii, H-2iii), 2.79 (m, 2H, H-1iii, H-2iii), 2.43 (m, 2H, H-1i, H-4), 2.31 (m, 1H, H-1i), 2.13 (m, 1H, H-3), 1.68 (m, 1H, H-3), 1.58 (s, 9H, C(CH3)3), 1.40 (s, 18H, C(CH3)3), 1.37 (s, 9H, C(CH3)3); 13C NMR (90.6 MHz, CDCl3) δ 173.9 (C-5), 171.7 (C-1), 165.4 (C, ester), 164.7 (C, amide), 155.4 (C, carbamate, C-1ii, C-1iv), 141.7 (C-1v), 133.5 (C-3ii), 133.3 (C-3iv), 132.9 (C-6iv), 131.1 (C-3i), 130.7 (C-3v), 130.5 (C-5iv), 130.1 (C-5ii), 127.9 (C-2i, C-4v), 126.4 (C-6ii), 126.0 (C-4iv), 125.3 (C-4ii), 119.2 (C-2v), 118.0 (C-2iv), 116.5 (C-2ii), 82.1 (C(CH3)), (81.0 (C(CH3)3), 79.8 (C(CH3)3), 52.7 (C-2), 42.7 (C-4), 36.2 (C-1i), 34.4 (C-3), 31.8 (C- 1i), 31.4 (C-2i), 28.4 (C(CH3)3), 28.3 (C(CH3)3), 28.1 (2xC(CH3)3); IR (ATR) 3343, 2983, 2931, 1710, 1528, 1148 cm-1; HRMS (HR-ESI) calcd. for [C47H60N4O9+H]: 825.4433; found: 825.4418. COSY and 1H/13C correlation were recorded. (1S,3R,E)-1,3-dicarboxy-6-((Z)-8-((4-carboxyphenyl)carbamoyl)-11,12-dihydro dibenzo[c,g][1,2]diazocin-3-yl)hex-5-en-1-aminium trifluoroacetate, Glu_brAzo2 To a solution of 9a (50 mg, 61 µmol) in CH2Cl2 (9 mL), trifluoroacetic acid (4.5 mL, 59 mmol) was added. The mixture was stirred for 2 h and the solvent was evaporated under vacuum. The resulting solid was washed with Et2O, hexane and EtOAc to give Glu_brAzo2 (33 mg, 0.049 mmol, 82% yield) as a pale orange solid: mp > 230 ºC (from EtOAc); [α]D20 -6.0 (c 0.25, DMSO-d6); 1H NMR (400 MHz, DMSO-d6) δ 10.50 (s, 1 H, -NHCO), 7.92 (d, J3iv,2iv = 8.7 Hz, 2H, H-3iv), 7.85 (d, J2iv,3iv = 8.7 Hz, 2H, H-2iv), 7.68 (d, J4iii,5iii = 7.9 Hz, 1H, H-4iii), 7.46 (s, 1H, H-2iii), 7.29 (d, J5iii,4iii = 7.9 Hz, 1H, H-5iii), 7.11 (d, J4i,5i = 8.0 Hz, 1H, H-4i), 7.06 (d, J5i,4i = 8.0 Hz, 1H, H-5i), 6.90 (s, 1H, H-2i), 6.34 (d, J6,5 = 15.5 Hz, 1H, H-6), 6.15 (m, 1H, H-5), 3.29 (m, 1H, H-1), 2.86 (m, 4H, H-1ii, H-2ii), 2.68 (m, 1H, H-3), 2.50 (m, 2H, H-4), 2.04 (m, 1H, H-2), 1.65 (m, 1H, H-2); 13C NMR (90.6 MHz, DMSO-d6) δ 172.8/171.8 (2xCOOH), 166.9(COOH), 164.7 (C, amide), 155.1 (C-1iii), 154.6 (C-1i), 143.0 (C-1iv), 135.9 (C-3i), 133.0 (C-3iii), 132.3 (C-6iii), 130.2 (C-5i, C-5iii, C-3iv), 128.2 (C-6), 126.6 (C-5, C-6i), 125.6 (C-4iii, C-4iv), 124.7 (C-4i), 119.5 C-2iv), 118.0 (C-2iii), 116.0 (C-2i), 50.7 (C-1), 40.1 (C-3), 35.5 (C-4), 31.1 (C-2), 30.9 (C- 1ii), 30.5 (C-2ii); IR (ATR) 2923, 1676, 1599, 1521, 1247, 1176 cm-1; HRMS (HR-ESI) calcd. for [C30H29N4O7]: 557.2031; found: 557.2040. COSY and 1H/13C correlation were recorded. S9 Computational, photochemical and biological supplementary methods Molecular docking calculations: Docking calculations were performed with GOLD5.6.33 program on the rigid crystallographic structure of GluK2 Ligand Binding Domain retrieved from the Protein Data Bank (PDB code: 4H8I).4 H++ web-server5 was used to add missing hydrogen atoms to the protein pdb file and to determine the protonation states of ionizable residues at pH = 7.0. All co-crystallized ligands, and water molecules were removed. The structures of trans-Glu_brAzo1 and cis- Glu_brAzo1 were built with standard bond lengths and angles using Gaussview6 and UCSF Chimera7 codes. Those structures were further minimized using Quantum Mechanics at the B3LYP/6-31G(d) level with the Gaussian09 package.8 The trans- Glu_brAzo2 and cis-Glu_brAzo2 initial structures were built from the coordinates of trans-Glu_brAzo1 and cis-Glu_brAzo1 QM minima, respectively, by replacing the cyano substituent by the bulky -CONH-p-C6H4COOH group, and the corresponding QM minima were also located. The four minimized compounds and the protein were further prepared for docking by assigning GOLD atom types to all the atoms by means of Hermes (GOLD 3D Visualizer). To ensure a complete exploration of the conformational space of the four ligands in the binding site of the protein, the maximum efficiency of the genetic algorithm was selected in the docking setup. All the parameters of the genetic algorithm were set to the default values of the program and partial flexibility was assigned to the ligands. The binding site of the receptor was defined as all atoms within 15 Å of a conveniently specified central point (given by its X, Y, Z coordinates). 100 binding solutions for each ligand were generated. The Chemscore fitness function9 was selected to estimate the binding affinities of the four ligands. The obtained poses were ranked based on those affinity docking scores (kJ/mol) and the best solution for each ligand was selected. Photochemical characterization of Glu_brAzo1 and Glu_brAzo2: All spectroscopic and photochemical experiments were carried out in HPLC- or spectroscopy-quality solvents. Steady-state UV-vis absorption measurements were recorded on a HP 8453 spectrophotometer with temperature control. The UV-vis absorption spectra of trans- Glu_brAzo1 and trans-Glu_brAzo2 in PBS:DMSO 1:1 was estimated from the absorption spectra of as-synthesized cis-Glu_brAzo1 and cis-Glu_brAzo2 and of a well-known mixture of their cis and trans isomers, the composition of which was measured by 1H NMR in DMSO-d6. Isomerization quantum yields were determined relative to 1,2-bis(5-chloro-2-methyl-3-thienyl)perfluorocyclopentene in hexane (ring opening = 0.1310). Different excitation sources were used in the photochemical S10 experiments depending on the spectral requirements: a cw DPSS laser (exc = 405 nm, SciTec) for cis-trans photoisomerization, and the second harmonic of a ns-pulsed Nd:YAG laser (exc = 532 nm, Brilliant, Quantel) for trans-cis photoisomerization. Cell line and transfection: HEK293 tsA201 cell line (SV40-transformed, Human Embryonic Kidney 293 cells) was maintained at 37 ºC in a 5% CO2 humid incubator with Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 media (DMEM) (1:1, Invitrogen) supplemented with 10% Fetal Bovine Serum (FBS) and 1% Penicillin/Streptomycin. Cells transiently expressed the receptor subunit GluK2 or GluK1(Q)-2b(GGAA). The expression plasmid for GluK2 was kindly provided by Ehud Y. Isacoff (University of California). GluK2 was subcloned into EcoRI site of the pCDNA3.1 expression vector. To obtain the GluK2-eGFP pCDNA3 construct, a fragment spanning the C-terminal domain of GluK2 and eGFP was amplified by PCR using AfeI/XbaI flanked primers and subcloned into GluK2-pCDNA3 plasmid by replacing the former AfeI/XbaI cassette. GluK1(Q)-2b(GGAA) was kindly provided by G. Swanson (Northwestern University, Feinberg School of Medicine), with the endoplasmic reticulum retention motif of the carboxy-terminal mutated to increase surface expression.11 DNA–X-tremGENE 9 Transfection Reagent (Roche) mix was used following manufacturer’s instructions with a Reagent:DNA ratio of 3:1. GluK1(Q)- 2b-GGAA plasmid was co-transfected with peGFP with a Transfection Reagent:GluK1:eGPF ratio of 3:1:0.1. The mix was incubated for 20 min at room temperature, meanwhile cells were detached and freshly plated into a 12-mulltiwell plate at a density of 3 × 105 cells before the DNA-Transfection Reagent mix was added dropwise into each well. Experiments were performed after 48–72 h, and the day before the experiment cells were plated at low density on 16-mm coverslips (Fisher Scientific) treated with collagen(Sigma-Aldrich) to allow cell adhesion. Electrophysiology: Whole-cell voltage-clamp recordings were done using an EPC-10 amplifier and data at 10 kHz was acquired with amplifier’s software Patch Master (HEKA). Bath solution was composed of (in mM): 140 NaCl, 1 MgCl2, 2.5 KCl, 10 HEPES, 2.5 CaCl2 and 10–20 glucose to fix osmolarity to 310 mOsm·kg–1, while pH 7.42 was adjusted with NaOH. Borosilicate glass pipettes were pulled with a typical resistance of 4–6 MΩ for HEK293 cells. Pipette solution contained (in mM): 120 cesium methanosulfonate, 10 TEA-Cl, 5 MgCl2, 3 Na2ATP, 1 Na2GTP, 20 HEPES, 0.5 EGTA; osmolarity was 290 mOsm·kg–1 and pH 7.2 was adjusted with CsOH. S11 Before starting the recording, cells were incubated 10 min with 0.3 mg mL–1 Concanavalin A (Sigma) —to block GluK1 and GluK2 desensitization— on an ES based on NMDG+ (to avoid depolarization due to open GluRs, in mM): 110 NMDG+, 2.5 KCl, 1 MgCl2, 10 HEPES, 10–20 glucose to fix osmolarity to 300 mOsm·kg–1, while pH 7.4 was adjusted with HCl. Before placing the coverslip to the recording chamber, cells were washed again with bath solution. Light stimulation was done by illumination of the entire focused field using a Polychrome V monochromator (TILL Photonics) connected through the back port of an IX71 inverted microscope (Olympus) with a XLUMPLFLN 20XW x20/1 water immersion objective (Olympus). For automatically controlling wavelength, the monochromator was connected to the EPC-10 amplifier via Photochromic Manual Control (TILL Photonics) and controlled with the photometry module of Patch Master. Light power density measured with a Newport 1916-C light meter after the objective was 22.0 μW mm−2 for 380 nm, 45.9 μW mm−2 for 460 nm and 47.4 μW mm−2 for 500 nm. One-photon action spectrum characterization was done during voltage-clamp recordings applying a train of 1 s light-pulses at different wavelengths (for the whole action spectrum, we ranged wavelengths from 300 to 600 nm, with 10 nm steps) with 5 s delay between pulses in which light was switched to 530 nm to trans-cis back- isomerize for activation spectrum, and to 390nm for deactivation spectrum. Rat hippocampal neural primary culture. All procedures were conducted in accordance with the European guidelines for animal care and use in research, and were approved by the Animal Experimentation Ethics Committee at the University of Barcelona (Spain). Low-density primary hippocampal cultures were prepared from newborn P0-P3 pups from Sprague Dawley rat and maintained in cell culture for 1-2 weeks in coverslips coated with poly-L-lysine (Sigma-Aldrich), as previously described.12-14. Cells were cultured with complete medium (Neurobasal A, B-27 5%, GlutaMax (0.5x), glucose 15mM, penicillin (5 U ml−1) and streptomycin (5 μg ml−1)). Within 48-72 h an anti-mitotic treatment with AraC 5 μM is done to avoid fibroblast and astrocyte proliferation. Half of the medium in each well was changed every 3-4 days. Electrophysiology recording conditions for rat hippocampal neurons: Voltage and current-clamp recordings under whole-cell configuration were done using an EPC- 10 amplifier and data at 10 kHz was acquired with amplifier’s software Patch Master S12 (HEKA). Bath solution was composed of (in mM): 140 NaCl, 2 MgCl, 2.5 KCl, 10 HEPES, 0.2 CaCl2 and 10–20 mM glucose to fix osmolarity to 310 mOsm·kg–1, pH 7.42 adjusted with NaOH. Borosilicate glass pipettes were pulled with a typical resistance of 6–8 MΩ for neurons. Pipette solution contained (in mM): 130 KCl, 5 MgCl2, 3 Na2ATP, 1 Na2GTP, 20 HEPES, 0.5 EGTA. Osmolarity is adjusted at 289 mOsm·kg–1 and pH 7.2 adjusted with KOH. During recordings, neurons were maintained at room temperature (r.t., 25-27 °C) in a continuous perfusion of bath solution. Data analysis and statistics: Amplitude of photocurrents were analyzed using IgorPro (Wavemetrics). Displayed whole-cell current traces have been filtered using the infinite impulse response digital filter from IgorPro (low-pass filter with cutoff of 50 Hz). The drift in current observed during light spectra recordings was corrected where appropriate with the IgorPro (WaveMetrics) software using a custom-made macro for drift correction. S13 Figure S1. Molecular docking simulations Figure S1. Best docking solutions for trans-Glu_brAzo1 (orange) and cis-Glu_brAzo1 (green) in GluK2. The protein residues interacting with the glutamate groups are also indicated. Oxygen, nitrogen and hydrogen atoms are depicted in red, blue and white, respectively. S14 Figures S2-S7 and Table S1. Photoisomerization of Glu_brAzo1 and Glu_brAzo2 Figure S2. Low-field region of the 1H NMR spectra (360 MHz, DMSO-d6) of Glu_brAzo1: (a) freshly prepared (cis isomer); (b) after irradiation at exc = 405 nm (power = 280 mW) for 3 h (cis:trans 1.4:1 mixture); (c) after irradiation of (b) at exc = 532 nm (power = 100 mW) for 3 h (cis isomer). S15 Figure S3. Low-field region of the 1H NMR spectra (400 MHz, DMSO-d6) of Glu_brAzo2: (a) freshly prepared (cis isomer); (b) after irradiation at exc = 405 nm (power = 280 mW) for 3 h (cis:trans 1:1.25 mixture); (c) after irradiation of (b) at exc = 532 nm (power = 100 mW) for 3 h (cis isomer). S16 Figure S4. Absorption spectra of Glu_brAzo1 and Glu_brAzo2 in PBS:DMSO 1:1 upon irradiation of their cis isomer at λexc = 405 nm (power = 88 mW). Glu_brAzo1 Glu_brAzo2 Figure S5. Absorption spectra of Glu_brAzo1 and Glu_brAzo2 in PBS:DMSO 1:1 upon irradiation at λexc = 532 nm (power = 10 mW) of the cis-trans photostationary mixture previously prepared at λexc = 405 nm. S17 Figure S6. Variation of the absorption at 395 and 478 nm of Glu_brAzo1 and Glu_brAzo2 in PBS:DMSO 1:1 upon consecutive cycles of cis-trans (λexc = 405 nm, 280 mW) and trans-cis photoisomerization (λexc = 532 nm, power = 230 mW). Figure S7. Variation of the absorption at λtrans abs,max of the cis-trans photostationary state mixture of Glu_brAzo1 and Glu_brAzo2 in the dark at 25ºC in PBS:DMSO 1:1. At these conditions, thermal trans-cis back-isomerization takes place, thus restoring the initial concentration of the cis state of the ligands. Points correspond to the experimental data, while lines were obtained from monoexponential fits. S18 Table S1. Photochemical properties of Glu_brAzo1 and Glu_brAzo2 in PBS:DMSO 1:1. %trans %cis trans λcis abs,max a λtrans abs,max a cis-trans trans-cis PSScis- PSStrans- (h) (nm) (nm) trans cis Glu_brAzo1 392 478 0.11 0.89 47 100 6.8 Glu_brAzo2 395 480 0.13 0.86 60 100 5.6 a Data for the n-π* band. b t1/2 = 4.7 and 3.9 h, respectively. S19 Figures S8-S11. Photomodulation of GluK1 and GluK2 in HEK293 cells using Glu_brAzo1 and Glu_brAzo2 Figure S8. Light-dependent inward currents measured for HEK293 cells expressing GluK1 and GluK2. (a-b) HEK293 cells transfected with GluK1 and perfused with (a) cis- Glu_brAzo1 (100μM), and (b) cis-Glu_brAzo2 (100μM). (c-d) HEK293 cells transfected with GluK2 and perfused with (c) cis-Glu_brAzo1 (100μM), and (d) cis- Glu_brAzo2 (100μM). Light pulses at λexc = 390 nm (purple bars) were employed to induce cis-trans photoisomerization, which in all the cases resulted in larger electrophysiological currents (i.e. receptor activation and channel opening). Irradiation at λexc = 530 nm (green bars) was used to trigger trans-cis back-isomerization and recover the initial signal (i.e. receptor deactivation and channel closing). Figure S9. Activation spectra of (a) Glu_brAzo1 (100μM) and (b) Glu_brAzo2 (100μM) in HEK293 cells transfected with GluK2. Wavelength excitation to promote cis- trans photoisomerization and receptor activation ranges from 300 nm to 530 nm. In all the cases, trans-cis back-isomerization was achieved with pulses at λexc = 530 nm. S20 Figure S10. Deactivation spectra of (a) Glu_brAzo1 (100μM) and (b) Glu_brAzo2 (100μM) in HEK293 cells transfected with GluK2. Wavelength excitation to promote trans-cis photoisomerization and receptor deactivation ranges from 300 nm to 600 nm. In all the cases, cis-trans isomerization was previously conducted with pulses at λexc = 390 nm. GluK1 cis isomer GluK1 cis isomer GluK1 PSScis-trans@390 nm GluK1 PSScis-trans@390 nm GluK2 cis isomer GluK2 cis isomer GluK2 PSScis-trans@390 nm GluK2 PSScis-trans@390 nm Glu_brAzo1 Glu_brAzo2 Figure S11. Dose-response curves of (a) Glu_brAzo1 and (b) Glu_brAzo2 in HEK293 cells transfected with GluK1 and GluK2. Measurements were conducted from 10 to 300 μM for their cis isomer and the trans-cis mixture obtained at λexc = 390 (PSScis-trans@390 nm). S21 Figures S12. Photoactivation of rat hippocampal neurons using Glu_brAzo1 and Glu_brAzo2 Figure S12. Whole-cell voltage clamp recording of rat hippocampal neurons in culture. Perfusion of 30 μM of (a) Glu_brAzo1 and (b) Glu_brAzo2. Light activation is induced by pulses at λexc = 390 and 530 nm (purple and green bars, respectively). S22 1 H NMR (500 MHz, CDCl3) 13 C NMR (125.8 MHz, CDCl3) S23 1 H NMR (500 MHz, CDCl3) 13 C NMR (125.8 MHz, CDCl3) S24 1 H NMR (500 MHz, CDCl3) 13 C NMR (125.8 MHz, CDCl3) S25 1 H NMR (500 MHz, CDCl3) 13 C NMR (125.8 MHz, CDCl3) S26 1 H NMR (500 MHz, CDCl3) 13 C NMR (125.8 MHz, CDCl3) S27 1 H NMR (360 MHz, DMSO-d6) 13 C NMR (90.5 MHz, DMSO-d6) S28 1 H NMR (400 MHz, acetone-d6) 13 C NMR (100.6 MHz, acetone-d6) S29 1 H NMR (400 MHz, CDCl3) 13 C NMR (100.6 MHz, CDCl3) S30 5iv 1iii 2iii 5ii 6iv 6ii 4ii 4iv H 3ii 2i 1v N iv 1iv ii 1i NHBoc 3 2iv N N1 2ii 3 2v O i tBuO2C 4v tBuO2C 4 2 CO2tBu 3v 5 3 1 9a 1 H NMR (360 MHz, CDCl3) 13 C NMR (90.6 MHz, CDCl3) S31 1 H NMR (400 MHz, DMSO-d6) 13 C NMR (90.6 MHz, DMSO-d6) S32 References [1] Gascón-Moya, M.; Pejoan, A.; Izquierdo-Serra, M.; Pittolo, S.; Cabré, G.; Hernando, J: Alibés, R.; Gorostiza, P.; Busqué, F. J. Org. Chem. 2015, 80, 9915-9925. [2] Venkatachalam, T. K.; Huang, H.; Yu, G.; Uckun, F. M. Synth. Commun. 2004, 34, 1489-1497. [3] Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins Struct. Funct. Biogenet. 2003, 52, 609–623. [4] Reiter, A.; Skerra, A.; Trauner, D.; Schiefner, A. Biochemistry 2013, 52, 8972-8974. [5] Anandakrishnan, R.; Aguilar, B.; Onufriev, A. V. Nucleic Acids Res. 2012, 40, 537- 541. [6] GaussView, Version 6, Dennington, R.; Keith, T. A.; Millam, J. M. Semichem Inc., Shawnee Mission, KS, 2016. [7] Pettersen E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. J. Comput. Chem. 2004, 25, 1605-1612. [8] Gaussian 09, Revision C.01, Frisch, M. J.; Trucks, G. W.; Sch-legel, H. B.; Scuseria, G. E.; Robb, M. A.; et al. Gaussian, Inc., Wallingford CT, 2009. [9] Eldridge, M. D.; Murray, C. W.; Auton, T. R.; Paolini, G. V.; Mee, R. P. J. Comput.- Aided Mol. Des. 1997, 11, 425–445. [10] Higashiguchi, K., Matsuda, K., Asano, Y.; Murakami, A.; Nakamura, S.; Irie, M. Eur. J. Org. Chem. 2005, 91-97. [11] Ren, Z.; Riley, N. J.; Needleman, L. A.; Sanders, J. M.; Swanson, G. T.; Marshall, J. J. Biol. Chem. 2003, 278, 52700–52709. [12] Halliwell, R. F.; Peters, J. A.; Lambert, J. J. Br. J. Pharmacol. 1989, 96, 480-494. [13] Beaudoin, G. M.; Lee, S. H.; Singh, D.; Yuan, Y.; Ng, Y. G.; Reichardt, L. F.; Arikkath, J. Nat. Protoc. 2012, 7, 1741-1754. [14] Kaech, S.; Banker; G. Nat. Protoc. 2006, 1, 2406-2415. S33