This is the accepted version of the journal article: Prischich, Davia; Gomila, Alexandre; Milla-Navarro, Santiago; [et al.]. «Adren- ergic Modulation With Photochromic Ligands». Angewandte Chemie (Inter- national ed. Internet), Vol. 60, Issue 7 (February 2021), p. 3625-3631. DOI 10.1002/anie.202010553 This version is available at https://ddd.uab.cat/record/266067 under the terms of the license Revised manuscript Click here to access/download;Manuscript;Revised Manuscript.pdf Adrenergic modulation with photochromic ligands 1 Davia Prischich1,2, Alexandre M. J. Gomila1,2, Santiago Milla-Navarro3, Gemma Sangüesa4,5, Rebeca 2 Diez-Alarcia6,7, Beatrice Preda1, Carlo Matera1,2, Montserrat Batlle4,5, Laura Ramírez3, Ernest Giralt8,9, 3 4 Jordi Hernando10, Eduard Guasch4,5, J. Javier Meana6,7, Pedro de la Villa3,11, Pau Gorostiza1,2,12,* 5 6 1. Institute for Bioengineering of Catalonia (IBEC), The Barcelona Institute for Science and Technology, Barcelona, Spain 7 2. Centro de Investigación Biomédica en Red – Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Madrid, Spain 8 3. Department of Systems Biology, University of Alcalá (UAH), Madrid, Spain 9 10 4. Institut Clínic Cardiovascular, Hospital Clinic, University of Barcelona (UB), IDIBAPS, Barcelona, Spain 11 5. Centro de Investigación Biomédica en Red – Enfermedades Cardiovasculares (CIBER-CV), Madrid, Spain 12 6. Department of Pharmacology, University of the Basque Country (UPV/EHU), Leioa, Bizkaia, Spain. 13 7. Centro de Investigación Biomédica en Red - Salud Mental (CIBER-SAM), Bilbao, Spain. 14 15 8. Department of Inorganic and Organic Chemistry, University of Barcelona (UB), Barcelona, Spain 16 9. Institute for Research in Biomedicine (IRB), Barcelona Institute for Science and Technology (BIST), Barcelona, Spain 17 10. Departament de Química, Universitat Autònoma de Barcelona (UAB), Cerdanyola del Vallès, Spain 18 11. Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), Madrid, Spain 19 20 12. Catalan Institution for Research and Advanced Studies (ICREA), Barcelona, Spain 21 22 (*) E-mail: pau@icrea.cat 23 24 25 Abstract: Adrenoceptors are ubiquitous and mediate important autonomic functions as well as modulating 26 arousal, cognition and pain on a central level. Understanding these physiological processes and their 27 underlying neural circuits requires manipulating adrenergic neurotransmission with high spatio-temporal 28 29 precision. Here we present a first generation of photochromic ligands (adrenoswitches) obtained via 30 azologization of a class of cyclic amidines related to the known ligand clonidine. Their pharmacology, 31 32 photochromism, bioavailability and lack of toxicity allow for broad biological applications, as demonstrated by 33 controlling locomotion in zebrafish and pupillary responses in mice. 34 35 Introduction 36 37 Adrenergic neurotransmission plays an essential role over the unconscious regulation of most vital functions 38 in the human body. Heart and respiratory rate, digestion, smooth-muscle contraction, gland secretion, 39 40 and pupil diameter are modulated by noradrenaline release from sympathetic fibres projecting from the 41 peripheral nervous system (PNS).[1] The neurotransmitter role is not limited to the PNS, though. Adrenergic 42 43 neurons firing from the locus coeruleus (LC) towards different areas of the central nervous system mediate 44 alertness, responses to acute stress and danger, pain modulation, arousal, sleep-wake cycles, as well as 45 neuroplasticity and cognitive behaviour.[2–4] Despite the physiological relevance of adrenergic 46 47 neurotransmission, tools to precisely modulate its activity and to functionally dissect adrenergic pathways 48 have been scarce. Classical pharmacology, which benefits from high receptor specificity, is inadequate to 49 address cellular processes with high spatio-temporal selectivity. In this regard, light offers a non-invasive and 50 51 versatile mean to precisely control a variety of biological targets. Photocontrol of adrenergic signalling has 52 been shown with optogenetics[5–7] and photo-uncaging.[8] However, while the first technique requires genetic 53 manipulation and can lead to unintended neuroplastic phenomena, the irreversible release of caged ligands 54 55 displays diffusion-limited resolution and is challenging in vivo. Photopharmacology poses an alternative to 56 precisely and reversibly deliver drug activity in space and time by introducing photochromic groups into the 57 58 structure of bioactive compounds.[9,10] Successful examples of photoswitchable drugs include ligands of ion 59 channels, receptors and enzymes, peptides, lipids and nucleic acids.[11–13] However, compounds to photoswitch 60 61 62 63 64 65 adrenergic neurotransmission are still not available. In this study we aimed at developing adrenergic drug-like molecules that can be reversibly controlled with light. 1 2 Results and Discussion 3 4 Among the large database of sympatholytic and sympathomimetic agents, we focused on a class of 5 cyclic amidines structurally related to clonidine and other commercialized drugs targeting α-ARs (Figure 1A). 6 7 All adrenergic agents share a common pharmacophoric motif: a primary or secondary aliphatic amine, 8 protonated at physiological pH, and separated by a short linker from a substituted benzene ring.[14] We 9 postulated that α-adrenergic ligands were suitable candidates for “azologization”[15,16] as the structural 10 11 changes required to obtain photo-responsive derivatives could be introduced without altering essential 12 pharmacophoric elements. In addition, the prospect of obtaining dark-inactive adrenergic azologs was 13 supported by the lack of biological activity of trans-like epoxydic analogues compared to cis-like compounds 14 15 described in the work of Gentili et al.[17] We thus designed a set of putative photoswitchable adrenergic 16 ligands, named “adrenoswitches”, by replacing the short linker with an azo group while constraining the cyclic 17 amidine moiety to the closest structurally related aromatic derivatives. The ortho-dichlorobenzene system, 18 19 common to clonidine and lofexidine, was maintained to favour both adrenergic activity and photochromism. 20 First, halogen substitution increases the lipophilicity of a molecule, thus improving its absorption and its 21 22 permeability to the blood brain barrier or the blood ocular barrier.[18] Moreover, lipophilic substituents at the 23 positions 2 and 6 of the phenyl ring are well tolerated in terms of pharmacological activity by both α1- and α2- 24 ARs.[19] Secondly, the photochromism of ortho-halogenated azobenzenes offers enhanced thermal stability, 25 26 longer photoisomerization wavelengths, and higher isomerization ratios.[20,21] 27 To identify suitable aromatic heterocycles, in our first analogue (adrenoswitch-1, Figure 1B) we substituted 28 29 the imidazoline ring with an imidazole. As phenylazoimidazoles are known to undergo fast cis-trans thermal 30 isomerization (i.e. fast-relaxing photoswitches), in adrenoswitch-2 we sought to reduce the rate of this process 31 32 with an N-methyl imidazole.[22] In adrenoswitch-3 we introduced a N,N-dimethyl imidazolium as a permanent 33 positive charge that might better mimic the electronic properties of the original cyclic amidine in its 34 physiologically protonated form. A similar strategy for the same purpose was using 2-aminothiazole in 35 36 adrenoswitch-4. 37 38 Our library was prepared via a divergent synthetic approach involving a standard azo coupling reaction (Figure 39 1B). Commercially available 2,6-dichloroaniline was converted into the corresponding diazonium salt and 40 reacted under mild basic conditions either with imidazole to provide adrenoswitch-1 or with 2-aminothiazole 41 42 to afford adrenoswitch-4. Adrenoswitch-2 and adrenoswitch-3 were obtained from adrenoswitch-1 through 43 reactions of mono- or di-N-methylation, respectively. 44 45 As these photoswitches have not been described among the reported azoheteroaryl[23], their photochromic 46 behaviours were then investigated. Slow-relaxing adrenoswitch-2 and -3 were characterized by steady-state 47 48 UV-Vis absorption spectroscopy (Figures 1C-F), while transient UV-Vis absorption spectroscopy was used for 49 fast-relaxing adrenoswitch-1 and -4 (Figures 1D-E). Our compounds can be photoisomerized from trans to cis 50 with (ultra)violet light (365–400 nm) and from cis to trans with blue or green light (450–500 nm). As intended 51 52 by design, thermal relaxation rates vary considerably along the series. Measured half-lives spanned from 53 milliseconds (adrenoswitch-4) to seconds (adrenoswitch-1), minutes (adrenoswitch-3), and months 54 (adrenoswitch-2) (Figures 1E-F). 55 56 57 58 59 60 61 62 63 64 65 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 C 22 1.2 0.9 1.2 23 0.3 Adrenoswitch-1 Adrenoswitch-4 Adrenoswitch-2 Adrenoswitch-3 Absorbance Absorbance Absorbance Absorbance 24 trans 25 0.4 trans 0.1 trans 0.3 trans cis 0.4 cis 26 27 200 400 600 200 400 600 200 400 600 200 400 600 28 D Wavelength (nm) Wavelength (nm) Wavelength (nm) Wavelength (nm) 29 30 200 400 600 200 400 600 F 1.4 2.5 10-3Abs 10-3Abs 31 0 0 t1/2  21 d t1/2  15.4 min (37C) (37C) 32 -4 cis -8 cis Abs (365 nm) Abs (365 nm) 33 E 34 0.7 1.5 t1/2  37.0 min 10-3Abs (400 nm) 10-3Abs (370 nm) 0 0 (25C) 35 t1/2  129 d 36 -4 -8 (25C) 37 t1/2  0.267 s t1/2  1.65 ms (25ºC) (25ºC) 38 0.0 0.5 0.0 0.4 0.8 0.000 0.004 0.008 0 1000 2000 0 200 400 39 Time (s) Time (s) Time (h) Time (min) 40 41 42 Figure 1. A) Chemical structures of adrenaline and some of the synthetic adrenergic ligands that inspired this work. B) Divergent 43 chemical synthesis of adrenoswitches 1-4. C) UV-Vis absorption spectra of the trans isomers. For slow-relaxing adrenoswitch-2 and -3, 44 UV-Vis absorption spectra of the cis isomers are also reported. Spectra were extracted from UPLC chromatograms after elution with a 45 mixture of water and acetonitrile supplemented with trifluoroacetic acid. Spectra of cis and trans isomers were normalized at their 46 isosbestic points (see SI – Figures S3A-C). D) Transient absorption spectra of adrenoswitch-1 and -4 at t = 0 upon pulsed excitation at 47 λexc=355 nm in physiological buffer (pH=7.4, 25°C). E) Absorption loss and recovery kinetics of adrenoswitch-1 and adrenoswitch-4, 48 measured at λ=370 and 400 nm respectively, upon irradiation with a single ns laser pulse (t=0) at λexc=355 nm in physiological buffer 49 (pH=7.4, 25°C). Red lines correspond to monoexponential fitting of the experimental data. F) Cis-to-trans thermal relaxation of 50 adrenoswitch-2 and -3 at 25˚C (in black) and 37˚C (in gray) under dark conditions. Data were fitted to a monoexponential decay model 51 52 for cis isomers half-lives determination. 53 54 With all the information to effectively photoswitch our ligands, we assessed their biological activity as a 55 function of illumination. The affinity of the library towards α2-ARs was measured by competitive radioligand 56 binding assay in pre-frontal cortex membranes obtained post mortem from human brains. All the 57 58 adrenoswitches competed in binding to the receptor albeit at weaker affinities than clonidine. Most notably, 59 slow-relaxing adrenoswitch-2 and -3 showed a significant change in binding potency upon UV illumination 60 61 62 63 64 65 (Figure 2A and Figures S4A-B). On the other hand, no difference in binding could be observed for adrenoswitch-1 and -4, possibly because these compounds display fast thermal isomerization and the binding 1 assay might reflect mainly the properties of the trans isomers. Nonetheless, since in the work of Gentili et al. 2 3 trans and cis-like epoxy analogues showed similar binding affinities, but produced different biological 4 activities[17], we decided to further test both slow- and fast-relaxing adrenoswitches. 5 6 Legend A B 7 8 0 Dark UV Aorta relaxation (%) 100 Total binding (%) 9 Adrenoswitch-1 Vehicle 10 11 50 50 Adrenoswitch-2 Clonidine 12 Adrenoswitch-3 UV light 13 0 14 100 Adrenoswitch-4 UV-Vis filter 15 Total 10 -6 10 -5 10 -4 10 -3 10 -2 10 -6 10 -5 10 -4 10 -3 16 Concentration (M) Concentration (M) 17 18 C D E 19 150 150 *** n.s. **** 20 40 Distance (mm) Distance (mm) 21 100 100 22 23 20 24 50 50 25 26 0 0 0 27 10 20 10 20 UV UV-Vis filter 28 Time (min) 29 30 Figure 2. A) Competition binding curves of dark-relaxed (trans, solid lines) and UV-illuminated (cis, dashed lines) adrenergic 31 photoswitches to α2-ARs in human brain pre-frontal cortex tissue. Binding is expressed as percentage of radioligand - [3H]RX821002 - 32 bound to the receptor in the presence of the competitor. Data are means ± SEM of two independent experiments run in duplicates or 33 triplicates. B) Dose-response curves comparing the vasodilatory potencies of dark-adapted (trans, solid lines) and cis-enriched (dashed 34 lines) adrenoswitches administered in the dark or under UV illumination to rat aortic rings where the endothelium had been preserved. 35 36 Vessels were pre-contracted with 10-6 M phenylephrine after treatment with 10-3 M NG-methyl-L-arginine acetate (L-NMMA), a nitric 37 oxide synthase inhibitor. Relaxation is expressed as percentages of the reference contraction induced by PE. Data are means ± SEM 38 (adrenoswitch-1, n=4 (per condition); adrenoswitch-2, (trans) n=6 (cis) n=5; adrenoswitch-3, (trans) n=3 (cis) n=4; adrenoswitch-4, n=2 39 (per condition). C-D) Danio rerio 7 days post-fertilisation larvae swimming distances and trajectories (movements with velocities over 40 6 mm·s-1) after treatment with vehicle (grey wells), 50 µM clonidine (light-blue wells) and 50 µM adrenoswitch-1 (yellow wells). 41 Conditions were simultaneously analysed under 365 nm UV illumination (left panels, cis-enriched adrenoswitch-1) and under a UV-Vis 42 filter (right panels, trans adrenoswitch-1) that only transmits infrared light for movement recording. E) Quantification of the swimming 43 distances shown in C) at t=20 minutes. Data are means ± SEM (n=12 per treatment). Statistical differences between UV- and non-UV- 44 exposed larvae were determined by two-way ANOVA with Tukey’s multiple comparison test (n.s., not significant; ***, p-value<0.01; 45 46 ****, p-value<0.001). 47 48 Since the adrenergic system is a major regulator of the vascular tone, we characterized the adrenergic effects 49 of our compounds by measuring their vascular activity in ex vivo rat aortic rings.[24–26] Besides being generally 50 recognized as a potent central hypotensive agent[27], clonidine is also known to induce vasodilation by acting 51 52 directly on vascular adrenoceptors.[25] This peripheral response involves stimulation of endothelial α2-ARs as 53 well as partial agonism towards α1-ARs of the vascular smooth muscle, and results in relaxation of 54 endothelium-intact strips previously contracted with phenylephrine (PE).[26,28–30] When testing if our analogues 55 displayed a similar behaviour, we observed that exposure to UV light alone could induce relaxation of the 56 57 vascular smooth muscle, substantially abolishing the contraction evoked by PE.[31] These intrinsic 58 photoresponses are due to nitric oxide (NO) release either from endogenous or exogenous donors 59 or from storage vesicles[32,33]. To minimize unintended photorelaxation, we pre-incubated the vessels in 60 61 62 63 64 65 a suitable nitric oxide synthase (NOS) inhibitor, namely NG-methyl-L-arginine acetate (L-NMMA) which is reported to be unaffected by light[34]. Remarkably, all adrenoswitches displayed high vasodilatory efficacies 1 (maximal evoked response Emax, Figure 2B) with adrenoswitch-1 coming through as the most potent 2 compound and the only one displaying a light-dependent behaviour in this assay (3x increase in potency upon 3 4 UV illumination, Figure S8A). Although this assay was useful to identify and rank the adrenergic efficacy and 5 potency of adrenoswitches, it had important shortcomings. Intrinsic photoresponses forced us to use a NOS 6 inhibitor, thus partially blocking endothelium-dependent relaxation responses to α2-AR agonists[26,35]. 7 8 Moreover, the model proved complex to clarify adrenoswitches vascular pharmacodynamics as the observed 9 antihypertensive responses might involve both α1 and α2-ARs and do not exclude further contributions from 10 other non-adrenergic targets, most notably imidazoline receptors.[36–38] 11 12 We next characterized the activity of adrenoswitches in vivo. As proof-of-concept applications, we used the 13 most promising compound of the series, adrenoswitch-1, to remotely manipulate locomotion in zebrafish and 14 15 pupillary responses in mice. 16 17 Zebrafish (Danio rerio) is a well-established animal model for both neurobehavioural[39] and toxicological 18 studies[40], and the transparency of their larvae makes them convenient for photopharmacological 19 applications.[41,42] Besides, the expression and function of zebrafish ARs have been studied[43] and the effect of 20 21 dexmedetomidine, a central-acting and structurally-related α2-AR agonist, has been described by Ruuskanen 22 et al.[44] In analogy to the sedative effect reported for dexmedetomidine, clonidine (50 µM) caused locomotor 23 24 inhibition when compared to untreated animals (Figure 2C-D). Clonidine effect was not significantly altered 25 by UV light, although in vehicle-treated animals we observed an increase of the swimming distances as a result 26 of their photomotor response to illumination.[39] In contrast, larvae treated with adrenoswitch-1 (50 µM) 27 28 responded to UV light by progressively reducing their locomotion to levels comparable to clonidine-treated 29 animals under UV (Figures 2D-E and Figures S9-S10). Therefore, adrenoswitch-1 displayed a clonidine-like 30 behaviour upon UV-activation, while its trans isomer was unable to evoke sedation. Although this response is 31 32 unequivocally light-induced, thermal relaxation of the compound was not enough to revert the behavioral 33 effect (Figure 1E and Figure S10). Since other non-adrenergic receptors might evoke locomotion responses, 34 we further studied adrenoswitch-1 in another in vivo model. 35 36 Among the available animal models for assessing autonomic responses, we sought one in which adrenoceptors 37 38 played a major physiological role and in which the use of photoswitchable drugs may have actual 39 investigational and therapeutic implications. Modulating pupil responses fulfilled both criteria and it seemed 40 particularly attractive as the target tissue is naturally accessible to light. In addition, pupillometry allows for a 41 42 selective evaluation of α-adrenergic activity and it excludes the involvement of imidazoline receptors.[45–47] As 43 pupils adapt their diameter to changes in light intensity (pupillary light reflex, PLR), we resorted to genetically 44 45 engineered blind mice (Opn4xRd10) for a first assessment of the compound activity. Opn4xRd10 mice do not 46 express melanopsin and their rods and cones degenerate two months after their birth, thus becoming 47 physiologically insensitive to any luminous stimuli. We tested if adrenoswitch-1 enabled photoregulation of 48 49 pupil diameter after topical administration (1 mM, 0.02% w/v) to isoflurane-anesthetized animals. The 50 compound exerted mydriasis only under UV illumination and the effect fully reversed in the dark, when 51 adrenoswitch-1 rapidly relaxes to the trans configuration (Figures 3A-B-C and Supplementary Movie S1). 52 53 These pupillary responses were reproducible in at least two cycles of alternating UV light and darkness, and 54 were neither elicited by application of the vehicle nor by exposure to UV light alone. The maximum 55 photoresponses were consistently observed after approximately 20 min from its administration (Figure 3C), 56 57 in agreement with the pharmacodynamics of adrenergic ligands with similar structure and lipophilicity.[48] As 58 mydriasis is mediated by postsynaptic α1-adrenoceptors of the iris dilator muscle, the results of Figures 3A-B- 59 60 C unambiguously demonstrate that adrenoswich-1 modulates endogenous ARs with light. Taking into account 61 62 63 64 65 its molecular structure and the pharmacological characterization, we consider that the compound is likely to act as a partial agonist towards α1-ARs. Nonetheless, a thorough characterization of adrenoswitches’ 1 pharmacological profiles would be needed to evaluate their specificity and selectivity over adrenergic and 2 3 non-adrenergic receptors. 4 5 A 6 7 8 9 10 11 12 B C 13 1.5 1.5 * 14 Legend  Pupil area (mm2) Pupil area (mm2) 15 1.0 1.0 16 Adrenoswitch-1 UV light 17 Vehicle Visible light 18 0.5 0.5 19 Clonidine UV-Vis filter 20 0.0 0.0 21 10 30 50 22 Time (min) 23 24 D E F 1.5 0.0 25  Pupil area (mm2)  Pupil area (mm2) 0.0 Pupil area (mm2) 26 1.0 27 -0.2 28 -0.2 29 0.5 30 -0.4 31 *** n.s. 0.0 -0.4 32 20 40 60 0 200 400 33 Time (min) Time (s) 34 35 Figure 3. A-B) Pupillary responses in an isoflurane-anesthetized Opn4xRd10 blind mouse first treated with vehicle and then 36 administered 1 mM adrenoswitch-1 in its right eye. Adrenoswitch-1 exerted mydriasis only upon UV illumination. Neither the vehicle 37 nor UV light alone elicited any pupillary responses in the animals. A complete movie of the experiment (100x speed) is available as SI 38 39 (Supplementary Movie S1). C) Change in area of vehicle vs. adrenoswitch-1 treated pupils of Opn4xRd10 blind mice at t=20 minutes 40 from administration of the drug under UV illumination. Data are means ± SEM (n=4 per condition). Statistical differences were 41 determined by Student’s t test for unpaired observations. D) Pupillary responses in an isoflurane-anesthetized C57BL/6J wildtype 42 mouse first treated with vehicle and then administered 1 mM adrenoswitch-1 in its right eye. The animal was exposed to 5-min cycles 43 of UV, darkness and green light. UV and green light elicited a pupillary light reflex (PLR). Adrenoswitch-1 abolished the PLR only under 44 concomitant UV illumination, while the response was maintained under visible light. E) Comparison of UV and green light evoked PLR 45 in adrenoswitch-1-treated animals. Data are means ± SEM (n=2 per condition). F) PLR quantification in anesthetized wildtype mice. 46 47 Pupil contraction evoked by exposure to UV and green light was compared between vehicle- and adrenoswitch-1-treated animals. 48 Data are mean ± SEM (n=2 per condition). Statistical differences were determined by Student’s t test for unpaired observations. 49 50 Finally, we tested if adrenoswitch-1 could be used in wildtype animals to selectively decouple pupil tone from 51 enviromental illumination. As shown in Figures 3D-F, vehicle-treated C57BL/6J mice respond to violet and 52 green light exposure with similar rapid PLRs. After topical administration of adrenoswitch-1, the miosis caused 53 54 by UV light was blocked whereas the response to green light was maintained (Figures 3D-E-F). As the UV 55 luminous stimulus is triggering a miotic reflex while simultaneously activating a photo-regulated mydriatic 56 agent (adrenoswitch-1), the two effects appear to cancel each other out. These pupillary responses were 57 58 reproducible in all the cycles of alternating UV, darkness and green light and persisted after several washings 59 of the cornea. 60 61 62 63 64 65 In contrast to anticholinergic mydriatics, the pharmacological action of adrenoswitch-1 can be selectively controlled using specific wavelengths, can be fully reversed on demand, and does not alter the pupil basal 1 tone. These unique properties might allow deeper insights into the role of PLR in vision adaptation dynamics. 2 3 Moreover, as retinal responses are closely related to both area and intensity of the visual stimulus, 4 adrenoswitches might become unique tools to investigate photoresponse kinetics and their mechanisms in 5 retinal ganglion cells.[49] 6 7 On-demand regulation of pupil diameter might also have opthalmic applications. For exemple, slow-relaxing 8 9 pre-illuminated adrenoswitches could induce a mydriatic response that spontaneously reverses upon 10 progressive exposure of the eye to natural light. Procedures like examination of the retina or other structures 11 of the eye’s fundus require fully dilated pupils. Although mydriasis can be readily achieved with topic 12 13 cholinergic antagonists, post-exam pupils’ dilation lasts hours and causes photophobia, increased intra-ocular 14 pressure and blurry vision. As these side effects are highly impairing, patients would benefit from drugs whose 15 16 mydriatic effect can be rapidly reverted. 17 Conclusion 18 19 In summary, we have rationally designed novel arylazoheteroarene photoswitches and characterised their 20 21 adrenergic action in vitro and in vivo, including the control of locomotor activity in zebrafish larvae and pupil 22 responses in mice. These results enable multiple applications that were previously unaccessible for adrenergic 23 neurotransmission and open the way to achieve optimized photopharmacological properties based on this 24 25 scaffold. On-demand modulation at specific locations might allow to single out adrenergic projections from 26 the LC and to dissect their function in circuits and behaviour with high resolution.[50,51] Moreover, 27 adrenoswitches can enable photocontrol over a variety of physiological functions, like pupillary responses, 28 29 and help unravel their physiology in vivo. 30 31 Acknowledgements: 32 Mass spectrometry was performed at the IRB Barcelona Mass Spectrometry Core Facility, which actively 33 34 participates in the BMBS European COST Action BM 1403 and is a member of Proteored, PRB2-ISCIII, supported 35 by grant PRB2 (IPT13/0001 − ISCIIISGEFI/FEDER). This research received funding from the European Union 36 37 Research and Innovation Programme Horizon 2020 (Human Brain Project SGA2 Grant Agreement 785907, 38 WaveScalES), European Research ERA-Net SynBio programme (Modulightor project), Agency for Management 39 of University and Research Grants/Generalitat de Catalunya (CERCA Programme; 2017-SGR-1442 and 2017- 40 41 SGR-00465 projects; RIS3CAT plan), Fonds Européen de Développement Économique et Régional (FEDER) 42 funds, Ministry of Economy and Competitiveness (Grant CTQ2016-80066-R), Institute of Health Carlos III 43 (IP18/00754), Fundaluce and “la Caixa” foundations (ID 100010434, agreement LCF/PR/HR19/52160010) and 44 45 Basque Government (IT-1211-19). D.P. was supported by fellowship BES-2015-072657. A.M.J.G. was 46 supported by fellowship BES-2017-083025. 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 [1] A. W. Tank, D. L. Wong, Compr. Physiol. 2015, 5, 1–15. [2] E. Samuels, E. Szabadi, Curr. Neuropharmacol. 2008, 6, 254–285. 1 2 [3] M. Llorca-Torralba, G. Borges, F. Neto, J. A. Mico, E. Berrocoso, Neuroscience 2016, 338, 93–113. 3 4 [4] S. J. Sara, Nat. Rev. Neurosci. 2009, 10, 211–223. 5 [5] J. M. Kim, J. Hwa, P. Garriga, P. J. Reeves, U. L. RajBhandary, H. G. Khorana, Biochemistry 2005, 44, 6 7 2284–2292. 8 [6] P. Makowka, T. Bruegmann, V. Dusend, D. Malan, T. Beiert, M. Hesse, B. K. Fleischmann, P. Sasse, 9 10 Nat. Commun. 2019, 10, 1281. 11 [7] E. R. Siuda, J. G. McCall, R. Al-Hasani, G. Shin, S. Il Park, M. J. Schmidt, S. L. Anderson, W. J. Planer, J. 12 A. Rogers, M. R. Bruchas, Nat. Commun. 2015, 6, 8480. 13 14 [8] S. Muralidharan, G. M. Maher, W. A. Boyle, J. M. Nerbonne, Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 15 5199–5203. 16 17 [9] W. A. Velema, W. Szymanski, B. L. Feringa, J. Am. Chem. Soc. 2014, 136, 2178–2191. 18 19 [10] M. J. Fuchter, J. Med. Chem. 2020, DOI 10.1021/acs.jmedchem.0c00629. 20 21 [11] K. Hüll, J. Morstein, D. Trauner, Chem. Rev. 2018, 118, 10710–10747. 22 [12] M. M. Lerch, M. J. Hansen, G. M. van Dam, W. Szymanski, B. L. Feringa, Angew. Chemie - Int. Ed. 2016, 23 55, 10978–10999. 24 25 [13] P. Paoletti, G. C. R. Ellis-Davies, A. Mourot, Nat. Rev. Neurosci. 2019, 20, 514–532. 26 27 [14] R. K. Griffith, in Foye’s Princ. Med. Chem. (Eds.: T.L. Lemke, D.A. Williams), Lippincott Williams & 28 Wilkins, 2013, pp. 348–356. 29 30 [15] S. Pittolo, X. Gómez-Santacana, K. Eckelt, X. Rovira, J. Dalton, C. Goudet, J. P. Pin, A. Llobet, J. Giraldo, 31 A. Llebaria, P. Gorostiza, Nat. Chem. Biol. 2014, 10, 813–815. 32 33 [16] M. Schoenberger, A. Damijonaitis, Z. Zhang, D. Nagel, D. Trauner, ACS Chem. Neurosci. 2014, 5, 514– 34 518. 35 36 [17] F. Gentili, F. Ghelfi, M. Giannella, A. Piergentili, M. Pigini, W. Quaglia, C. Vesprini, P. A. Crassous, H. 37 Paris, A. Carrieri, J. Med. Chem. 2004, 47, 6160–6173. 38 [18] G. Thomas, in Med. Chem. An Introd., Wiley, 2008, pp. 83–84. 39 40 [19] A. J. Nichols, R. R. Ruffolo, in Alpha-Adrenoceptors Mol. Biol. Biochem. Pharmacol., 1991, pp. 75–114. 41 42 [20] D. Bléger, J. Schwarz, A. M. Brouwer, S. Hecht, J. Am. Chem. Soc. 2012, 134, 20597–20600. 43 44 [21] J. Calbo, A. R. Thawani, R. S. L. Gibson, A. J. P. White, M. J. Fuchter, Beilstein J. Org. Chem. 2019, 15, 45 2753–2764. 46 47 [22] J. Otsuki, K. Suwa, K. K. Sarker, C. Sinha, J. Phys. Chem. A 2007, 111, 1403–1409. 48 [23] S. Crespi, N. A. Simeth, B. König, Nat. Rev. Chem. 2019, 3, 133–146. 49 50 [24] B. Civantos Calzada, A. Aleixandre De Artiñano, Pharmacol. Res. 2001, 44, 195–208. 51 52 [25] S. Guimarães, D. Moura, Pharmacol. Rev. 2001, 53, 319–356. 53 [26] P. M. Vanhoutte, J. Cardiovasc. Pharmacol. 2001, 38, 796–808. 54 55 [27] L. Isaac, J. Cardiovasc. Pharmacol. 1980, 2, 5–20. 56 57 [28] E. G. Silva, T. Feres, L. M. Vianna, P. Okuyama, T. B. Paiva, J. Pharmacol. Exp. Ther. 1996, 277, 872– 58 876. 59 60 [29] S. Iwanaga, O. Shibata, A. Tsuda, S. Hashimoto, T. Makita, S. Cho, K. Sumikawa, Res Commun Mol 61 62 63 64 65 Pathol Pharmacol. 1998, 102, 137–47. [30] E. F. Castillo, C. S. Ortíz, R. M. López, A. Ruíz, J. M. Vélez, C. Castillo, Fundam. Clin. Pharmacol. 2006, 1 2 20, 339–349. 3 [31] R. F. Furchgott, S. J. Ehrreich, E. Greenblatt, J. Gen. Physiol. 1961, 44, 499–519. 4 5 [32] K. L. Andrews, J. J. McGuire, C. R. Triggle, Br. J. Pharmacol. 2003, 138, 932–940. 6 7 [33] F. W. Flitney, I. L. Megson, J. Physiol. 2003, 550, 819–828. 8 [34] J. A. Bauer, H. Fung, Pharmacol. Lett. 1993, 54, 1–4. 9 10 [35] J. C. Molin, L. M. Bendhack, Vascul. Pharmacol. 2004, 42, 1–6. 11 12 [36] S. Regunathan, C. Youngson, H. Wang, D. J. Reis, Ann. New York Acad. Sci. 1995, 12, 580–90. 13 14 [37] G. J. Molderings, M. Göthert, Gen. Pharmacol. 1999, 32, 17–22. 15 [38] M. F. Chen, J. T. Tsai, L. J. Chen, T. P. Wu, J. J. Yang, L. Te Yin, Y. L. Yang, T. A. Chiang, H. L. Lu, M. C. 16 Wu, Biomed Res. Int. 2014, 2014, 1–7. 17 18 [39] D. Kokel, J. Bryan, C. Laggner, R. White, C. Y. J, R. Mateus, D. Healey, S. Kim, A. A. Werdich, J. Stephen, 19 C. A. Macrae, B. Shoichet, R. T. Peterson, Nat Chem Biol 2010, 6, 231–237. 20 21 [40] A. L. Rubinstein, Expert Opin. Drug Metab. Toxicol. 2006, 2, 231–240. 22 23 [41] C. Matera, A. M. J. Gomila, N. Camarero, M. Libergoli, C. Soler, P. Gorostiza, J. Am. Chem. Soc. 2018, 24 140, 15764–15773. 25 26 [42] K. Rustler, A. Gomila, G. Maleeva, P. Gorostiza, P. Bregestovski, B. König, Chem. – A Eur. J. 2020, 26, in 27 press. 28 29 [43] J. O. Ruuskanen, J. Laurila, H. Xhaard, V. V. Rantanen, K. Vuoriluoto, S. Wurster, A. Marjamäki, M. 30 Vainio, M. S. Johnson, M. Scheinin, Br. J. Pharmacol. 2005, 144, 165–177. 31 [44] J. O. Ruuskanen, N. Peitsaro, J. V. M. Kaslin, P. Panula, M. Scheinin, J. Neurochem. 2005, 94, 1559– 32 33 1569. 34 [45] J. Raczak-Gutknecht, T. Frackowiak, A. Nasal, R. Kaliszan, Pharmacol. Reports 2013, 65, 305–312. 35 36 [46] Y. Yu, M. C. Koss, Auton. Neurosci. Basic Clin. 2005, 117, 17–24. 37 38 [47] H. Ishikawa, D. D. Miller, N. Patil, Naunyn. Schmiedebergs. Arch. Pharmacol. 1996, 354, 765–772. 39 40 [48] H. C. A. Innemee, A. de Jonge, J. C. A. van Meel, P. B. M. W. M. Timmermans, P. A. van Zwieten, 41 NaunynSchmiedebergs Arch. Pharmacol. 1981, 316, 294–298. 42 [49] K. Y. Wong, F. A. Dunn, D. M. Berson, Neuron 2005, 48, 1001–1010. 43 44 [50] S. Pittolo, H. Lee, A. Lladó, S. Tosi, M. Bosch, L. Bardia, X. Gómez-Santacana, A. Llebaria, E. Soriano, J. 45 Colombelli, K. E. Poskanzer, G. Perea, P. Gorostiza, Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 13680– 46 47 13689. 48 [51] G. Cabré, A. Garrido-Charles, M. Moreno, M. Bosch, M. Porta-de-la-Riva, M. Krieg, M. Gascón-Moya, 49 N. Camarero, R. Gelabert, J. M. Lluch, F. Busqué, J. Hernando, P. Gorostiza, R. Alibés, Nat. Commun. 50 51 2019, 10, 907. 52 53 54 55 56 57 58 59 60 61 62 63 64 65 Table of Contents Text and Artwork 1 2 3 Adrenoswitches 4 5 UV 6 7 8 Δ 9 10 11 12 13 14 15 16 17 18 19 Adrenoswitches, a class of novel arylazoheteroarene photoswitches, were developed to achieve on-demand 20 modulation of adrenergic neurotransmission by means of illumination at specific wavelenghts. Their versatility 21 to photocontrol biological systems is demonstrated in vivo employing two different animal models, zebrafish 22 and mice. 23 24 Keywords: Adrenergic – Azo compounds – Biological activity – Neurotransmitters – Photochromism 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65