Recent Developments in the Optical Control of Adrenergic Signaling

ABSTRACT

Adrenoceptors (ARs) play a vital role in various physiological processes and are key therapeutic targets. The advent of optical control techniques, including optogenetics and photopharmacology, offers the potential to modulate AR signaling with precise temporal and spatial resolution. In this review, we summarize the latest advancements in the optical control of AR signaling, encompassing optogenetics, photocaged compounds, and photoswitchable compounds. We also discuss the limitations of current tools and provide an outlook on the next generation of optogenetic and photopharmacological tools. These emerging optical technologies not only enhance our understanding of AR signaling but also pave the way for potential therapeutic developments.

1. Introduction

G protein‐coupled receptors (GPCRs), the largest family of human membrane proteins, play a critical role in drug discovery. Approximately 36% of marketed medicines approved by the US Food and Drug Administration (FDA) target GPCRs, the most valuable class of druggable targets, with total sales of about $917 billion for 2011–2015 [1, 2, 3]. As a superfamily of membrane proteins, GPCRs are encoded by over 800 genes in the human genome and consist of six subfamilies: Rhodopsin‐like receptors (Class A), Secretin receptors (Class B1), Adhesion receptor (Class B2), Metabotropic/Glutamate receptors (Class C), Frizzled/smoothened receptors (Class F), and Taste 2 receptors (Class T) [4]. GPCRs regulate broad physiological functions like vision, smell, taste, immune response, brain and cardiovascular functions [5].

Adrenoceptors (ARs), belonging to the class A GPCRs, are activated by the endogenous catecholamines adrenaline and noradrenaline, also known as epinephrine and norepinephrine, respectively (Figure 1A). Notably, there are three subgroups of ARs, α_1, α₂, and β receptors, each comprising of three structurally different subtypes (Figure 1B) [8]. The α₁‐ARs are primarily coupled to heterotrimeric G_q/11 proteins, leading to the hydrolysis of phosphatidylinositol 4,5‐bisphosphate (PIP2) and ultimately increasing intracellular Ca²⁺ levels and protein kinase C (PKC) activity. The α₂‐ARs subgroup is Gα_i‐coupled and inhibits the activation of adenylyl cyclase (AC) resulting in a decrease in cellular cAMP levels. β‐ARs are mainly coupled with Gα_s, which prompt the activation of AC and increase the cellular production of cAMP, leading to further downstream signaling via activation of protein kinase A (PKA). In addition to Gα_s, β‐ARs can also couple to some extent to Gα_i proteins [9, 10, 11]. Next to signaling via the G‐protein pathways, ARs are phosphorylated by G protein receptor kinases (GRKs) resulting in the recruitment of β‐arrestins, causing desensitization of G‐protein signaling pathways. Interestingly, β₃‐AR (ADRB3) is resistant to phosphorylation‐mediated desensitization due to the lack of GRK phosphorylation sites [12, 13, 14].

Figure 1.

Classification of GPCRs and the crystal structure of β₂‐AR. (A) The phylogenetic tree (modified from Hauser et al. [4]) of GPCR‐targeted drugs with the receptors in UniProt names. Approved drugs with targeted receptor families are indicated using red circles and drugs in clinical trials with targeted receptor families are indicated in green circles. There are 117 approved drugs and 41 drugs in clinical trials targeting ARs. (B) The phylogenetic tree of ARs. The data are collected from GPCRdb (https://gpcrdb.org/) and visualized using iTOL [6]. (C) The crystal structure of human β₂‐AR (PDB code: 2RH1). β₂‐AR was crystallized with carazolol bound (inverse agonist, green) [7]. The seven transmembrane helices are indicated by the gray numbers. Source: Figure is generated using PyMOL Molecular Graphics System, Version 2.5.4 Schroödinger LLC (https://pymol.org). [Color figure can be viewed at wileyonlinelibrary.com]

ARs share common GPCR structural features, including seven transmembrane (7TM) α‐helices linked by three intracellular loops and extracellular loops [4]. β₂‐AR (ADRB2) was the first hormone‐activated GPCR whose crystal structure (PDB: 2R4R, 2RH1, antagonist carazolol‐bound structure) was solved using X‐ray crystallography in 2007 (Figure 1C) [7, 15]. This β₂‐AR crystal structure provided an important template for the overall 7TM structure of nonrhodopsin GPCRs and enabled the identification of new β₂‐AR ligands through in silico screening [16]. The crystal structure of turkey β₁‐AR (ADRB1, PDB: 2VT4) bound with cyanopindolol was solved 1 year later [17]. Over the next 10 years, the structures for α_1A‐AR (ADA1A, PDB: 7YM8, 7YMH and 7YMJ), α_1B‐AR (ADA1B, PDB: 7B6W), α_2A‐AR (ADA2A, PDB: 6KUX and 6KUY), α_2B‐AR (ADA2B, PDB: 6K41 and 6K42), α_2C‐AR (ADA2C, PDB:6KUW), and β₃‐AR (PDB: 7DH5) have also been elucidated, offering new ways forward for structure–function and structure‐based drug discovery studies [18, 19, 20, 21, 22, 23].

ARs are expressed in a broad variety of cell types, with each subtype primarily located in specific tissues. α₁‐ARs contribute to regulating the cardiovascular system, the genitourinary system and the brain. α₂‐ARs are present in the blood vessels and also play an important role in the central nervous system (CNS). β‐ARs are widely distributed throughout peripheral tissues and the CNS, playing a crucial role in regulating the well‐known “fight or flight” response. Activation of β‐ARs enhances cardiac output, modulates blood vessel function, and influences essential organs, including skeletal muscle, to prepare the body for rapid action [9, 24, 25, 26, 27, 28].

Along with the discovery of ARs, synthetic ligands have been continuously developed to selectively modulate AR physiological signaling. The first selective β‐ARs antagonist (in fact a weak partial agonist), dichloroisoproterenol, was developed by Powell and colleagues in 1958 [29]. In 1964, Black and colleagues developed another efficient β‐blocker, propranolol, which is still used clinically today [30]. Since then, synthetic agonists, partial agonists, inverse agonists and antagonists with different degrees of ARs subtype selectivity have been developed [31, 32, 33, 34].

As important therapeutic targets (Figure 1A), many FDA‐approved drugs acting on ARs are currently used in the clinic to treat a variety of diseases (Figure 2). Agonists of α₁‐ARs work as vasoconstrictors and antagonists are used to treat hypertension and benign prostatic hyperplasia (BPH). α₂‐AR agonists are mainly used for diseases in the CNS whereas the clinical use of α₂‐AR antagonists is limited. β₂‐ARs agonists are mainly used for diseases in the respiratory tract and β‐AR antagonists are the first‐line treatment for arrhythmias, next to their use in, for example, glaucoma, anxiety, and hypertension (Figure 2) [9]. Unfortunately, adverse effects can arise along with desired therapeutic outcomes due to ubiquitous target expression or, for example, overdosing [43, 44].

Figure 2.

Major clinical use of FDA‐approved AR therapeutics. The use of AR drugs is indicated in the central nervous system, cardiovascular system, visual system, respiratory tract, and urinary system [35, 36, 37, 38, 39, 40, 41, 42]. [Color figure can be viewed at wileyonlinelibrary.com]

Therapeutic AR targeting can potentially be more effective if modulated with precise spatial and temporal control to avoid on‐target side effects by ARs that are expressed elsewhere in the body. The recent development of new innovative approaches toward GPCR modulation with optical tools promises superior spatiotemporal resolution by local illumination of the desired tissue, which can be applied with precise control of duration and intensity [45, 46]. These light‐based technologies to modulate biology are mainly classified into three subgroups: (1) optogenetics, (2) freely diffusible photopharmacology, and (3) tethered photopharmacology (also known as optogenetic photopharmacology) [47, 48]. These technologies potentially overcome some of the issues associated with conventional pharmacology, such as drug toxicity and resistance [49, 50]. The optogenetic strategy allows light‐control of cell function by genetically introducing engineered light‐sensitive target proteins (Figure 3A–D). In contrast, photopharmacology (also known as optopharmacology) uses light‐regulated, drug‐like molecules to control (unmodified) target proteins and consequently cell function (Figure 3E,F) [51, 52]. In general, two strategies are applied in this field: freely diffusible photopharmacology and tethered photopharmacology. Tethered photopharmacology utilizes a bioconjugation reaction to covalently bind light‐sensitive ligands to their target protein [48]. A well‐known natural tethered photopharmacological ligand is retinal, which is covalently tethered with rhodopsin, leading to the activation of rhodopsin upon exposure to light (see Section 2.1) [53]. Although some tethered photopharmacology examples are known in the GPCR field (e.g., for dopamine receptors and metabotropic glutamate receptors [54, 55]), this strategy has not been applied to ARs so far. Therefore, tethered photopharmacology will not be discussed in this review. The fundamental application and introduction for optical control of GPCRs have been comprehensively reviewed elsewhere [47, 48]. Here, we summarize the optical AR tools based on optogenetics and freely diffusible photopharmacology that have been developed to spatially and temporally control AR signaling over the past 30 years and provide a perspective on the next generation of optical AR tools.

Figure 3.

Optical control strategies for ARs. (A) Opto‐AR, a chimeric receptor by fusing rhodopsin with the intracellular part of adrenergic receptor to gain optical control of AR signaling. (B) OptoNb60 with photodimerization protein nMagHigh1 and pMagHigh1 to control AR signaling. (C) Optical modulation of β‐arrestin2–β₂‐AR complex formation using dimerization protein CIB–CRY. (D) OptoNb60 with photocleavage protein PhoCl to control AR signaling. (E) Photocaged compounds irreversibly release active compounds upon illumination. (F) The configuration of photoswitchable compounds can be reversibly switched by illumination with light of appropriate wavelengths. The different configurations can have different interactions with its GPCR target. [Color figure can be viewed at wileyonlinelibrary.com]

2. Optogenetics

Optogenetics, first introduced by Deisseroth and colleagues in 2006, is a technology in which cells are genetically modified to express light‐sensitive proteins that control signaling pathways following illumination [56]. This technique has been proven very powerful and was in 2010 selected by Nature Methods as Method of the Year across all fields of science and engineering [57]. Multiple photoreceptor proteins from nature, such as rhodopsin and channelrhodopsins (ChRs), have been successfully applied as optogenetic tools in the field of GPCRs [58, 59]. Herein, we summarize the optogenetic tools to investigate and modulate AR signaling, which has so far mainly been applied in the field of neuroscience.

2.1. Opto‐ARs

An effective optogenetic strategy for the optical modulation of AR signaling is the Opto‐AR system (Figure 3A), which employs engineered chimeric proteins based on the shared 7TM structure of ARs and rhodopsin. The GPCR rhodopsin is expressed in the eye and involved in dim light vision by having light‐sensitive retinal covalently bound to its TM7, which upon illumination photoisomerizes from 11‐cis‐retinal to all‐trans‐retinal leading receptor activation and signaling [60]. In 2005, Kim and colleagues engineered the first Opto‐GPCR by replacing the intracellular signaling domain of the Gα_t‐coupled bovine rhodopsin with the corresponding ICL1‐3 and C‐terminal tail of hamster β₂‐AR (Figure 3A) [61]. This chimeric protein activates the Gα_s signaling pathway of β₂‐AR in transfected cells upon illumination with 500 nm light. Airan and colleagues generated two OptoARs and targeted the expression of Opto‐β₂‐AR and a related Opto‐α₁‐AR to neurons in the nucleus accumbens of mice by lentiviral vectors with a synapsin‐I promoter [62]. Importantly, Opto‐β₂‐AR induced cAMP generation, MAP kinase activation and receptor internalization in response to optodynamic stimulation with similar kinetics as isoproterenol‐stimulated wild type β₂‐AR [63]. In addition, Opto‐β₂‐AR was used to investigate cAMP capture by the olfactory marker protein in HEK293T cells, resulting in rapid deactivation of olfactory cyclic nucleotide‐gated channel activity [64]. Neural activity regulated by these two Opto‐ARs was detected using an optrode consisting of an optical fiber and electrode to simultaneously deliver light and record electrophysiological responses, respectively, and by analyzing reward‐related behavior. This work showed that Opto‐ARs can be functionally expressed in specific cells in vivo, allowing optical modulation of intracellular signaling cascades and network physiology without supplementation of exogenous retinoids. More recently, structure‐guided optimization of Opto‐β₂‐AR has resulted in Opto‐β₂‐AR‐2.0, in which not only the ICLs of β₂‐AR but also α‐helical TM residues that are involved in the interaction with Gα_s (e.g., I139^3.54 and L280^6.37) have been introduced in rhodopsin [65]. Opto‐β₂‐AR‐2.0 induces a more profound increase of intracellular cAMP levels as compared to the first‐generation Opto‐β₂‐AR and provides a basis for examining the contribution of specific contact residues to GPCR function. Given the robust functional activity, Opto‐β₂‐AR‐2.0 also allows light‐induced detection of cAMP levels upon the introduction of three mutations in TM3 and TM7 (T118^3.33A, E122^3.37D, A318^7.39S) which were previously shown to blue shift rhodopsin absorption [65]. No light‐induced cAMP production can be detected after introducing these three mutations into Opto‐β₂‐AR [65].

The application of Opto‐ARs to activate signaling locally and rapidly without any ligands facilitates the dissection of receptor function within the neural circuit. Opto‐α₁‐AR has been expressed in astrocytes to investigate its effects on memory, synaptic plasticity, exploratory, sensory evoked response, and the development of Alzheimer's disease [66, 67, 68, 69, 70, 71]. Additionally, Opto‐β₂‐AR has been expressed in vivo to demonstrate the role of β₂‐AR in the modulation of anxiety‐like behavioral states and other stress‐induced disorders [62, 72, 73, 74]. Moreover, Lee and colleagues developed multiple transgenic CHROMus (Cornel/National Heart Lung Blood Resource for Optogenetic Mouse Signaling) mouse lines which in vivo express optogenetic Opto‐ARs [75]. These new in vivo models provide interesting new approaches to study AR‐modulation of physiology. Light‐induced activation of Opto‐α₁‐AR and Opto‐β₂‐AR leads, for example, to ex vivo vessel contraction in isolated cerebral‐ and mesenteric arteries, respectively.

2.2. Photoactivatable Proteins to Optically Control AR Signaling

Photodimerizing proteins are utilized to induce the interaction of specific proteins in living cells and have also been successfully used to control GPCR signaling [76]. The single‐domain camelid antibody (also known as nanobody) Nb60 allosterically binds to the intracellular site of β₂‐AR and stabilizes its inactive conformation [77, 78]. The β₂‐AR optobody was engineered in 2019 by fusing the N‐ and C‐terminal fragments of Nb60 with the Magnet photodimerization proteins nMagHigh1 and pMagHigh1 (Figure 3B). Upon illumination with 488 nm light, Nb60‐N‐nMagHigh1 and pMagHigh1‐Nb60‐C associate which drives the functional reconstitution of Nb60, subsequently resulting in the irreversible targeting of the β₂‐AR and the inhibition of agonist‐induced signaling [79, 80]. Cryptochrome (CRY), derived from the plant Arabidopsis thaliana, can reversibly interact with the cryptochrome‐interacting basic‐helix‐loop‐helix 1 (CIB) protein upon absorbing blue light [81]. Takenouchi and colleagues fused CIB with a flexible linker to the intracellular C‐tail of ADRB2 and CRY to the N‐terminus of mCherry–β‐arrestin2 [82]. The ADRB2_CIB–Arrestin_CRY complex forms upon exposure to blue light and Arrestin_CRY is redistributed back to the cytosol upon termination of irradiation due to the dissociation of CRY and CIB (Figure 3C). This light‐activated CRY–CIB system allows the temporal modulation of β₂‐AR interaction with β‐arrestin2, and revealed that the duration of this interaction regulates intracellular trafficking of β₂‐AR. In addition to the photosensitive β‐arrestin, several other optogenetic tools have been described to control downstream GPCR effectors [83]. For example, the CRY2–C1BN system was used to optically induce the translocation of regulator of G protein signaling 4 to the plasma membrane upon illumination with blue light, resulting in a reversal of chemokine receptor CXCR4‐induced G_βγ translocation by increasing the GTPase activity of Gα subunits [84]. Moreover, optical control of Gα_q and Gα_s translocation from the cytosol to the plasma membrane using membrane‐anchored‐nMagHigh1 in combination pMagFast1 fused to the Gα subunits, or the red light‐responsive dimerization system phytochrome B (PhyB) in combination with phytochrome‐interacting factor 6 (PIF6), allows light‐induced activation of the phospholipase Cβ and AC signaling pathways, respectively [85]. Applications of these photodimerizing approaches can be considered for the optical control of AR signaling in further studies. In addition, another light‐controlled Nb60 was more recently developed by fusing the photocleavable protein termed PhoCl [86] to the N‐terminus of the nanobody [87]. Illumination of this OptoNb60 with 405 nm light photocleaves off PhoCl, allowing the uncaged Nb60 to target β₂‐AR and inhibit agonist‐mediated signaling (Figure 3D).

3. Photopharmacology

The idea of using photo‐controlled compounds to target human disease was first conceived in the 1970s [88] and has been developed more recently for GPCRs, including ARs [47, 52, 89, 90]. Nowadays, photopharmacology is applied as a molecular tool to allow temporal and spatial control of endogenous ARs to investigate their functional properties, including activation and deactivation kinetics and trafficking [91, 92]. There are generally two strategies to achieve optical control of small molecule drugs: (1) irreversible uncaging and (2) reversible photoswitching (Figure 3E,F) [45, 52, 93]. In this section, we summarize the status of these two strategies for the control of AR activity with photoresponsive ligands.

3.1. Photocaged Ligands to Control AR Function

Photocaged compounds contain a photoreactive protecting group (PPG) covalently attached to a biologically active ligand (Figure 3E). The biologically active ligand is irreversibly released from the photoreactive protecting moiety (“cages”) upon exposure to light with a suitable wavelength. Photolysis of caged compounds is a robust technique for temporal and spatial control of biochemical response and new generations of photocages continue to emerge [94, 95, 96, 97]. Several aspects need to be considered for the design of caged ligands: (1) the most important property is that the caged analog as well as any side products released upon illumination should have negligible biological activity. (2) Rapid photorelease (uncaging) of the active ligand upon illumination by using a suitable cage moiety and optimization of the illumination conditions. (3) Stable chemical and metabolic properties of the active ligands under the used uncaging conditions [96, 98, 99].

The first photocaged compound applied in biological systems enabled the rapid release of ATP from its protected analog upon illumination at 340 nm [100]. Nowadays, numerous photocaged ligands have been reported, including AR agonists and antagonists. The first studied photocaged agonist in the field of ARs was caged‐phenylephrine (caged‐PE) for α‐ARs (Table 1) [94]. Caged‐PE exploits a 2‐nitrobenzyl PPG to inactivate phenylephrine as measured by vasoconstriction of isolated adult rat mesenteric arterioles. Vasoconstrictive activity by PE could be recovered upon flash photolysis with 365 ± 10 nm light. Subsequently, a similar strategy was applied to obtain caged‐epinephrines, as well as caged‐isoproterenol (Table 1) [101]. Caged‐epinephrine lacks bioactivity on both α‐ and β‐ARs. However, some residual activity of vessel relaxation via β‐ARs was observed in isolated rat mesenteric arterioles induced by caged‐isoproterenol and its derivatives, though it was much smaller than the effect observed with isoproterenol. Eight years later, Vaniotis and colleagues developed a more hydrophobic caged‐isoproterenol analog, allowing the resulting ZCS‐1‐67 (Table 1) to freely cross membranes [102]. Uncaging ZCS‐1‐67 upon illumination with 405 nm induced the release of NO release by activating β₃‐AR in nuclear membranes, as observed using live‐cell confocal fluorescence microscopy. Recently, photocaged optoIso (Table 1), masking the agonist with three photocleavable 2‐nitrobenzyl PPGs, was reported [98]. OptoIso allowed spatial and temporal control of β‐ARs in both endomembrane and plasma membrane regions, as monitored using live cell imaging combined with Venus‐miniG_s and Nb80‐mCherry activation sensors in HeLa cells.

Table 1.

Photocaged ligands for ARs.

Compound [ref]	AR targeta	Chemical structureb	Photochemical properties		Pharmacological properties
			Irradiation wavelength	Photolysis kinetics	Efficacy	Binding assay/functional assay
Caged‐phenylephrine [94]	α‐ARs		355 nm	Time constant (τ)≈50 μs	Agonist	Ex vivo vasoconstriction
Caged‐epinephrine [101]	ARs		355 nm	1% conversion per 0.5 ms	Agonist	Ex vivo vasoconstriction
Caged‐isoproterenol [101]	β1‐AR β2‐AR		355 nm	n.d.	Agonist	Ex vivo vasoconstriction
ZCS‐1‐67 [102]	β3‐AR		405 nm	n.d.	Agonist	[125I]CYP binding assay/cAMP production NO production
OptoIso [98]	β1‐AR β2‐AR		445 nm	Completely released after 15 min	Agonist	Nb80, miniGs recruitment G‐protein translocation cAMP production
Caged‐timolol [103]	β‐ARs		400–430 nm	Completely released after 26 min	Antagonist	In vivo IOP measurement, Cornea thickness analysis Retinal ganglion cell density measurement
Caged‐carvedilol [104]	β‐ARs		405 nm	15% release after 3 min	Antagonist	cAMP production Langendorff heart experiments Behavioral studies

Abbreviation: n.d., not determined.

^a Receptor specificity is not shown if there is no evaluation in the original study.

^b Photocleavage cages are highlighted in blue.

The strategy of photocaging was first applied to β‐blockers with the development of caged‐timolol (CT) (Table 1) [103]. Timolol, a potent inhibitor of β‐ARs, is the first‐line drug (Betimol) for decreasing the intraocular pressure (IOP) in glaucoma patients, but it is associated with overdose‐related side effects such as cardiac dysfunction [106, 107]. To allow spatial and temporal control, a dimethoxy‐substituted 2‐nitrobenzene PPG [105] was applied to achieve photocleavage upon exposure to violet–blue wavelengths (400–430 nm) of daylight. A preclinical mouse model introduced contact lenses with this caged compound that release timolol to the eyes for the treatment of IOP upon exposure to natural daylight and revealed inhibition of β‐ARs for over 10 h. More recently, caged‐carvedilol (Table 1) was generated to modulate β‐ARs [104]. Carvedilol, a nonselective β‐ARs antagonist, was conjugated to [7‐(diethylamino) coumarin‐4‐yl] methyl (DEACM) and was successfully irreversibly released upon 405 nm illumination. Introducing the photoreactive moiety of DEACM abolished the physiology response of carvedilol on β‐adrenergic signaling, but β‐ARs regulation in isolated mice hearts and zebrafish larvae can be recovered upon illumination.

In general, PPGs used for AR ligands are mainly o‐nitrobenzyl groups and the light conditions for uncaging are therefore still primarily limited to the UV range. Of note is the fact that computational studies become increasingly important for the design of caged ligands and help to determine the optimal position to attach a PPG [98, 104].

3.2. Photoswitchable Ligands to Control AR Function

Whereas the photocage strategy results in irreversible optical “ON” switches, the photoswitch strategy (Figure 3F) offers optical “ON‐and‐OFF” tools to regulate biological processes by introducing a photoisomerizable moiety in ligands that can reversibly switch between two configurations upon illumination with different wavelengths [108, 109]. Particularly, (heteroaryl) azobenzene is often used as photoswitchable moiety, due to its suitable photochemical behavior and good synthetic accessibility [52, 108]. So far, six studies have reported photoswitchable ligands for ARs [110, 111, 112, 113, 114, 115]. The first photoswitchable AR ligand adrenoswitch‐1 was developed based on the structure of clonidine, using an “azologization” strategy by replacing its short linker with an azo group and restricting the cyclic amidine moiety to the closest structurally related aromatic derivatives, such as imidazole and thiazole (Table 2) [110]. Clonidine is an FDA‐approved α₂‐AR agonist for the treatment of attention deficit/hyperactivity disorder and hypertension [116, 117, 118, 119]. Radioligand competition binding assay with α₂‐ARs and ex vivo measurements of contraction of rat aortic rings indicated adrenoswitch‐1 to be the most promising compound with a short thermal relaxation time (half‐life) of seconds. Illumination of adrenoswitch‐1 with 365 nm reduced zebrafish larvae locomotion and increased pupil diameter in both blind mice and wild‐type mice models [110]. Therefore, adrenoswitch‐1 might contribute to the development of pupil dilators that can be precisely controlled by light. It is worth noting that, unlike the parental compound clonidine, the authors hypothesized adrenoswitch‐1 to be a partial agonist of α₁‐AR, but this statement requires further pharmacological characterization.

Table 2.

Photoswitchable ligands for ARs.

Compound [ref]	AR targeta	Chemical structureb	Photochemical properties (to reach PSS cis /PSS trans )		Pharmacological properties
			Irradiation wavelength	Half‐life (min) or isomerization rate (s−1)	Activityc (photoinduced‐potency shift)	Binding assay/functional assayd
Adrenoswitch‐1 [110]	α‐ARs		365–400 nm/450–500 nm	n.d.	cis‐on (3.0‐fold) agonist	[3H]RX821002 binding assay In vitro vasodilation
pAzo‐2 [114]	β1‐AR		365 nm/525 nm	3.1 s−1/68.0 s−1	trans‐on (83‐fold) partial agonist	Alprenolol‐green binding assay cAMP production
Photo‐adrenaline [115]	β2‐AR		365 nm/400 nm	n.d.	trans‐on (2.1‐fold) agonist	[3H]CGP‐12177 binding assay β2‐AR conformation
PZL‐1 [111]	β2‐AR		380 nm/550 nm	n.d.	trans‐on (16.8‐fold) antagonist	cAMP production
PZL‐2 [111]	β2‐AR		380 nm/550 nm	n.d.	cis‐on (3.7‐fold) antagonist	cAMP production
Opto‐prop‐2 [113] (VUF17062)	β2‐AR		365 nm/430 nm	4 min/3 min	cis‐on (31.6‐fold) antagonist	[3H]DHA binding assay β2‐AR conformation
(S)‐opto‐prop‐2 [112] (VUF25474)	β2‐AR		365 nm/430 nm	4 min/3 mine	cis‐on (398.1‐fold) antagonist	[3H]DHA binding assay cAMP production

Abbreviation: n.d., not determined.

^a Receptor specificity is not shown if there is no evaluation in the original studies.

^b Photoswitch moiety is highlighted in blue.

^c The fold change is determined from functional assays in the original studies (the potency of active isomer divided by the value of the inactive isomer).

^d Assay used for activity determination.

^e Data from its racemic mixture opto‐prop‐2.

Duran‐Corbera and colleagues designed and synthesized the first photoswitchable β₁‐AR agonist pAzo‐2 using an “azologization” strategy, introducing a para‐methoxy‐substituted azobenzene, while retaining the common oxyaminoalcohol substructure to yield a high degree of β₁‐/β₂‐AR selectivity (Table 2) [114]. The ethanolamine backbone is vital for β‐AR activity by forming H‐bond interactions with key residues in the orthosteric binding pocket [114, 120]. pAzo‐2 displays quite good photochemical properties with high photostationary states (PSS) for both cis and trans isomers (more than 90%) and an appropriate thermodynamic half‐life for GPCR assays (7.1 h). Photopharmacological characterization showed trans‐on partial agonism and proper selectivity of pAzo‐2 for β₁‐AR using a FRET‐based cAMP assay in HEK293 cells stable expressing β₁‐AR. Moreover, pAzo‐2 allowed the optical modulation of the cardiac rhythm in zebrafish larvae. Recently, the photoswitchable β₂‐AR full agonist photo‐adrenaline (Table 2) was developed using an “azoextention” strategy to introduce an arylazopyrazole moiety to the N‐Me group of adrenaline via linker at the ortho position [115]. Photo‐adrenaline displays a trans‐on β₂‐AR binding affinity and full agonism in a BRET‐based β₂‐AR conformation sensor with a comparable potency as isoproterenol. Yet, the optical modulation of the agonist activity is rather limited (Table 2).

The first photoswitchable β‐blockers were designed by Duran‐Corbera et al. [111] (Table 2) based on the 3‐aryloxypropan‐2‐olamine molecular scaffold which is the pharmacophore of β‐ARs antagonists [121]. All synthesized photoswitchable antagonists achieved a relatively high PSS _cis value of more than 85% and PSS _trans value of more than 70%. Notably, PZL‐1 with an ortho‐substituted azobenzene moiety shows a 17.4‐fold decreased inhibitory potency (trans‐on) for β₂‐AR upon the illumination of 380 nm monitored by a FRET‐based EPAC cAMP assay. Additionally, PZL‐2 exhibits increased inhibitory ability of cimaterol‐induced β₂‐AR response. In parallel, Bosma and colleagues designed and synthesized the opto‐prop‐2 series (Table 2) by applying an “azologization” strategy to the naphthalene core of the β‐blocker propranolol [113]. Opto‐prop‐2 exhibits a thermodynamic half‐life of more than 10 days and a PSS value of more than 75% for both trans and cis isomers. Photopharmacological characterization revealed that opto‐prop‐2 shows a good selectivity for β₂‐AR and an increased affinity and inhibitory ability of isoproterenol‐induced β₂‐AR response upon trans‐cis isomerization after illumination at 365 nm using [³H]DHA displacement binding experiments and BRET‐based conformational assay, respectively. Also, preilluminated opto‐prop‐2 attenuated isoproterenol‐induced contractility of isolated adult rat cardiomyocytes. To improve the properties of opto‐prop‐2, Shi and colleagues recently described the two enantiomers of opto‐prop‐2, that is, (S)‐opto‐prop‐2 (Table 2) and (R)‐opto‐prop‐2 [112]. (S)‐opto‐prop‐2 was identified as the most active enantiomer with similar photochemical properties but an increased gain of photo‐induced affinity for β₂‐AR as compared to racemic opto‐prop‐2; the cis‐(S)‐opto‐prop‐2 shows a 1000‐fold increase in affinity after illumination of trans‐(S)‐opto‐prop‐2 with 365 nm. Moreover, (S)‐opto‐prop‐2 allowed optical modulation of β₂‐AR activity in a number of assays in vitro (FRET‐based cAMP assay and Red‐upward cADDis cAMP sensor assay) and in vivo (rabbit IOP measurements). Molecular docking of (S)‐opto‐prop‐2 (Table 2) in β₂‐AR crystal structures has been validated with site‐directed mutagenesis studies and these studies have highlighted crucial residues D113^3.32, N312^7.39, and F289^6.51 for the β₂‐AR binding of cis‐(S)‐opto‐prop‐2.

4. Future Perspectives

Only 19 years ago, optogenetics was first introduced to the field of ARs, nowadays allowing the precise control of genetically defined subsets of neurons to primarily investigate AR in brain (patho)physiology. Studies have shown promising optogenetic applications in nonhuman primates [122]. Recently, the partial recovery of visual function was reported after optogenetic therapy of humans with microbial opsin, which is a breakthrough in the field of optogenetics [123]. Currently, an open‐label clinical trial evaluates the safety and efficacy of this optogenetic approach (NCT03326336, details in clinicaltrials.gov). The results of this study are eagerly awaited and will hopefully pave the way for actual clinical applications. Complimentary to optogenetics, photopharmacology is an emerging field with promises for future clinical applications of light‐sensitive drugs. Currently, KIO‐301, an azobenzene‐based photoswitchable ligand which blocks voltage‐gated ion channels including hyperpolarization‐activated cyclic nucleotide‐gated (HCN) and potassium channels during exposure to visible light, is in Phase I/II clinical trial to evaluate its safety, tolerability, and efficacy in terms of vision recovery (NCT05282953, details in clinicaltrials.gov).

Hitherto, various in vivo applications of AR optogenetics tools have been applied in the field of neuroscience in combination with optical fibers. Due to the straightforward nature of light delivery, synthesized photopharmacological tools targeting ARs have predominantly been utilized in the visual system of experimental animals, such as the eyes of mice and rabbits, but also in transparent model organisms (such as zebrafish) [110, 112]. Although ARs are well‐expressed in the heart, only few studies focus on cardiac systems due to limited light delivery and complex cardiac pathophysiology [124, 125]. Indeed, efficient photoresponsivity is strongly affected by optimal penetration of light in the targeted tissue. The best penetration depth (4–5 mm) is obtained with light in the region of near‐infrared (NIR; 700–900 nm) with limited scattering and photobleaching, compared to the less than 1 mm or 1–2.5 mm penetration by UV (315–400 nm) or visible light (400–700 nm), respectively [126, 127]. Also, the in vivo safety‐limiting irradiance dose increases from UV to NIR with high‐energy UV light inducing phototoxicity, including DNA damage or cell death [128, 129]. However, at higher intensities, red‐shifted light might generate heat within illuminated tissues resulting in thermal damage to cells [130, 131]. Hence, NIR light (700–900 nm) has been suggested as the optimal light condition for in vivo (clinical) photopharmacology [132].

4.1. Design of the Next Generation Phototools for ARs

4.1.1. Optogenetic Tools

The next generation of optogenetics for ARs should be based on the recently developed red‐shifted opsins, such as channelrhodopsin (ReaChR) [133] and ChrimsonR [134]. Their responsiveness to longer wavelengths with deeper tissue penetration will allow broader application of future red‐shifted Opto‐AR in whole‐body studies. Optogenetic tools are mostly expressed in vivo using an adeno‐associated virus (AAV) vectors that can be injected in a specific region, and use cell‐type specific promoters to acquire targeted expression of the optogenetic protein. Yet, immunogenicity and genotoxicity need to be carefully evaluated for AAV‐based delivery of optogenetics proteins. In addition, optogenetic tools might autoactivate in the dark due to high expression level and the duration of expression resulting in undesired cell toxicity [135].

4.1.2. Photopharmacology Tools

Since 2007, ~80 structures of ARs have been determined by X‐ray crystallography and cryo‐electron microscopy (cryo‐EM) [136]. The β₂‐AR is both the best characterized and one of the most pharmacologically significant AR subtypes. It was one of the first GPCRs to be characterized through radioligand binding, the first non‐rhodopsin GPCR gene to be cloned [137] and the first non‐rhodopsin GPCR crystallized in its inactive state [7, 15]. Additionally, the crystal structure of the active β₂‐AR–G_s complex was obtained by Brian K. Kobilka and Roger Sunahara, providing the first high‐resolution view of transmembrane GPCR signaling [138]. In 2012, Brian K. Kobilka as well as Robert J. Lefkowitz were jointly awarded the Nobel Prize in Chemistry because of their studies on GPCRs. A number of β‐AR structures have now been determined in the active and inactive states [139, 140, 141]. Moreover, recent structures of active and inactive α_1A‐AR [23], inactive α_1B‐AR [22], active and inactive α_2A‐AR [18, 142], active α_2B‐AR‐G_i/o complex [20], and inactive α_2C‐AR [19] have revealed the basis of ligand subtype selectivity between α₂‐ARs and β‐ARs. This progress in the structural characterization of ARs has been instrumental in the understanding of the mechanism of ligand recognition by ARs and will be important for the future development of photopharmacological tools with a specific focus on where to insert photoresponsive moieties and how to acquire appropriate AR subtype selectivity.

Members of each major subtype of ARs share 51%–64% overall sequence identities and still 30%–40% identities among different major types [31, 143]. Particularly, β₁‐AR and β₂‐AR are highly similar in sequence (57% overall, 70% within binding pocket). There is only one amino acid difference between the antagonist binding pocket of β₁‐AR and β₂‐AR while the amino acids lining the edge of the pocket exhibited some structural diversity [143]. Polypharmacology (drug interacting with multiple targets) is common in the AR field due to the sequence conservation [31]. To date, 14 photopharmacological tools target ARs. Some of them are reported to display subtype receptor selectivity (based on affinity or functional properties), such as pAzo‐2, PZL1/2, opto‐prop‐2, and (S)‐opto‐prop‐2, but in general evidence for receptor subtype selectivity is limited and should be accounted for in future AR photopharmacology studies.

The second generation of AR photopharmacology tools is expected to show high specificity via either structure‐based design approaches or the use of ligands with high AR subtype as starting point for the design of photopharmacological tools. Numerous compounds have been developed to target ARs. Some exhibit good subtype selectivity, such as oxymetazoline binding to α_1A‐AR and guanfacine binding to α_2A‐AR. Dobutamine and salmeterol specifically bind to β₁‐AR and β₂‐AR, respectively, and mirabegron only binds β₃‐AR. Also, there is extensive knowledge on the structure–activity relationship (SAR) of various adrenergic drugs. This SAR will be very useful to properly design new AR photocages by coupling AR ligands with PPGs and to develop new photoswitchable ligands. Kobauri and colleagues recently reported on various general photoswitch strategies [144], whereas Wijtmans and colleagues focussed specifically on photoswitchable GPCR ligands [52].

Several factors must be considered in the design of photoresponsive molecules. First consideration is the synthetic feasibility of the molecule, which includes the availability of starting materials, number of synthetic steps, yields, purification methods, safety, etc. The second factor to consider is the solubility and stability of the molecule. Solubility needs to be suitable for pharmacology assays. Stability, both in storage and in buffer, also needs to be assessed and has been reported to be problematic for some (substituted) azobenzene‐derived phototools [145]. There are several methods to improve stability. For example, we recently observed that converting amine‐bearing compounds to their salt forms can resolve instability issues during storage (manuscript in preparation). A third consideration is the photochemical properties of the molecule, which is very important for selection of the proper GPCR assays and pharmacological outcomes. For photocaged ligands, key parameters include absorption maxima (λ _max), molar absorptivity (ε), quantum yield (QY), and irradiation wavelengths for photocleavage. For photoswitchable molecules, important factors include absorption maxima (λ _max), PSS values, wavelengths for photoswitching, time required for isomerization, half‐life (thermal relaxation of the metastable photoisomer), and resistance to photobleaching. The optimal thermal isomerization half‐life, which can range from seconds to years, is determined by the specific application of interest. Numerous studies have shown that different substituents at different positions are vital in shaping both photochemical and biological properties [146, 147, 148, 149].

As discussed above, the use of red‐shifted wavelengths is preferred in vivo to allow deeper tissue penetration and avoid UV‐associated cell damaging. Therefore, second generation AR photopharmacology tools should aim for uncaging or photo‐isomerization at higher wavelengths.

The first reported visible‐light‐absorbing PPG is based on ruthenium polypyridyl complexes [150] whereas different BODIPY, coumarin, cyanine, and other PPGs can now also be used to increase the absorption wavelengths (as reviewed by Josa‐Cullere and Llebaria [151]). In these approaches, known PPGs are chemically modified, that is, by extending the conjugation system of the chromophore. Another strategy involves the development of new PPGs by transforming known visible‐light‐absorbing fluorophores (400–700 nm).

Photoswitchable molecules targeting ARs have so far designed based on azobenzene and the conventional derivatives (arylazopyrazole and azoimidazole) with the photoisomerization under blue or green light. In the field of azoswitches, significant progress has been made in the development of bathochromically shifted photoswitches, which can be switched in both directions with visible light due to the separation in n–π* transitions of the trans and cis photoisomers. The most prominent red‐shifted azobenzenes are tetra‐ortho‐substituted azobenzenes, such as tetra‐ortho‐fluoroazobenzene [152], tetra‐ortho‐methoxyazobenzenes [153, 154], and tetra‐ortho‐chloro‐azobenzenes [155]. Recently, beyond conventional azobenzene derivatives, a variety of visible‐light heteroarylazobenzenes have been reported, and some of them show impressive photoswitching properties, such as high thermal stability and (near‐)quantitative bidirectional photoisomerization [156]. Bis‐heteroaryl azo switches further expand the chemical diversity and provide a promising way to improve photochemical properties. He et al. reported a family of azobispyrazoles, capable of achieving (near)‐quantitative bidirectional photoisomerization and widely tunable thermal half‐life, ranging from hours to years [157]. Other fully‐visible‐light‐operated photoswitches have also been developed, such as hemithioindigos [158, 159] and iminothioindoxyls [160]. Fink and colleagues reported aryl‐azocyclopropeniums, the minimalist visible‐light photoswitches, which exhibit a redshifted π–π* absorbance bond compared to azobenzenes. This allows photoisomerization under purple or green light irradiation (385 and 505 nm, respectively) with high‐yielding bidirectional switching up to 90%. The thermal half‐life of these photoswitches ranges from minutes to hours [161]. All the discussed new photoswitch moieties exhibit promising applications in photopharmacology, but mostly await application in the GPCR field. A visible‐light phenylazobenzimidazole derivative, benzimidazole azo‐arenes, behaving as a trans‐on agonist, has, however, been shown to be a suitable molecular probe to study β‐arrestin2 signaling via the cannabinoid 2 receptor (CB₂R) [162].

Besides one‐photon excitation (1PE) mentioned above, two‐photon (2PE or TPE), or even three‐photon excitation (3PE) of photoresponsive molecules, utilizing pulsed femtosecond NIR illumination has also been developed [163]. The advantages of 2PE and 3PE include higher spatiotemporal resolution and deeper tissue penetration with minimal toxicity, compared to UV light excitation. Many photocages have been designed by combining known PPGs, especially red‐shifted ones, with suitable two‐photon light‐harvesting antennas [164, 165]. This strategy has also been successfully applied to photoswitchable ligands [166]. There is sufficient room for new approaches in this field, mainly improving the two‐photon absorption, the water solubility at sufficient concentrations, and the hydrolytic stability. Another effective strategy for shifting the wavelength of photoresponsive reactions toward the visible to NIR region is the use of upconverting nanoparticles (UCNPs) [167]. UCNPs can be applied to convert multiple low‐energy incident photons (long wavelength) into the emission of a higher‐energy photon (short wavelength). Currently, these strategies have not found use in the AR photopharmacology field and deserve attention in the future.

4.2. In Vivo Light Delivery

Successful in vivo regulation of AR optogenetic and photopharmacological tools rely on the ability to deliver sufficient light at targeted tissue. Photodynamic therapy (PDT) has already proven its clinical use in the form of light‐activated therapies. This approach consists of a light‐activated chemical (photosensitizing agent) and the local administration of light to induce cell damage or death through the generation of reactive oxygen species (ROS) (singlet oxygen, hydroxyl radical, peroxides, and superoxide) [168]. Diode lasers and LEDs are used for the noninvasive and direct irradiation of superficial tumors or skin, while optical fibers involving side‐emission zones or lenses have been adopted for deep tumors [126]. PDT provides effective clinical treatments in the fields of dermatology and oncology and shows a good tolerance with patients. The field is developing rapidly and recently some remarkable light management technologies have been developed and applied for local light delivery in vivo, such as wireless subdermal electronics, multimode optical fibers [90], and wireless optofluidics [169]. For example, Photofrin, as a photosensitizer for PDT, can be activated to an excited electronic state and contribute to ROS upon the illumination with 630 nm [170]. During the treatment of non‐small cell lung cancer, a fiber‐optic light diffuser is placed using navigational bronchoscopy and offers illumination for Photofrin [171]. With the future availability of more optogenetic and photopharmacology tools, the development and application of sophisticated in vivo light delivery systems will most likely push the therapeutic use of light‐controlled GPCR ligands.

5. Summary and Outlook

Over the past 30 years, optical control of ARs has been advanced through both optogenetic approaches and freely diffusible photopharmacological ligands. This review highlights four major types of optogenetics strategies targeting ARs: Opto‐ARs to study AR‐mediated signaling, photo‐induced β‐arrestin2 coupling to β₂‐AR to study subcellular trafficking, photodimerization‐based Opto‐Nb60, and photocleavage‐based Opto‐Nb60 to inhibit receptor signaling. Photopharmacological tools targeting ARs include 14 ligands, with a focus on β‐ARs over the past 5 years. These tools have been systemically tested using in vitro, ex vivo, and in vivo assays, providing spatial–temporal control of AR signaling. Currently, most photopharmacological tools of ARs are controlled by UV light. The development of red‐shifted photopharmacological AR tools is highly anticipated, as they would significantly enhance their in vivo applicability and potential therapeutic use. Despite these challenges, the future of optical AR tools is promising, and they provide valuable contribution to dissecting the complex pharmacology of AR subtypes in human physiology. Such tools will not only be sophisticated research tools but might in the future also be investigated for potential localized therapeutic strategies.

Data Availability Statement

The authors have nothing to report.