Optogenetic approaches to control Ca²⁺-modulated physiological processes

Abstract

As a versatile intracellular second messenger, calcium ion (Ca²⁺) regulates a plethora of physiological processes. To achieve precise control over Ca²⁺ signals in living cells and organisms, a set of optogenetic tools have recently been crafted by engineering photosensitive domains into intracellular signaling proteins, G-protein coupled receptors (GPCRs), receptor tyrosine kinases (RTKs), and Ca²⁺ channels. We highlight herein the optogenetic engineering strategies, kinetic properties, advantages and limitations of these genetically-encoded Ca²⁺ channel actuators (GECAs) and modulators. In parallel, we present exemplary applications in both excitable and non-excitable cells and tissues. Furthermore, we briefly discuss potential solutions for wireless optogenetics to accelerate the in vivo applications of GECAs under physiological conditions, with an emphasis on integrating near-infrared (NIR) light-excitable upconversion nanoparticles (UCNPs) and bioluminescence with optogenetics.

Graphical Abstract

1. Introduction

Ca²⁺, as a signal for life and death and secondary messenger, controls a myriad of cellular events, including muscle contraction, neurotransmitter release, lymphocyte activation, gene expression and metabolism^1–3. In mammalian cells, the intracellular Ca²⁺ concentration is tightly regulated through coordinated actions of a repertoire of Ca²⁺-signaling components, including G-protein coupled receptors (GPCRs), the receptor tyrosine kinases (RTKs), various Ca²⁺ channels and transporters/exchangers, and intracellular Ca²⁺ sensing and buffering proteins^1–3. Upon engagement of membrane receptors by ligands or antigens, phospholipase C gamma (PLCγ) is phosphorylated and activated to catalyze the hydrolysis of phosphatidylinositol 4,5-biphosphate (PI(4,5)P₂), yielding two important secondary messengers, diacylglycerol (DAG) and inositol 1,4,5-triphosphate (InsP₃). InsP₃ further binds to ER-resident InsP₃ receptors to induce Ca²⁺ release from the ER Ca²⁺ store, followed by the activation of store-operated Ca²⁺ entry (SOCE)². SOCE is best exemplified by the Ca²⁺ release-activated Ca²⁺ (CRAC) channel composed of ORAI1 and the stromal interaction molecule 1 (STIM1), and constitutes the primary route of Ca²⁺ entry in non-excitable tissues^4–6. By taking advantage of naturally-existing photoreceptors derived from microbes, plants or mammals, these signalling events can be recapitulated through optogenetic engineering to achieve tailored function. Compared to pharmacological, chemogenetic and genetic approaches, these genetically-encoded Ca²⁺ channel actuators (GECAs) have the advantages of rapid reversibility, superior spatiotemporal resolution, and non-invasiveness⁴. In this review, we present the latest progress in this endeavor, with a focus on the design principles, kinetic features and caveats to be taken into consideration during the application of these tools. When used in combination with genetically-encoded Ca²⁺ indicators, GECAs permit the simultaneous monitoring and interrogation of Ca²⁺-modulated physiological processes both in vitro and in vivo.

2. Engineered photoresponsive GPCRs

As the founding member of the optogenetic field, channelrhodopsin-2 (ChR2) derived from the green alga Chlamydomonas reinhardtii has transformed the neuroscience field by enabling precise control of neural circuits with light both in vitro and in vivo^7–9. As a cation permeant channel, ChR2 permits the non-selective entry of Na⁺, H⁺, and Ca²⁺ into the cytosol upon blue-green light stimulation (450–510 nm), thereby depolarizing the plasma membrane (PM) to trigger neuronal firing¹⁰ (Fig. 1a). However, given the over 75-fold higher concentrations of extracellular Na⁺ over Ca²⁺ (136–151 mM versus 1.4–1.8 mM), the Ca²⁺ permeability of ChR2 under physiological conditions is very low (P_Ca/P_Na: 0.15)¹¹, thus greatly hampering its application in non-excitable cells and tissues. One ChR2 variant, designated CatCh for Calcium translocating Channelrhodopsin, harbors a point mutation L132C with the Ca²⁺ permeability enhanced by 1.6-fold (P_Ca/P_NA: 0.24)^{11, 12}, but still falls far behind voltage- or store-operated Ca²⁺ channels (P_Ca/P_Na: > 1000)⁶. More Ca²⁺-specific ChR2 variants are yet to be developed.

Figure 1. Optogenetic tools for remote activation of intracellular Ca²⁺ signaling.

a. Channelrhodopsin-2 (ChR2) as a cation-permeant channel is non-selective toward H⁺, K⁺, Na⁺, and Ca²⁺.

b. OptoXRs as light-sensing rhodopsin-GPCR chimeras. The extracellular and TM regions are derived from a light-sensitive rhodopsin, while intracellular regions from different adrenergic receptors (β₂AR or α₁AR).

c. Blue light-mediated isomerization of the light-absorbing retinal cofactor triggers the conformation change of melanopsin (OPN4) and subsequent activation of Gα_q-phospholipase C (PLC) to induce Ca²⁺ influx via TRPC channel.

d. Opto-RTKs as photoactivatable membrane-anchored receptor tyrosine kinases. The intracellular domain of RTKs (FGFR or TrK), is fused to CRY2 or other optical dimerizers, thereby enabling light-inducible clustering to activate downstream signaling, including the PLC pathway and subsequent Ca²⁺ mobilization.

e. LOV2-CaM-M13 hybrid protein as a photoactivatable Ca²⁺ releaser (PACR). Photostimulation results in the release of bound Ca²⁺ into the surrounding environment.

f. Domain architecture of STIM1. SP, signal peptide; EF, EF-hand Ca²⁺-binding motif; SAM, sterile alpha motif; TM, transmembrane domain; CC1, coiled-coil domain 1; SOAR, STIM-Orai activating region; PS, proline-serine rich domain; PB, polybasic domain.

g. Genetically-encoded Ca²⁺ channel actuators (GECAs) derived from STIM1.

(Left) schematic illustrating functional STIM1-ORAI1 coupling in response to store depletion. Ca²⁺ depletion within the ER lumen induces oligomerization of the luminal EF-SAM domain to initiate STIM1 activation and triggers conformational switch to expose SOAR/CAD and the C terminal PB domain. Next, STIM1 undergoes further oligomerization and migration toward the PM, a process facilitated by the association between SOAR/CAD and ORAI1, as well as the interaction between positively-charged PB and negatively-charged phosphoinositides embedded in PM. SOAR/CAD is responsible for direct engagement and activation of ORAI channels to mediate Ca²⁺ influx.

(right) Optogenetic engineering to convert store-operated Ca²⁺ entry (SOCE) into light-operated Ca²⁺ entry (LOCE). (i) A protein oligomerization-based engineering strategy using various optical dimerizers (Opto-CRAC2) and CRY2-STIM1 fusion proteins (OptoSTIM1, monSTIM1, and eOS1). Upon light illumination, CRY2 or optical dimerizers undergoes oligomerization, mimicking Ca²⁺ -depletion induced EF-SAM multimerization in ER lumen, to trigger STIM1 activation and the subsequent opening of ORAI Ca²⁺ channels in the PM. (ii) The design of GECAs based on LOV2 conformational switch (Opto-CRAC1, LOVS1K or BACCS), in which the CC1-SOAR interaction-mediated intramolecular autoinhibition is mimicked by LOV2-SOAR fusion in the dark. Upon photostimulation, LOV2 undergoes conformational changes to expose SOAR/CAD, which engages PM-resident ORAI Ca²⁺ channel to induce Ca²⁺ influx.

Instead of site-directed mutagenesis, Airan et al. explored an alternative approach by generating a series of rhodopsin-GPCR chimeras, designated “OptoXRs”¹³, to photo-manipulate intracellular second messengers, including cAMP and Ca²⁺. In a typical OptoXR, the intracellular loops of rhodopsin were replaced with the intracellular regions of G_q-coupled adrenergic receptors (Fig. 1b). Upon photostimulation at 504 nm, OptoXR evoked cytosolic Ca²⁺ elevation via the activation of the PLC pathway in HEK cells, and was further applied to photo-control behaviors in awake mice¹³. A limitation of ChR2 variants and OptoXRs is that the broad rhodopsin absorption peak (440–590 nm) may potentially conflict with biosensors used to report the functional consequences of optogenetic tools^{14, 15}. In addition, owing to the rapid trans-to-cis isomerization of the retinal co-factor (tens of ms), sustained Ca²⁺ entry requires constant photostimulation and tends to cause undesired photo-toxicity. The latter concern is partially mitigated by the use of naturally-existing melanopsin (encoded by OPN4), a photopigment protein found in human retinal ganglion cells and the iris of mice and primates¹⁶ (Fig. 1c). Melanopsin can generate both sustained and oscillatory Ca²⁺ signals within seconds (t_{1/2, ON}: < 10 s) to activate a downstream Ca²⁺-responsive master transcription factor, the nuclear factor of activated-T cells (NFAT)¹⁷. Melanopsin has been introduced into mouse embryonic stem cell-derived embryoid bodies (EBs) to generate pacemaker-like activity in cardiomyocytes within EBs following photostimulation¹⁸. When expressed in non-excitable primary T cells, melanopsin could photo-boost tumor killing in a mouse model of hepatocarcinoma¹⁹. Similar to the design of OptoXRs, a chimeric receptor, Opto-mGluR6, composed of melanopsin and metabotropic glutamate receptor 6 (mGluR6), has been devised and expressed in retina to restore the vision of blind mice with retinal degeneration²⁰. One potential caveat associated with these GPCR-based optogenetic actuators is the paralleled stimulation of the DAG/PKC pathway following PLC activation, which could cause non-Ca²⁺-related functional crosstalk to complicate data interpretation.

3. Photoactivatable RTKs

Similar to GPCR, the engagement of RTKs by their cognate ligands often leads to receptor multimerization to activate the PLCγ pathway for Ca²⁺ mobilization. To faithfully recapitulate this process, optical multimerizers, including Arabidopsis cryptochrome 2 (CRY2), the light-oxygen-voltage domains (LOV) of aureochorme, and Deinococcus phytochrome BphP (DrBphP), have been fused to the catalytic intracellular kinase domain of RTKs to generate Opto-RTKs^21–24 (Fig. 1d). Most engineering efforts have been devoted to two RTKs, the fibroblast growth factor receptor 1 (FGFR1)^{21, 22} and the tropomyosin-related kinase (Trk)²⁴. Upon light stimulation, the photosensitive moieties undergo dimerization or oligomerization to drive the activation of the intracellular kinase domain of RTKs. Repetitive Ca²⁺ signals can be generated by switching on and off blue light for CRY2 or LOV-based tools^21–24, or toggling between near-infrared (780 nm) and far-red light (660 nm) for the case of DrBphP²⁵. Opto-RTKs can thus be applied to control intracellular Ca²⁺ signals using both visible and non-visible light. The use of NIR light further enables multiplexed experiments using fluorescent probes and biosensors in the blue-green range without activation of the engineered receptors. Opto-RTKs, however, are fraught with the lack of Ca²⁺ signaling fidelity because of simultaneous activation of multiple downstream pathways. For example, aside from the Ca²⁺-associated PLC pathway, photostimulation of Opto-FGFR1 could co-activate proliferation-and survival-related RAS-MAPK, PI3K-AKT and STAT-dependent signaling cascades^21–24. Extra cautions, therefore, have to be taken when using Opto-RTKs to interrogate Ca²⁺-regulated signaling events.

4. Photoswitchable Ca²⁺ releaser

Optogenetic engineering has also been extended to intracellular Ca²⁺ sensor proteins, such as the four EF-hand-containing calmodulin. A photoactivatable Ca²⁺ releaser (PACR) has been generated via optimized insertion of LOV2 into an engineered calmodulin-M13 fusion protein (CaM-M13) (Fig. 1e). In the dark, PACR interacts with Ca²⁺ with a dissociation constant (K_d) of approximately 16 nM. Upon blue light stimulation, structural changes within LOV2 disrupt the CaM-M13 interaction, which in turn causes Ca²⁺ release from PACR because of a pronounced reduction in the affinity for Ca²⁺ (K_d = 3.75 μM)²⁶, thereby leading to an elevation of [Ca²⁺]_cyto. Even though highly specific for Ca²⁺, the application of PACR faces three drawbacks: (i) the limited Ca²⁺ releasing capability (<90 nM increase of [Ca²⁺]_cyto) and (ii) cell-to-cell variations because of varying levels of PACR expression; and (iii) the potential perturbation to resting cytosolic Ca²⁺ level and complications from other CaM-regulated targets.

5. Optogenetic actuators engineered from the CRAC channel

During SOCE activation, STIM1 dynamically couples to ORAI1 via two major molecular steps (Fig. 1f–g): (i) the initiation of STIM1 activation via Ca²⁺-depletion induced oligomerization of the luminal domain containing the EF-SAM domain; (ii) inside-out signal propagation toward the STIM1 cytoplasmic domain (STIM1ct) to overcome an intramolecular inhibition primarily mediated by the interaction between the coiled-coil 1 region (CC1) and the STIM-Orai activating domain (SOAR)^{2, 5, 6, 27}. Two major optogenetic engineering approaches, using either CRY2 or LOV2 photosensory domains derived from plants, have been applied to reconstruct these key molecular steps to enable optical control of Ca²⁺ influx through CRAC channels (Fig. 1g).

5.1. CRY2-based OptoSTIM1 and variants

To recapitulate EF-SAM oligomerization during initiation of SOCE, the luminal and transmembrane domains of STIM1 have been replaced by the photolyase-homology region (PHR) of CRY2 to yield OptoSTIM1²⁸, which has an activation kinetics of <1 min in response to blue light stimulation and a deactivation kinetics of around 4–6 min upon the withdrawal of light (Fig 1g and Table 1). Very recently, two further improved versions, monster-OptoSTIM1 (monSTIM1) and enhanced OptoSTIM1 (eOS1), have been generated to enhance the light sensitivity and dynamic range of Ca²⁺ response by 55-fold and 5~10-fold, respectively. The deactivation half-life of the OptoSTIM1 toolset generally ranges from 4 to 7 min^{28, 29}. For eOS1, it takes about 10 min to return to the basal level³⁰. In addition to CRY2-mediated homo-oligomerization, the fusion of optical dimerizers, made of iLID/SspB or CRY2/CIBN, with STIM1ct can likewise photo-trigger Ca²⁺ influx in mammalian cells, with the activation and deactivation half-lives falling in the range of ~30 sec and 40–190 sec, respectively³¹. Furthermore, ER-tethered versions were developed to manipulate membrane contact site (MCS) assembly at ER-PM junctions and induce localized Ca²⁺ influx (t_{1/2, on} = 37.8 ± 5.3 s and t_{1/2, off} = 97.5 ± 16.5 s) in response to blue light stimulation³¹. These tools set the stage for more physiologically-relevant mimicry of native STIM1 puncta formation and Ca²⁺ microdomain generation.

Table 1.

Summary of optogenetic tools for noninvasive and precise control Ca²⁺ signals in mammals.

Names	Engineering templates	Photosensitive components	Stimulation	Activation (t1/2, sec)	Deactivation (t1/2, sec)	Cell / Tissue Types	Refs
Photoswitchable GPCRs and G proteins
ChR2	Channelrhodopsin-2	opsin	Blue light	~0.002	~0.01	Multiple mammalian cells, model organism, rodents and primates	7
OptoGPCR / OptoXRs	Gq-coupled α1AR	opsin	Blue/green light	N/A	N/A	HEK293 and mice	13
	Melanopsin	opsin	Blue light	< 10	N/A	Multiple mammalian cells and mice	47, 48
Photoactivatable RTKs
OptoRTKs	FGFR1398–822	CRY2phr	Blue light	<30	N/A	HeLa/HUVECs	21, 22
	TrkA/B/C					HeLa	24
	TrkA	DrBphP-PCM	nir	~7	~10	HeLa	25
STIM1-based GECAs
BACCS variants	STIM1347–448	LOV2	Blue light	< 30	~30–60	HEK293T	33
	STIM1347–448	2*LOV2
	ORAI1 + STIM1347–448	2*LOV2
	dOrai + dmStim413–514	2*LOV2				S2
LOVS1K	STIM1233–450	LOV2	Blue light	N/A	N/A	HeLa	34
Opto-CRAC1 variants	Mouse STIM1336–486	LOV2	Blue light	23.4 ± 4.2	24.9 ± 4.8	HeLa and multiple cell lines	32
			NIR	<30	<30	BMDCs and mice
	Human STIM1336–685	LOV2	Blue light	36.3 ± 6.4	n/a	HeLa
	chicken STIM1262–412	LOV2	Blue light	~10	<30	Skin mesenchymal cells	35
	zebrafish STIM1341–442	LOV2	Blue light	18.7	24.5	HeLa	36
Opto-CRAC2 variants	STIM1233–685	iLID/SspB	Blue light	28.5 ± 3.2	48.6 ± 5.4	HeLa	31
	STIM1233–685	CRY2PHR/CIBN		23.4 ± 2.6	153.0 ± 26.2
	ER-tethered-STIM1233–685	CRY2PHR		37.8 ± 5.3	97.5 ± 16.5
OptoSTIM1 variants	STIM1238–685	CRY2PHR	Blue light	64.5 ± 4.8	274 ± 23.7	HeLa and multiple cell lines	4, 28
	STIM1238–463			43.2 ± 3.8	383 ± 52.3
	STIM1342–685			48.2 ± 5.4	344 ± 34.6
monSTIM1	STIM1238–685	CRY2PHR-E281A-9A	Blue light	24.8 ± 1.0	513.5 ± 25.8	HEK and neurons and mice	29
eOS1	STIM1238–685	CRY2PHR-E490G	Blue light	<60	<600	B3Z T cell	30
			Two-photon			In vivo CD8+ T dells
Light-switchable inhibition of Ca2+ signaling
OptoRGK	Rem1–266	iLID/SspB	Blue light	3.2 ± 1.0*	23.0 ± 2.4*	HEK293; HL-1	37

^*cytosol-plasma membrane translocation

Abbreviations: BACCS, blue light-activated Ca²⁺ channel switch; BMDCs, bone marrow-derived DCs; CaM-M13, calmodulin fused with the M13 peptide; ChR2, Channelrhodopsin-2; CIBN, the N-terminal domain of CIB1; CRY2, cryptochrome2; eOS1, enhanced OptoSTIM1; FGFR1, fibroblast growth factor receptor 1; GECA, genetically encoded Ca²⁺ actuator; HUVECs, human umbilical vein endothelial cells; iLID, improved light-induced dimer (LOV2-SsrA); LOV2, light-oxygen-voltage sensing) domain; LOVS1K, LOV2 fused STIM1 fragments (233–450); monSTIM1, monster-OptoSTIM1; PACR, photoactivatable Ca²⁺ releaser; Opto-α₁AR, Optogenetic G_q-coupled α1-adrenergic receptor; Opto-CRAC, Optogenetic Ca²⁺ release-activated Ca²⁺ channel; OptoRGK, Optogenetic Ras-like GTPase Rad/Rem/Gem/Kir; Opto-RGS2, Optogenetic regulators of G-protein signaling; Opto-RTKs, Optogenetic receptor tyrosine kinases; OptoSTIM1, Optogenetic stromal interaction molecule 1; OptoXRs, optogenetic control of intracellular signal transduction; S1, somatosensory cortex; S2, Drosophila S2 cells; SP, Signal peptide of STIM1; NIR, near-infrared; TM, transmembrane domain; Trk, tropomyosin-related kinase;

5.2. Opto-CRAC series and other LOV2-based GECAs

To mimic the intramolecular autoinhibition of STIM1 mediated by the juxtamembrane CC1 and SOAR, CC1 was replaced by LOV2 and various STIM1ct fragments were optimized to generate LOV2-STIM1ct chimeras (Opto-CRAC^{31, 32}, BACCS³³, and LOVS1K³⁴). These LOV2-based GECAs enable more rapid and readily reversible control of Ca²⁺ signals, with the activation kinetics ranging from 10 sec to 30 sec and the deactivation half-lives in the range of 30–50 sec (Table 1). In these chimeric constructs, LOV2 cages the fused STIM1ct fragment in a largely inactive conformation presumably through steric hindrance. Upon photostimulation, the photochemical reaction within LOV2 leads to the unfolding of the C-terminal Jα helix to unleash the caged STIM1ct fragments, thereby exposing the SOAR/CAD region to engage and gate ORAI Ca²⁺ channels in the PM (Fig. 1g). Opto-CRAC³² or BACCS variants³³, either anchored to PM or using STIM1ct fragments derived from species other than human and rodent (fruit fly³³, chicken³⁵, and zebrafish³⁶) further diversify the kinetic properties of the LOV2-based GECAs. In addition, these LOV2-based GECAs can be directly fused to ORAI1 channels, in the form of a single or two copies^{32, 35}, to enable optogenetic control of Ca²⁺ signaling in cells or tissues with no or little expression of endogenous ORAI proteins.

Collectively, compared to Opto-GPCRs and Opto-RTKs, STIM1-based GECAs directly gate one of the most Ca²⁺-selective channels, the ORAI protein family, and therefore, retain high Ca²⁺ selectivity over other cations³². Nonetheless, one has to bear in mind the caveat that STIM1, and presumably STIM1-derived GECAs too, might cross-talk with other types of Ca²⁺ channels (e.g., certain types of transient receptor potential (TRP) channels and voltage gated Ca²⁺ channels) in selected cell types.

6. Genetically-encoded photoswitchable inhibitors of Ca²⁺ signaling

Complementary to the GECA toolkit, two optogenetic tools capable of suppressing Ca²⁺ entry and/or ER Ca²⁺ release have been generated. To develop a light-inducible OFF-switch for voltage-gated Ca²⁺ (Ca_V) channel, Ma et al. developed an optogenetic construct, termed OptoRGK³⁷ (Fig. 2a), by using a light-sensitive heterodimerization system (iLID/SspB) to control the subcellular localization of a small G protein, Rem_1–266. Rem belongs to the RGK family of G proteins and functions as negative regulator of Ca²⁺ entry mediated by Ca_V channels in excitable cells. OptoRGK has been successfully used to perturb Ca²⁺ oscillations in HL-1 cardiomyocytes, and shows promise to rescue cardiac arrhythmia³⁷. In a second case, Hannanta-anan et al. utilized a similar strategy using the CRY2/CIBN optical dimerizer to recruit regulators of G-protein signaling (RGS) toward PM, thereby reducing the lifespan of GTP-bound α-subunits to terminate agonist-evoked calcium oscillations via inhibition of Gα_q-mediated downstream signaling (including Ca²⁺ mobilization from internal stores) in response to blue light stimulation³⁸ (Fig. 2b). Complementary to the genetic approach, photopharmacological tools building upon azopyrazoles have also been invented to selectively suppress CRAC channel activity in a light-dependent manner³⁹.

Figure 2. Genetically-encoded photoswitchable inhibitors of intracellular Ca²⁺ signaling.

a. Cartoon illustration of OptoRGK to photo-control Ca_V channel activity in cardiomyocytes. An optical dimerizer pair made of LOV2-SsrA (iLID) and SspB is utilized to induce the translocation of Rem1 core domain (a member of the RGK family of GTP-binding proteins) from the cytosol to PM, where it engages the auxiliary beta subunits of Ca_V channel to suppress its activity upon photostimulation.

b. Design of optogenetic regulators of G-protein signaling (Opto-RGS2). In this design, RGS2 is split into two components, including the catalytic box (R2-box) which is critical for inhibition of Gα_q signaling and the N-terminal amphipathic helix (R2-helix). The two split parts are fused with either CRY2_PHR or its binding partner CIBN. In the dark, R2-box is distributed in the cytoplasm. Upon blue light illumination, the forced dimerization between CIBN-R2-helix and CRY2-R2-box triggers reconstitution of the active protein and subsequently inhibit Gα_q signaling and terminates Ca²⁺ release originated from GPCR signaling.

7. Applications of GECAs and wireless optogenetics

Given the strict Ca²⁺ selectivity and high signalling fidelity, STIM1-derived GECAs have been used ex vivo to control Ca²⁺/NFAT-dependent gene expression (Fig. 3a) and cytokine production in cells of the immune system (Fig. 3b). When coupled with the CRISPR activation (CRISPRa) technique⁴⁰, Opto-CRAC can further be repurposed to control the expression of endogenous genes. The calcium-responsive transcriptional reprogramming tool (CaRROT)³⁶, made of NFAT1–460-dCas9-VP64, has been devised to serve this purpose, which allows precise control of transcriptional outputs at user-defined genomic loci, thus expanding the Ca²⁺-responsive effectors to almost any endogenous genes (Fig. 3c). Moving beyond in vitro applications, recent studies have reported encouraging progress in the use of non-opsin-based optogenetics in living animals.

Figure 3. Applications of GECAs and potential solutions for wireless optogenetics.

a. GECAs (e.g., Opto-CRAC) utilized to control Ca²⁺-dependent gene expression using synthetic response elements (RE) derived from various Ca²⁺-responsive transcription factors. NFAT, nuclear factor of activated-T cells; CREB: cAMP response element-binding domain; SRE, serum response element.

b. GECAs used to control Ca²⁺/NFAT-driven cytokine production in T cells in the presence of co-stimulatory signals to activate activator protein 1 (AP-1).

c. Coupling GECAs with CRISPRa to enable light-tunable transcriptional programming. Upon blue light illumination, Opto-CRAC1 induces cytosolic Ca²⁺ influx, which in turn activates CaM/calcineurin to dephosphorylate the N-terminal non-DNA binding fragment of NFAT (NFAT_1–460) and eventually drive the nuclear entry of the fusion protein (NFAT_1–460-dCas9-VP64) – a calcium-responsive transcriptional reprogramming tool (designated as CaRROT). In the presence of small guide RNAs (sgRNAs), CaRROT can be precisely targeted to the promoter regions of any desired genomic loci to photo-tune the expression of endogenous genes.

d. Optogenetic neuromodulation with GECAs (e.g., OptoSTIM1 or monSTIM1). Blue light can be delivered into a specific brain region via implantation of optical fiber (i) or μLED (ii). Photoactivatable Ca²⁺ influx can lead to induction of Ca²⁺-responsive gene expression (e.g., c-Fos) in both neurons and astrocytes in deep-brain regions such as hippocampal and thalamic regions (iii), thereby modulating the behavior (e.g., social fear learning) in awake mice (iv).

e. A nano-optogenetic platform for wireless optogenetic immunomodulation in a mouse model of melanoma. (i) schematic illustration of the core/shell structure of lanthanide doped upconversion nanoparticles (UCNPs), which act as nanoilluminator to emit visible light upon near-infrared irradiation (NIR); (ii-iii) NIR light-controllable activation of Ca²⁺ influx in dendritic cells (DCs) expressing Opto-CRAC1 could boost antigen presentation and DC maturation to more efficiently prime T cells in the draining lymph nodes, thereby facilitating dendritic cell-based immunomodulatory therapy against melanoma.

f. Bioluminescence-aided optogenetic stimulation (BLOGS). (i) Cartoon illustrating the photoactivation of a chimeric luminopsin (LMO) receptor comprising a microbial opsin and Gaussia luciferase (GLuc) or NanoLuc luciferase (NLuc) in the presence of their cognate substrates coelenterazine (CTZ) or furimazine (FZ). (ii) Photochemical reactions catalyzed by GLuc or NLuc. (iii) Schematic depiction of wireless optogenetics using BLOGS in mice administered with luciferase substrates.

7.1. Neuromodulation in awake mice

A bottleneck with the existing optogenetic approach is a situation where the depth of tissue penetration of visible light is shallow (<2–3 mm)⁴¹, requiring optogenetic protocols to use invasive fiber optic probes or microLED to deliver visible light (in the range of 400–500 nm) into tissues (Fig. 3d). These invasive procedures are technically demanding, tend to elicit inflammatory responses, and remain incompatible with chronic experiments. To overcome these issues, monSTIM1 was generated by using a hyper-sensitive CRY2 variant with one mutation and nine additional amino acids in the C-terminus of CRY2_PHR (CRY_2E281A-A9) to boost its clustering capability. The new construct displayed 55-fold increase in light-sensitivity compared to the parental OptoSTIM1²⁹ (Fig 1g). Hence, monSTIM1 was utilized to manipulate Ca²⁺ signaling and activate Ca²⁺-dependent gene expression in deep-brain regions (hippocampal and thalamic regions), and to boost the social learning capability of awake mice without invasive surgical procedures²⁹ (Fig. 3d). However, upon non-invasive light stimulation, how deep the tissues can be reached to effectively photoactivate monSTIM1 remains elusive. Furthermore, blue light emitting at 450–470 nm may generate local heat and oxidative stress while STIM1 is also sensitive to both heat and ROS²⁷. It remains to be rigorously validated if the observed behavior changes are indeed caused by Ca²⁺ signals per se rather than undesired side effects of photostimulation. Another attractive strategy to overcome this bottleneck is to exploit optogenetic modules that can be switched on and off by NIR that has a greater depth of tissue penetration (up to centimeters)⁴².

7.2. Immunomodulation in living animals

STIM1-based GECAs have been successfully applied in vivo to control the motility and function of cells of the immune system. eOS1 has been employed to manipulate Ca²⁺ signals in single T cells and photo-tune T cell migration, adhesion and chemokine release in vivo via two-photon-based photoactivation³⁰. Although two-photon optogenetics provide a better spatial specificity and deep penetration in scattering tissue⁴³, potential heat generation during photostimulation and special equipment requirements make it non-ideal for daily use in the laboratory setting, let alone further clinical applications. A wireless nano-optogenetic approach based on NIR-excitable upconversion nanoparticles (UCNPs; Fig. 3e) has been developed to accelerate in vivo application using immune cells. The most significant advantage of UCNPs is their unusual inverse excitation and emission profiles: UCNPs are excited using low power (mW) and deep tissue-penetrating (1–2 cm) NIR light. This low energy input is efficiently converted to a higher energy output emission at diverse shorter wavelengths, including blue light emission at 470 nm that can activate most optogenetic constructs (Fig. 3e). Since UCNPs can be illuminated from NIR sources located outside the body, they can act as “nano-illuminator relays” to activate optogenetic constructs in a wireless manner. The nano-optogenetic concept has been validated in a dendritic cell-based immunomodulatory therapy against tumor in a mouse mode of melanoma³² (Fig. 3e).

8. Conclusions

In summary, a variety of optogenetic tools, based on both opsin or non-opsin photoreceptors, have been crafted to enable remote control of intracellular Ca²⁺ signaling both in cellulo and in living organisms (summarized in Table 1). Most tools are engineered from membrane receptors (GPCRS and RTKs) and signaling proteins (CaM, G proteins and STIM1) that are directly or indirectly involved in Ca²⁺ signaling. Among these tools, GECAs inspired by STIM1 are regarded as the most Ca²⁺ selective and have been most widely applied to modulate the central nervous system and the immune system. The in vivo application of these tools will likely be accelerated by developing NIR-responsive optogenetic modules and combining existing tools with nanotechnologies or bioluminescence to enable in vivo optogenetics. The feasibility of bioluminescence-aided optogenetic stimulation (Fig. 3f), at the cost of partially sacrificing the spatial resolution, has been validated in several recent studies^44–46. These innovative platforms might represent alternative solutions for future wireless optogenetics.

Highlights.

• Overview of optogenetic tools to control Ca²⁺-modulated physiological processes

• Engineering strategies for genetically-encoded Ca²⁺ channel actuators (GECAs)

• Guidelines for selecting appropriate GECAs to achieve tailored function

• Insights into potential solutions for wireless optogenetics in vivo