Principles and applications of optogenetics in developmental biology

Abstract

The development of multicellular organisms is controlled by highly dynamic molecular and cellular processes organized in spatially restricted patterns. Recent advances in optogenetics are allowing protein function to be controlled with the precision of a pulse of laser light in vivo, providing a powerful new tool to perturb developmental processes at a wide range of spatiotemporal scales. In this Primer, we describe the most commonly used optogenetic tools, their application in developmental biology and in the nascent field of synthetic morphogenesis.

Keywords: Embryonic development; Optogenetics; Signaling; Synthetic biology; Tissue morphogenesis.

Fig. 1. Approaches to controlling protein activity using optogenetics.

In all panels, photoreceptors are depicted in blue, their binding partners in green and proteins of interest in gray. (A) Light-induced protein dimerization can be used to recruit a protein of interest to a specific intracellular location, where it can pursue its function. (B) Light-dependent oligomerization (clustering) can induce active functional signaling hubs or inhibit protein function. (C) Light-induced dimerization can also be adopted to sequester a protein of interest away from its site of action. (D) Photo-uncaging based on LOV domains can be used to directly control protein activity with light.

Fig. 2. Overview of LOV domain-based optogenetic systems.

LOV domains are the most versatile group of photoreceptors. In all panels, LOV domains are depicted in blue, heteromeric interacting partners in green and proteins of interest in gray. (A) The LOV core domain is composed of a Per-Arnt-Sim (PAS) domain that binds to a flavin co-factor and with which it forms a covalent bond upon blue-light illumination. This causes unfolding of an α-helical connector element (e.g. LOV2 Jα). (B) Light-induced conformational changes of LOV domains can be used to photo-uncage a protein of interest and stimulate its activity. (C) The Jα-helix can also be engineered to mask a protein motif that becomes exposed upon light-induced unfolding. (D-H) A variety of different optogenetic dimerization systems are based on LOV domains. These include iLID (D), TULIP (E), TAEL (F), Vivid (VVD) (G) and magnets (H). Whereas iLID and TULIP function by unmasking a protein-interaction domain (e.g. SsrA, LOVpep) upon photoactivation that can be bound by a specific interactor (e.g. SspB, ePDZ) (D,E), TAEL and VVD undergo light-induced homodimerization (through either an adjacent dimerization domain or through the light-responsive N-terminal Ncap fold, respectively) (F,G). (H) Magnets are derived from VVD and engineered to undergo heterodimerization.

Fig. 3. LOV domain-based optogenetic manipulation of animal development.

In all panels, LOV domains are depicted in blue, heteromeric interacting partners in green, protein of interests in gray, and an engineered recognition motif in yellow. (A,B) In this example, the iLID heterodimerization system is used to recruit the Ras-GEF Sos to the plasma membrane and activate Erk signaling upon blue-light illumination during early Drosophila embryogenesis. By varying the temporal pattern and intensity of light activation, cell signaling and tissue patterning can be controlled. (C,D) The TAEL homodimerization system has been applied in zebrafish embryos to induce gene expression upon light activation. Light-dependent conformational changes in TAEL cause homodimerization and DNA binding of the dimer to a specific promoter region (dark blue) triggering gene expression. (E,F) A Jα of AsLOV2 engineered to contain a nuclear localization signal (NLS) that is exposed only upon light-induced Jα unfolding causes target proteins to shuttle into nuclei. Similarly, the LANS system has been used in C. elegans to induce nuclear shuttling of the transcription factor Lin1. (G,H) The small GTPase Rac1 can be photo-caged using the AsLOV2 domain (PA-Rac1). Upon light-induced unfolding of the LOV domain, PA-Rac1 becomes active inducing remodeling of the actomyosin network (red) and lamellipodia formation (pink). PA-Rac1 has been used in Drosophila oocytes to guide the movement of border cells using light.

Fig. 4. Cryptochrome- and phytochrome-based optogenetic regulation of tissue morphogenesis.

In all panels, CRY2 is depicted in blue and PHYB in red, the respective interaction partners in green, and proteins of interest in gray. (A) Upon blue-light illumination, the photosensitive protein CRY2 undergoes a conformational change and binds to its interaction partner CIBN. (B) The cryptochrome system has been applied to recruit the phosphoinositide phosphatase OCRL (tagged with CRY2) to the plasma membrane by triggering the light-dependent interaction of CRY2 with a membrane-anchored CIBN during Drosophila gastrulation. At the plasma membrane, OCRL depletes PI(4,5)P₂ (PIP₂, yellow circles), which function as anchoring points for cortical actin fibers (red lines). This results in an inhibition of actomyosin contractility and apical constriction during tissue invagination. (C) CIBN/CRY2-mediated recruitment of RhoGEF2 to the apical plasma membrane has been used to trigger Rho signaling upon light exposure culminating in myosin-II-dependent apical constriction and tissue invagination during Drosophila embryogenesis. (D,E) The phytochrome system consists of the PHYB photoreceptor, which binds to its interaction partner PIF upon red-light illumination. Far-red illumination causes the PHYB/PIF interaction to dissociate, making the optogenetic system reversible. PHYB activity depends on the plant-specific co-factor PCB. PHYB, anchored at the plasma membrane, recruits the cell polarity determinant Pard3 (fused to PIF) to specific plasma membrane domains (highlighted in purple) upon red light illumination (red region). Illumination of the entire cell with far-red light (pink region) allows PHYB/PIF complex formation only in the region that was simultaneously illuminated with red light. With this strategy it is possible to induce asymmetric inheritance of Pard3 during cell division.

Fig. 5. Subcellular optogenetics.

Three different approaches have been so far employed to achieve subcellular photoactivation (activated photoreceptor colored in blue, membrane-anchored components in green). Left: Using the PHYB photoreceptor anchored uniformly at the plasma membrane, it is possible to locally photoactivate a subcellular region. Red light-induced PHYB/PIF6 dimer formation is approximately seven times faster than far-red light-induced dissociation. Stimulation of a subregion of interest using red light (red region) with simultaneous deactivation of the whole cell using far-red light (pink region) results in locally restricted photoactivation. Middle: Two-photon excitation using near-infrared light enables locally restricted light delivery (blue blurred line) and photoactivation deep inside living tissues by temporally and spatially restricting the laser light to a focal volume in the femtoliter range. Right: Subcellular optogenetic activation can also be achieved in tissues of complex morphology by engineering an optogenetic anchor in such a way that it localizes only to the site of the cell where optogenetic activation is desired (green cell outline). Components of the cell polarity machinery are ideal candidates for designing optogenetic anchors, as recently demonstrated by the use of PatJ to manipulate myosin-II activity specific at the cell base during Drosophila gastrulation. Using this approach, even whole-cell photoactivation results in a locally confined activation of the optogenetic system. See text for more details.

Fig. 6. Reconstructing morphogenesis using synthetic biology approaches.

(A) Morphogenesis relies on a common set of mechanisms (modules) involving changes in cell behaviors that occur at specific time points and locations, and that give rise to highly complex forms and patterns. (B) By enabling the delivery of precise spatiotemporally controlled inputs, optogenetics allow individual modules to be triggered at will and determine the minimum set of requirements sufficient to drive morphological remodeling. Computerized feedback control could be used to automatically tune optogenetic inputs in real time according to the desired morphogenetic outputs (synthetic morphogenesis). Such an experimental set-up has been recently developed to achieve robust perfect adaptation (RPA) (Aoki et al., 2019) of gene expression in single Saccharomyces cerevisiae yeast cells (Rullan et al., 2018). This combination of optogenetic and control theory concepts should allow us to eventually reconstruct complex morphogenetic processes and build synthetic embryos. The embryos depicted in this figure represent the corral Monoxenia darwinii during gastrulation as drawn by Ernst Haeckel (Haeckel, 1891).