Detection of neuropeptides in vivo and open questions for current and upcoming fluorescent sensors for neuropeptides

During a stress response, various neuropeptides are secreted in a spatiotemporally coordinated way in the brain and in the periphery. For a precise understanding of peptide functions in a stress response, it is important to investigate when and where they are released, how they diffuse, and how they are broken down in the brain. In the past two decades, genetically encoded fluorescent calcium indicators have greatly advanced our knowledge of the functions of specific neuronal activity in regulation of behavioral changes and physiological responses during stress. Recently, various kinds of structural information on G-protein-coupled receptors (GPCRs) for neuropeptides have been revealed. Genetically encoded fluorescent sensors have been developed for detection of neurotransmitters by making use of conformational changes induced by ligand binding. In this review, we summarize the recent and upcoming advances of techniques for detection of neuropeptides and then present several open questions that will be solved by application of recent or upcoming technical advances in detection of neuropeptides in vivo.

During a stress response, various neuropeptides are secreted in a spatiotemporally coordinated way in the brain. For a precise understanding of peptide functions in a stress response, it is important to investigate when and where they are released, how they diffuse, and how they are broken down in the brain. In the past two decades, genetically encoded fluorescent calcium indicators have greatly advanced our knowledge of the functions of specific neuronal activity in regulation of behavioral changes and physiological responses during stress. In addition, various kinds of structural information on G-protein-coupled receptors (GPCRs) for neuropeptides have been revealed. Recently, genetically encoded fluorescent sensors have been developed for detection of neurotransmitters by making use of conformational changes induced by ligand binding. In this review, we summarize the recent and upcoming advances of techniques for detection of neuropeptides and then present several open questions that will be solved by application of recent or upcoming technical advances in detection of neuropeptides in vivo.

Response to stress and release of neuropeptides
Various peptides are secreted within the brain to cope with stress [1][2][3] (Table 1). Corticotropin-releasing hormone (CRH) and vasopressin are essential for the coordination of behavioral and metabolic responses to stress [1]. These two neuropeptides regulate the hypothalamic-pituitary-adrenocortical (HPA) axis and induce the release of glucocorticoids from the adrenal cortex into blood. Glucocorticoids reach every organ by blood circulation and regulate brain and body functions. The mineralocorticoid receptor (MR) and the glucocorticoid receptor (GR) comprise the receptor system that mediates the genomic actions of glucocorticoids. In the recovery phase of an adaptive stress reaction, GR monomers repress peptide synthesis of CRH and vasopressin. Corticosteroid receptors and neuropeptides mutually affect each other [4,5].
There are many neuropeptides other than CRH and vasopressin that are deeply involved in stress responses and mood disorders such as depression. Neuropeptide Y (NPY) has unique stress-relieving, anxiolytic and neuroprotective properties [6,7]. Although oxytocin in the hypothalamic paraventricular nucleus and the supraoptic nucleus is well known for its role in sociality, oxytocin neurons are strongly activated by social defeat stress or noxious stimuli [8,9] and exert anxiolytic and analgesic effects [10,11]. Orexin in the lateral hypothalamic area is regarded as a major regulator of sleep/wakefulness [12,13], but it is also involved in stress responses including autonomic changes and nociception [14][15][16]. Substance P is an 11-residue neuropeptide that is involved in nociception, stress regulation, and autonomic control including control of body temperature [17,18]. Endogenous opioid peptides such as dynorphins and enkephalins act on opioid receptors, and opioid systems contribute to not only addiction but also the acute and delayed molecular and behavioral effects of stress [19,20]. In this review, we do not go into detail about each of the neuropeptide functions, but we focus on the common features of neuropeptide release and how to detect neuropeptides in vivo.

Neuropeptides as neuromodulators
In the brain, neuropeptides modulate information processing. They are released from subsets of neurons located in various brain regions such as the hypothalamus and act on neurons by binding to their specific receptors. Although many neuropeptides are produced in the hypothalamus and brainstem, cortical neurons are also subject to the strong influence of neuropeptides. A recent study showed that transcripts of neuropeptide precursor and neuropeptide-selective GPCR genes are abundant in almost all cortical neurons [21]. Neuropeptides function as neurotransmitters, but there are several differences between neuropeptides and classical neurotransmitters such as glutamate or GABA. In principle, most neuropeptides act at GPCRs, while ionotropic receptors as well as GPCRs mediate the actions of glutamate and GABA [22,23]. Transmission by glutamate or GABA has a spatial specificity that is achieved by the structure of synapses and temporal specificity conferred by tight coupling to electrical activity and rapid clearance. In contrast, neuropeptides are delivered more roughly but can induce coherent behavioral changes.
The half-lives of neuropeptides in the brain are relatively long compared with those of glutamate and GABA. For example, the half-lives of oxytocin and vasopressin are approximately 20 min in the brain and 2 min in the blood [24]. Orexin-A is more stable because of post-translational modifications of both termini and two intra-chain disulfide bonds, and its half-life in plasma is 27 min [25]. These features of neuropeptides contribute to their unique actions such as volume transmission [26,27]. Inactivation of neuropeptides is caused by internalization after receptor binding or degradation by extracellular proteases, which in some cases can actually create different bioactive peptides. The time scale and functional regulation of this inactivation for most peptides remain unclear.

Releasing points of neuropeptides
Secretion sites of neuropeptides are not restricted to presynaptic axon terminals. Neuropeptides are packaged in large dense-core vesicles, while glutamate and GABA are packaged in small synaptic vesicles [27]. These two types of vesicles are stored differently, and large dense-core vesicles are released from all parts of a neuron including dendrites. It is also known that dendritic release of neuropeptides can be regulated differently from axonal release. For example, α-MSH evokes dendritic release of oxytocin, while it inhibits the electrical activity of oxytocin neurons and reduces the axonal secretion of oxytocin into the blood [28]. Most neuropeptides act on GPCRs, and they have high affinity for their receptors. Neuropeptides act not only as general messengers to transmit information to various brain regions at the same time but also send specific information depending on brain regions of projection sites where peptides are released. The same neuropeptide from different secretion sites can function differently. For example, somatostatin released into the portal blood supply of the median eminence from nearby hypothalamic neurons can decrease growth hormone secretion from the pituitary gland, while somatostatin-synthesizing neurons in the cortex have no functional relation to hormone regulation [22]. Our understanding of the functional organization of neural circuits is derived from the recent availability of advanced technologies using optogenetics to probe synaptic transmission at projection sites [29,30]. However, in the case of MCH neurons, the expression pattern of the MCH-1R is apparently more widespread than that of MCH-containing terminals. If you express channelrhodopsin fused with EGFP in MCH neurons and perform immunohistochemistry to depict MCH peptide in MCH neurons, you will easily realize that there are many "blank" fibers with EGFP but without MCH peptide. In the case of oxytocin, the olfactory bulb is an important site for oxytocin-dependent behaviors and is rich in oxytocin receptors but only sparsely innervated by oxytocin fibers [31,32]. These findings indicate the possibility that alternative modes of communication might be involved in the effects caused by neuronal activation other than synaptic wiring transmission.

Neuropeptides in the periphery
Many neuropeptides are released from neurons in the brain, but some of their actions seem analogous to hormones in peripheral tissues. Vasoactive intestinal peptide (VIP) of cortical GABAergic interneurons dilates vessels, whereas neuropeptide Y (NPY) and somatostatin induce vasoconstriction [33]. Oxytocin and vasopressin are well known as hypophysial hormones. Oxytocin receptors are enriched in the uterus and the mammillary gland, and they are required for uterus contraction during labor and milk ejection in response to suckling by a child [34]. Vasopressin V2 receptor (V2R) is expressed in renal principal cells in the kidney. In a state of hypernatremia or hypovolemia, vasopressin is released from the pituitary and induces transportation of aquaporin-2 to the apical plasma membrane [35].Vasopressin is delivered through blood circulation, and therefore monitoring blood concentration is useful for investigating renal functions. On the other hand, neuropeptides including oxytocin, VIP and NPY are also expressed in neurons in the enteric nervous system [36][37][38]. Considering these local neuropeptides, spatiotemporal measurement techniques for specific neuropeptides are required to investigate their physiological roles in the periphery. Table 1 Stress-related neuropeptides, their direct detection, and their receptors. The table lists neuropeptide precursor protein (NPP) genes, predicted neuropeptides, direct detection reports, each neuropeptide length, neuropeptide-selective G-protein-coupled receptor (NP-GPCR) genes, neuropeptide receptors, GPCR class, and primary Gα family of NP-GPCRs. Detection references show reports of direct detection for each neuropeptide.

Methods for detecting neuropeptides in vivo
In this chapter, we explain several techniques for measurements of neurotransmitters including neuropeptides (Fig. 1), and we highlight the fundamental principles and potential advantages of genetically encoded fluorescent sensors for neuropeptides.

Microdialysis
Intracerebral microdialysis is a well-established technique for monitoring the neuropeptide concentration in the extracellular environment in the brain [39][40][41][42][43][44][45][46][47], although it is still challenging because of the low concentrations of peptides in dialysates. The probe membrane of microdialysis is semi-permeable, and microdialysis can be used to collect/deliver molecules from/to the extracellular fluid in living animals [48]. This feature is a unique advantage of microdialysis because it enables investigation of the neuropeptide secretion while stimulating the adjacent tissue by the perfusate including a bioactive substance [48]. It is also possible to combine optogenetics with microdialysis to detect optically evoked peptide release in vivo [49]. The collected solution is analyzed by an immunoassay or HPLC coupled with mass spectrometry. In some cases, several molecules can be measured from the same samples at the same time [50]. Using microdialysis, it was suggested that dendritic and axonal release of oxytocin and vasopressin can be temporally dissociated [51]. However, the sampling rate of microdialysis is relatively slow, and it typically takes several minutes or longer (approximately 30 min for neuropeptides). Microdialysis is not suitable for detecting rapid changes of neuropeptide concentrations in a second or sub-second time scale.

Fast-scan cyclic voltammetry
Fast-scan cyclic voltammetry (FSCV) is an electrochemical technique that was developed to measure changes in electroactive neurotransmitters [52]. In brief, fast real-time detection of neurotransmitters can be achieved by monitoring the changes in electrical responses that are induced by the redox reactions of analytes on the surface of a carbon microelectrode. Although FSCV has been mainly used to detect electroactive monoamine transmitters such as dopamine [53,54], noradrenaline [55], serotonin [56], and histamine [57], this technique is also applicable to a fraction of neuropeptides that show electrochemical oxidation on the surface of the electrode. It should be noted that traditional FSCV has limits of detection for dopamine in the low nM range, but neuropeptides are present at even lower concentrations. This means that some improvements are required to detect small faradaic signals from background drift and noise in vivo. Recently, a modified sawhorse waveform consisting of multiple scan rates was proposed for the detection of the opioid neuropeptide methionine-enkephalin [58]. However, detection of most peptides in the brain is still not possible by FSCV. In addition, FSCV requires implantation of a relatively large probe (approximately 70-300 μm in diameter) into brain tissue. These features of FCSV limit the achievement of spatially precise measurement of neuropeptides in living animals.

Genetically encoded fluorescent sensors
For a clear understanding of spatial and temporal regulation of neuropeptides, it is essential to monitor the dynamics of neuropeptides in the brain at a cellular resolution with a sub-second time scale. Specific measurement of a neuropeptide concentration with a high spatiotemporal precision in free-moving animals cannot be achieved by microdialysis or FSCV. Recent advances in the development of genetically encoded fluorescent sensors have a potential to present a solution for this longstanding problem.
In the past two decades, the use of genetically encoded fluorescent calcium indicators (GECIs) such as GCaMP and Cameleon has resulted in enormous advances in neuroscience. It has enabled comprehensive recordings of neuronal activities in specific neuronal circuits in animals exhibiting complex behaviors such as feeding [59], memory formation [60], and social interactions [61]. Similarly, genetically encoded fluorescent sensors for extracellular neurotransmitters have recently emerged as powerful tools for direct measurement of neurotransmitter dynamics. First, fluorescence probes that monitor the activity of a neurotransmitter receptor using Förster resonance energy transfer (FRET) were generated ( Fig. 1(3)a). Martin Lohse's group reported α2A-adrenergic receptor and parathyroid hormone receptor-based FRET probes [62]. Unfortunately, these GPCR-based FRET sensors have two disadvantages: (1) the FRET signal is extremely small and (2) the sensor, which contains two fluorescent proteins, shows poor plasma membrane localization. Although fluorescein arsenical hairpin binder (FlAsH) has been developed to improve the expression and delivery of these sensors [63,64], the signal is still too weak for in vivo application. Yulong Li's group and Lin Tian's group have recently reported genetically encoded GPCR-based intentiometric fluorescent sensors for neurotransmitters ( Fig. 1(3)b) such as acetylcholine [65], dopamine [66,67], and noradrenaline [68]. To detect binding of the ligand to the receptor, cyclic permutated EGFP (cpEGFP), a derivative of EGFP with fluorescence properties that are highly sensitive to external environments, was inserted into the third intracellular loop of seven-transmembrane GPCRs, where a large conformation change is expected following receptor activation. In addition, yellow and red fluorescent sensors for dopamine have recently been developed by using a similar method [69]. Expansion of the color pallet of the fluorescent probes will enable simultaneous use of the sensors with either other genetically encoded sensors or optogenetic tools. Compared with the previous FRET-based sensors, intentiometric neurotransmitter sensors have a much higher signal dynamic range. For example, the first generation of green-fluorescent dopamine sensors such as GRAB DA and dLight1 enabled measurement of dopamine dynamics not only in vitro but also in various model organisms such as flies, fish, and mice with a millisecond-time resolution. Since the receptors for most neuropeptides are GPCRs, development of genetically encoded fluorescent sensors for neuropeptides is promising. Indeed, small fluorescence responses were observed in a few instances such as enkephalin and oxytocin [67,70] There will be various GRAB sensors for neuropeptides in the near future.
The third group of fluorescent sensors utilize downstream pathways of GPCR activation. Ligand binding to GPCRs activates phosphorylation of the GPCR and results in binding of β-arrestin to the phosphorylated receptor. The Tango assay (named after the association of two proteins to stimulate a response) was designed to transmitβ-arrestin binding after GPCR activation into a stable expression of reporter genes [71,72]. In this system, the transcription factor tTA is fused to the C-terminal domain of the GPCR by a peptide sequence containing a tobacco etch virus (TEV) protease cleavage site. This construct can be cleaved when TEV protease-fused β-arrestin is recruited upon GPCR activation induced by ligand binding. Cleaved tTA translocates to the nucleus and initiates stable expression of a reporter gene. Recently, an improved iTango and its simplified version iTango2 have been developed [73]. In this system, two modifications were added: (1) TEV protease is split into two fragments (TEV-N and TEV-C) and (2) the TEV protease cleavage site is exposed only when the fused LOV2 domain is activated by blue light illumination ( Fig. 1(3)c). These modifications reduce background signals and yield better temporal resolution for detecting neuromodulators. By signal amplification, the Tango assay and its related methods can provide single-cell resolution, nanomolar sensitivity for specific neuropeptides such as oxytocin [74] and vasopressin [71]. However, the temporal resolution of the Tango assay is poor because of their principles utilizing transcription for signal amplification cascades.
Although these GPCR-based sensors provides spatial information that cannot be obtained by FSCV, and genetically encoded sensors are suitable for monitoring specific neuromodulators in living animals, several limitations should be considered. GPCR-based fluorescent sensors are difficult to express, and their fluorescent intensity is weak compared with that of GECI because of their seven-transmembrane structure. The dynamic range of fluorescent sensors might be too small for detecting dynamic changes in vivoin some cases. Virus vectors are very useful for expression of these sensors, but the strong ectopic expression of transmembrane proteins sometimes induces aggregation within the cell and reduces efficient ΔF/F 0 of fluorescent sensors. It is important to confirm that the disturbance of endogenous signaling by fluorescent sensors is carefully minimized. Nevertheless, genetically encoded fluorescent sensors provide high specificity, single-cell spatial resolution, and physiologically relevant temporal resolution.
Several optical devices can be used for recording of fluorescent imaging. Fiber photometry is a simple way to detect temporal changes of total fluorescent intensity ( Fig. 1(3)x). It can be used to detect orexin neuronal activation during nociception [14]. A microdialysis study in the human amygdala has shown that the concentration of orexin peptides is low during periods of pain despite high levels of arousal [42]; however, the temporal resolution of microdialysis is insufficient for detecting real-time transient changes in the activity of orexin neurons. On the other hand, the temporal resolution of a fiber photometry system is sufficiently high for detection of a transient fast change of neuronal activity during physiological responses. It should be noted that fiber photometry receives an integrated signal of the whole neuronal population around the fiber tip. This limits the spatial resolution, which is one of the great advantages of fluorescent sensors. Simultaneous measurement of multiple brain regions using multiple optical fibers is one of the solutions for this problem [75]. A miniature head-fix microscope is another powerful candidate for recording of optical imaging of fluorescent sensors (Fig. 1(3)y) [76]. The use of transparent organisms such as zebrafish might be suitable for comprehensive analysis of deep brain neuronal populations including neuropeptidergic cells [77]. Two-photon microscopy can achieve maximum resolution, while the large and complex components limit free moving of target animals ( Fig. 1(3)z).

Future perspectives and open questions for coming fluorescent sensors
As shown by the fact that Hilbelt's problems greatly advanced 20thcentury mathematics, open questions are important for subsequent research in the field. Here we present several open questions on neuropeptides that might be solved by application of recent or upcoming technical advances in detection of neuropeptides in vivo.

What is the volume transmission of a neuropeptide like?
The term "volume transmission" has been widely used to describe the unique features of neuropeptides; however, no one has ever visualized the volume transmission in vivo. Although it represents an important mode of neuronal communication, the mechanism of volume transmission has not been extensively investigated. Dendritic secretion of neuropeptides should be an important research target that can be powerfully investigated with fluorescent probes of neuropeptides. Dendritic release in vivo sometimes does not correlate with axonal release [27]. Although changes in extracellular peptide concentrations have been measured in many brain areas in response to physiological stimuli, it is still not clear if these changes reflect local release from axonal terminal or diffusion of dendritic release from distant sites.
Visualization of the difference between neuropeptide concentrations inside and outside of blood vessels is also an interesting topic of studies with fluorescent sensors. Confocal live imaging of blood vessels in the brain will show the regulation of peptide transport through the bloodbrain barrier and will contribute to a precise understanding of neurodegenerative diseases related to deficiencies in regulation of the bloodbrain barrier.
Sleep affects the interstitial space in the brain and drives metabolite clearance in mice [78] and humans [79]. Convective fluxes of interstitial fluid increase the rate of β-amyloid clearance during sleep, and CSF dynamics are interlinked with neural and hemodynamic rhythms in the brain [79]. Therefore, it is reasonable to assume that the sleep/wakefulness state modulates the diffusion pattern of secreted neuropeptides and their actions on physiological responses. Wide-field observations using fluorescent sensors of neuropeptides will provide valuable information on the regulation of volume transmission of various neuropeptides in living animals.

How does an intranasally administered neuropeptide reach the brain?
Intranasal administration is widely used to convey neuropeptides that cannot easily go through the blood-brain barrier in humans [80,81]. Experimental and therapeutic use of neuropeptides in humans greatly relies on intranasal administration because systemically administered neuropeptides do not easily pass through the blood-brain barrier and sometimes evoke side effects when circulating in the blood. Oxytocin is regarded as one of the treatment candidates for neuropsychiatric disorders such as autism [82], mood disorders [81], and addiction [83]. So far, intranasal administration of oxytocin has been reported to increase brain oxytocin levels in rodents [84,85] and monkeys [86]. However, enormous doses that exceed the total amount of oxytocin in the pituitary, diverse peripheral actions, and discredited detection methodology are still criticized [87,88], and it is still not clear how and where oxytocin reaches to the brain after intranasal administration. Lee et al. directly showed that deuterated oxytocin administered via the intranasal route reaches the brain, but their study had limitations in spatial and temporal resolution [86]. The development of fluorescent sensors of oxytocin will make it possible to visualize the propagation of intranasally administered oxytocin in the brain. Recently, simultaneous calcium measurements using multiple optic fibers inserted in different brain regions revealed that different populations of VTA dopamine neurons are encoding rewarding or aversive stimuli depending on the projection target [77]. Similar investigation will shed light on the difference in the propagation pattern of oxytocin between intranasal administration and endogenous oxytocin release.

What type of neuropeptides are co-released by neuronal activation?
It is well known that neurons expressing neuropeptide Y (NPY) in the arcuate nucleus also express agouti-related protein (AgRP) and GABA, whereas neurons expressing proopiomelanocortin (POMC) in the arcuate nucleus express cocaine-and amphetamine-regulated transcript (CART) peptides. Their mutually coordinated regulations play pivotal roles in feeding behaviors and energy homeostasis [89]. Orexin neurons in the lateral hypothalamic area contain not only orexin but also other neuropeptides such as dynorphin and neurotensin [14,90]. Orexin and dynorphin are packaged within the same vesicles [91]. Oxytocin neurons in the supraoptic nucleus have been shown to produce various neuropeptides including CART [92], endocannabinoids, glutamate and nitric oxide. These findings suggest that co-transmission of multiple neuropeptides is a very common phenomenon and that neuronal activation using optogenetics or chemogenetics should be carefully interpreted considering the effect of co-transmitters. Therefore, simultaneous recording of multiple molecules using multicolor fluorescent sensors will be useful for clarification of co-transmission released. Multicolor genetically encoded fluorescent calcium indicators such as XCaMPs [93] are good guideposts for upcoming fluorescent sensors for neuropeptides.
In addition, it is important to note that some neuropeptides exert their actions through multiple receptors. Oxytocin and vasopressin are nonapeptide products of genes derived by duplication of a common ancestral gene, and they have similar structures with a disulfide bridge between Cys residues 1 and 6 [34]. Oxytocin receptors and V1aRs are expressed in many brain regions, but their distributions are different. The affinities of vasopressin to vasopressin or oxytocin receptors are similar [94], and non-synaptic release of oxytocin and vasopressin thus has the potential to result in substantial crosstalk among their receptors [95]. It was reported that oxytocin can act via V1Rs and influence social behaviors [96]. V1aR activation facilitates social recognition [97]. These findings imply that it is important to monitor both oxytocin and vasopressin concentrations for a clear understanding of their overall functions in vivo.

How does the size of the brain affect the transmission of neuropeptides?
Body size affects the homeostasis and metabolism. The mouse brain weighs about 0.4 g, the human brain weighs about 1500 g, and the blue whale brain weighs about 9000 g [98]. Considering these huge differences in brain size, is it reasonable to presume that the volume transmissions of neuropeptides are used in the same way? Not only passive diffusion but also an active transport system might be involved in the volume transmission in a large brain. Comprehensive recordings with a wide optical field might reveal the different regulations of volume transmission among species. Intravenous injection of an AAV vector can induce a sufficient amount of expression in the brains of mice [99], rats [100], and even non-human primates [101]. Given that sparse expression is still useful for optical observation [102], intravenous injection of AAV vectors expressing genetically encoded fluorescent probes of neuropeptides can be used for optical observation in a wide variety of animals including non-laboratory model animals.

What is the intracellular dynamics of peptides like?
A neuropeptide precursor protein is translated into the lumen of the rough endoplasmic reticulum and is cleaved into one or more neuropeptide products during or after transportation to the Golgi complex. Genetically encoded fluorescent sensors for neuropeptides might be used to investigate intracellular dynamics of peptides. Ligand-binding sites of GPCRs are located on the extracellular side of the seventransmembrane protein. It might be possible to express GPCR-based sensors within the endoplasmic reticulum (ER) or mitochondria since it is known that intracellular GPCRs such as mGlu 5 and MT 1 play some roles in synaptic plasticity in the ER [103]. Delta opioid receptors are localized in the membrane of large dense-core vesicles containing calcitonin gene-related peptide [104]. Although it is not guaranteed that intracellular GPCRs maintain their affinity for ligands and conformational properties, live imaging of fluorescent sensors for neuropeptides has the potential to investigate the processing of neuropeptides during transportation of prerpropeptides within vesicles by detecting mature target peptides. We might find some differences between peptide processing in the ER and that in mitochondria. Spatial and temporal high resolution of genetically encoded fluorescent sensors will be fully utilized in such high-resolution analyses of intracellular processes.

Concluding remarks
As described in this review, we are facing the decade of rapid development of techniques for detection of neuropeptide dynamics. Advances in technology lead to scientific discoveries. Structural information obtained by electron cryomicroscopy has revealed a molecular mechanism for simultaneous achievement of water permeability and proton impermeability [105]. Calcium indicators have revealed various kinds of intracellular signaling such as calcium oscillation [106][107][108]. We strongly hope that recent and coming technologies for detection of neuropeptides will lead to qualitatively different discoveries in neuroendocrinology that are not possible by traditional technologies. In addition, great advances in structural biology have provided much information on various GPCRs. Recently, the crystal structure of the human oxytocin receptor was reported [109]. Given that structural information on channelrhodopsin [110] enabled efficient modification of channelrhodopsin based on the conformational mechanism [111], it is reasonable to expect rapid development of GPCR-based genetically encoded neuropeptide sensors. Simultaneous application of optical actuators and fluorescent sensors of neuropeptides will reveal the cause-and-effect relationships between specific neuropeptides and behavioral changes.