Acute hippocampal slices, stereotaxic surgery for viral injection, and LED implant protocols and animal procedures for mouse behavioral assays were approved by the animal care committees at The Centre for Phenogenomics (TCP; Toronto, ON, Canada) and University of Toyama (Japan). Organotypic slice cultures of hippocampus were prepared from postnatal day 6–7 Sprague Dawley rat pups as previously described (Stoppini et al., 1991; Okamoto et al., 2004) and in accordance with the guidelines of the University Health Network animal care committee approved protocol and oversight (TCP, Canada).

BlgC (Ryu et al., 2010): human codon-optimized with S27A mutation to reduce dark activity (Stierl et al., 2014) was fused with RFP (tdTomato) (Shaner et al., 2004) at the N-terminus and subcloned into pCAG plasmid vectors (Niwa et al., 1991). The cGMP FLIM probe (cGiYR: cGMP indicators with YFP/RFP) was constructed from the Gln⁷⁹—Tyr³⁴⁵ in cGMP-dependent protein kinase I (cGKI) (Russwurm et al., 2007) fused with YFP (Citrine) (Heikal et al., 2000) as a FRET/FLIM donor and RFP (tdTomato) as a FRET/FLIM accepter into pCAG plasmid vector. The cAMP FLIM probe (REY: RFP-Epac-YFP) (Luyben et al., 2024), and a control YFP (Citrine) were also subcloned into pCAG plasmid vectors.

AAV1-CaMKIIα promoter-mGFP-BlgC (titer: 2.07 × 10¹³ VG/mL for behavioral assays, 2.11 × 10¹³ VG/mL for electrophysiology) was packaged and purified by SignaGen laboratories (SL100863 and SL100864, Rockville, MD, United States).

For in vitro cGMP detection by two-photon FLIM, HEK293 cell lysates expressing cGMP FLIM probe (cGiYR) were placed in a microscope chamber and we measured the dose-dependent lifetime change by serially adding 8-Br-cGMP and comparing with a control cell lysate which express YFP alone using a two-photon FLIM microscope equipped with a time-correlated single photon counting (TCSPC) system (PicoHarp 300, FLIM upgrade kit for Olympus FV1000MPE, SymPhoTime software; PicoQuant, Berlin, Germany) using a single plane of YFP fluorescence (520–542 nm) images (donor of cGiYR: 900 nm excitation, 30-s repetitive FLIM imaging time) (Luyben et al., 2024). Fluorescence lifetime data were analyzed by SymPhoTime software (PicoQuant, Berlin, Germany) (Hille et al., 2009). For average time-domain fluorescence lifetime measurements, the lifetime decay curves were fit by a double exponential decay model using the tail-fitting analysis. To detect BlgC photoactivation in living neurons, organotypic hippocampal slice cultures were prepared (Okamoto et al., 2004) and biolistically transfected cGMP FLIM probe cGiYR with or without BlgC (or cAMP FLIM probe REY with BlgC as a control). The expressed BlgC was photoactivated by single photon light (414 ± 21 nm, mercury arc lamp, 60 s) in the neurons. The lifetime change of the FLIM probes were measured by two-photon FLIM microscopy as described above for the in vitro experiments. The fluorescence lifetime pseudo-color image of neurons was constructed using a pixel by pixel fitting function.

Next, 14- to 16-week-old C57BL/6J male mice were anesthetized with isoflurane (2% vol/vol) and then transferred to a stereotaxic apparatus. Mice were secured to the apparatus by fixing ear bars to the head and inserting the incisor adaptor. A nose cone was used to administer isoflurane during surgery to maintain anesthesia as confirmed by mild pinching of the tail. Mice were injected with 0.1 mL of meloxicam (0.5 mg/mL) and 0.1 mL of saline subcutaneously prior to surgery. Hair atop the scalp was removed with electric clippers, and antiseptic solution was applied topically to clean the skin. In addition, Tear-Gel (Alcon, TX, United States) was applied to lubricate the eyes and reapplied when necessary. A scalpel was used to make a midline incision and expose the bregma and lambda landmarks of the skull. After leveling the head, a drill was used to make two small holes 2 mm posterior from bregma and 1.3- or 1.6-mm lateral to the midline. Injection volumes were 2 μL per hemisphere and delivered at a rate of 0.1–0.2 μL per min using a gastight syringe with a 26-gauge needle (Hamilton, Reno, NV, United States). Following injection, the syringe was kept in place for an additional 5 min to allow diffusion of the virus particles. The syringe was then slowly withdrawn from the head, and the injection site was wiped clean. The skin was sutured, and antibiotic ointment was applied over the wound. The mouse was removed from the stereotaxic apparatus and placed in a recovery cage on a heat pad. Meloxicam was administered subcutaneously every 24 h for 2 to 3 days following surgery. Mice were given at least 3 weeks of postoperative recovery. For validation of the expression, mouse brain coronal sections were prepared and imaged with DAPI staining as previously described (Yokose et al., 2017).

Acute hippocampal slices were prepared as previously described (Luyben et al., 2020) from adult BlgC-expressing mice and their wild-type littermates. After 1.5- to 2-h recovery, slices were transferred to a chamber perfused with an ACSF solution containing 124 mM NaCl, 3 mM KCl, 2.5 mM CaCl₂, 1.3 mM MgSO₄, 1.25 mM NaH₂PO₄, 26 mM NaHCO₃, and 10 mM glucose (pH 7.4, 30°C, 1.5 mL/min) equilibrated with 5% CO₂/95% O₂. Recordings of field excitatory postsynaptic potentials (fEPSPs) were recorded as previously described (Luyben et al., 2020) in the DG while blocking inhibitory synaptic function using 10 μM bicuculline (Saab et al., 2009). The stimulation electrode was positioned in the dorsal blade of the dentate molecular layer for MPP stimulation. Paired field responses were evoked by stimulating with an intensity (0.05 ms pulses, 40 ms apart) that yielded fEPSPs that were 40% of the maximum spike-free fEPSP size. Responses were evoked and acquired every 20 s throughout the experiment using an Axopatch 1D amplifier (Axon Instruments) digitized at 20 kHz and measured by slope (10–50% of fEPSP rising phase). The expressed BlgC in DG granule cells was photoactivated using a blue LED light: 1.5 mW, 4 min for baseline synaptic responses, 5 min for LTP (4 min before tetanus and 1 min during tetanus) under an objective lens (4X, NA0.1). Tetanus was induced with a bipolar tetanic stimulation (100 Hz, 0.15 ms pulses delivered in 4 trains of 0.5 s duration, 20 s apart). In the time course experiments, field responses were plotted by normalizing to the baseline fEPSP slope (average of the 10-min period prior to tetanic stimulation).

The hippocampal slices from BlgC and WT littermate mice were homogenized in buffer (40 mM HEPES/Na, pH 8.0, 0.1 mM EGTA, 5 mM magnesium acetate, 1 mM DTT, and 0.01% Tween-20) by sonication and centrifuged at 16,000 g for 15 min to clear large tissue debris as previously described (Luyben et al., 2020). The expressed hippocampal lysates were photoactivated for 5 min with a 455 nm LED (4.5 mW/mm²; ThorLabs, NJ, United States) on a plastic paraffin film (Parafilm M^®, Bemis, United States) covered glass slide at room temperature in the dark with 200 μM GTP. After photoactivation, the reactions were stopped by application of 0.1 M HCl following the ELISA kit protocol. The synthesized cGMP concentrations were measured by a cGMP ELISA kit (Enzo Life Sciences, NY, United States).

For behavioral testing, C57BL/6J male mice with BlgC AAV injection were surgically equipped with a bilateral LED (or dummy) implant targeting the same site where virus was injected. A bilateral 470 nm wavelength LED device (0.75 mm diameter, 20 mW, Eicom USA, CA, United States) was inserted slowly to minimize damage to brain tissue. The LED device was fixed to the skull using two kinds of dental cement (Bistite II DC, Tokuyama Dental America Inc., Japan, and UNIFAST Trad Liquid, GC America Inc., United States) and cured. The mouse was then removed from the stereotaxic device and placed in a recovery cage on a heat pad (the implant work was done in Japan). Analgesia (METACAM) was continued by subcutaneous injection every 24 h for 3 days post-surgery. At least 14 days were interspaced between surgery and behavioral testing.

Mice were surgically equipped with either a bilateral fiber optic LED system (Eicom USA, CA, United States) or, as a control, a “dummy” optic fiber, which produces no light but exhibits the same physical properties. The mice were housed in a home cage monitoring system with automated video tracking and habituated to wearing a detachable battery (necessary to power the LED). The battery (or dummy battery) was attached and turned on to produce cGMP during each test (light-activation details available in each figure). All behavioral tests were performed as previously described (Ageta-Ishihara et al., 2013; Shoji et al., 2016). The control and BlgC mouse cohorts were counterbalanced such that each test was performed on equal number of control and experimental mice simultaneously (in the case of multiple chambers). In single chamber tests, a control mouse test was followed by BlgC mouse test until all mice were evaluated. After the behavior testing was complete, the mice were sacrificed and checked for the expression of BlgC.

Statistical methods are indicated in the figure legends. All data are presented as mean ± SEM. ***p < 0.001; **, p < 0.01; *, p < 0.05; NS, p > 0.05.

The study was designed to determine the role of cGMP in synaptic plasticity and cognitive functions, with a specific focus on MPP fibers synapsing onto dentate gyrus (DG) granule neurons (MPP-DG synapses). Bilateral DG injections of virus expressing a green fluorescence protein (GFP)-tagged photoactivatable BlgC (Ryu et al., 2010) under the CaMKIIα promoter were carried out, and the expression was detected in granule cells including their mossy fiber projections to CA3 (Figures 1A,B). We know that the BlgC actuator system generates cGMP since we tracked its dynamics within neurons from organotypic hippocampal slices in cultures. Specifically, we introduced a cGMP FLIM probe, cGiYR (cGMP indicators with YFP/RFP cGMP FLIM) and, upon blue light stimulation (414 ± 21 nm), determined that the fluorescence lifetime of cGiYR increased and was maintained for at least 10 min, specifically in the neurons biolistically co-transfected both BlgC and cGiYR probe, but not cGiYR probe alone or REY (cAMP probe) + BlgC (Supplementary Figure S1). We then returned to the mouse brain to evaluate whether the BlgC-expressing DG neurons can produce a cGMP response. Hippocampal slice lysate was photostimulated (455 nm LED, 4.5 mW/mm², for 5 min), and subsequently, cGMP levels were determined by ELISA. Compared with the no-light control, the photoactivated group showed elevated cGMP levels, thus demonstrating enzymatic functionality and effective application of the BlgC actuator system within the DG (Figure 1C).

To test whether the expression of BlgC (without photoactivation) has any impact on the basal synaptic transmission at intact MPP-DG synapses, we measured the field excitatory postsynaptic potentials (fEPSPs) in acutely prepared hippocampal slices from control mice and those expressing BlgC in the DG (Figure 2A). The input/output (I/O) curves and paired-pulse ratio (PPR) at MPP-DG synapses showed no significant difference between BlgC and control groups (Figures 2B,C), indicating that expression of BlgC does not affect basal synaptic properties. In addition, synaptic responses remained unchanged after 4 min of LED illumination (480 ± 15 nm, 1.5 mW) (Figure 2D), indicating that there is no immediate effect of cGMP on basal synaptic transmission.

We next examined whether photoactivation of BlgC has any effect on LTP at MPP-DG synapses. To relieve the strong endogenous inhibition at dentate granule dendrites and maximize synaptic plasticity (Wigstrom and Gustafsson, 1983; Nguyen and Kandel, 1996), we applied strong tetanic stimulation (tetanus: 4 × 100 Hz) in the presence of 10 μM bicuculline to block GABA_A receptor-mediated inhibition (Saab et al., 2009). Immediately prior to induction, control or BlgC-expressing hippocampal slices received either 5 min of photostimulation (+Light) or no light (−Light) (Figures 2E,F). In the control mice, the light did not affect the extent of LTP (Figure 2E right). However, upon photostimulation of BlgC in combination with tetanic stimulation of the MPP, there was enhanced LTP that persisted for the additional 90-min recording period (Figure 2F). An analysis of the paired-pulse ratio (PPR), a measure of the probability of presynaptic transmitter release, was not significantly altered following LTP induction either in the absence or presence of BlgC and regardless of photostimulation (Figure 2G). The data show that tetanic stimulation in the presence of elevated postsynaptic cGMP significantly enhances LTP at MPP-DG synapses.

To evaluate the effect of cGMP on a battery of mouse behavioral tests, we surgically implanted BlgC-expressing mice with either (i) a bilateral fiber optic-coupled and wireless LED system to deliver light to the DG that we refer to as the “cGMP” group or (ii) a control (Cntrl) group with “dummy” optic fiber which produces no light but exhibits the same physical properties (Figure 3A, upper). After a 2-week recovery following surgery, we carried out a battery of behavioral tests including general health and neurological screening, anxiety-like behavior, social interaction, and memory tests (Figure 3A, lower). We performed general health and neurological screening (GHNS), rotarod (RR), and hotplate (HP) test to determine the health condition and sensory/motor function of the animals (Supplementary Figure S2). The GHNS indicated no obvious abnormalities in general health, gross appearance, or nerve reflexes in the injected mice (Supplementary Figure S2A white bars). In the RR (Supplementary Figure S2B white bars) and HP tests (Supplementary Figure S2C white bars), the latency to fall and response to noxious temperature were indicative of good motor function and pain sensitivity in these mice. To examine the effect of cGMP on mouse condition and performance, we performed a 5-min BlgC photostimulation prior to the screen (Supplementary Figures S2A–C black bars, cGMP). The photostimulation of BlgC in hippocampal DG neurons did not show any effect on the basal performance in the mice, indicating that there is no significant effect of cGMP on the general health condition, motor function, and nociception of the mice.

We then examined the effect of optogenetic cGMP manipulation on anxiety-like behavior, social behavior, learning, and memory. The overall results of the study are shown in Figure 3B and the data for each significant test appear in Figures 4–6 and the others in Supplementary Figures S2–S7. Among the test battery performed in this study, the tests for anxiety-like behavior which last a total of 10 min, such as the light-dark (LD) (Supplementary Figure S3) test and the elevated plus maze (EP) test (Supplementary Figure S4), did not reveal any effect of BlgC photoactivation (cGMP generation in DG neurons). However, the open field (OF) test (Figure 4) showed that the cGMP group spent an increasingly reduced time in the center of the arena with stay time (Figure 4E), such that a statistically significant increase in anxiety-like behavior persisted at the end of the 120 min of observation. The single chamber social interaction (SI) test also showed an effect in the group receiving optogenetic production of cGMP in DG neurons (Figure 5), consisting of an increase in the number of contacts (Figure 5D) that signifies increased sociability. However, using Crawley’s social interaction (CSI) test, the DG cGMP group had no significant difference in the time spent interacting with a caged stranger over an empty cage (Supplementary Figure S5), suggesting social novelty was not affected, and therefore, cGMP may be specifically involved in sociability.

In the object location test (OLT) (Supplementary Figure S6) and T-maze (TM) test (Supplementary Figure S7), there were no significant differences in performance between controls (Cntrl) and photoactivated (cGMP) mice. In addition, the Barnes maze (BM) test (Figure 6) was used to examine training-dependent reference memory, where we photostimulated BlgC-expressing mice for 5 min in each daily training (Figure 6A). During the training trials, the photostimulated BlgC mice (cGMP) did not show significant differences in the latency, error, and distance to find the escape hole compared to control mice (Cntrl) (Figure 6B). In probe test 1 (PT1), which was conducted 24 h after the last training, there were fewer errors in the cGMP mice (Figure 6C), suggesting the improvement of spatial reference memory retrieval. However, there were no significant differences in the time spent around the target escape hole between groups (Figure 6D). Both of the groups maintained a strong preference to the target hole compared to the adjunct holes (Figure 6E) in PT1. In contrast, during probe test 2 (PT2) which was performed 2 weeks after the last training session (Figures 6C,D,F), photostimulated cGMP mice did not show any differences in the latency, error, or distance to find the escape hole compared to control mice (Figure 6C). In addition, there was no differences in the time spent around the target escape hole between groups (Figure 6D). While the control mice lost the preference to the target hole, cGMP mice maintained the preference after 2 weeks (Figure 6F), suggesting that there is an effect of cGMP on the enduring retention of reference memory. Altogether, cGMP signaling was associated with increased sociability, improved long-term reference memory, possibly at the expense of increased anxiety-like behavior over the longer term.

To manipulate cGMP in neurons, we virally expressed photoactivatable guanylyl cyclase (BlgC) in the granule neurons of the hippocampal DG region. We prepared the brain slices and imaged the CaMKIIα-driven expression of GFP-tagged BlgC. Confocal imaging revealed that the mice expressed abundant levels of GFP within the dentate granule neurons including the mossy fiber axons, indicating the localized expression of BlgC within the DG granule neurons, but not the presynaptic perforant path fibers. We next validated photoactivation of the expressed BlgC. The lysates from the acute slices showed cGMP production upon delivery of blue light. To directly observe the synthesized cGMP by photoactivation of BlgC in living neurons, we coexpressed a cGMP FLIM probe (cGiYR) with the BlgC in living hippocampal organotypic cultured neurons and detected the FLIM change immediately after blue light stimulation. Expression of cAMP FLIM probe (REY) with BlgC did not show the FLIM change upon the light stimulation, indicating the increase of cGMP, but no detectable change of cAMP levels in the neurons.

The expression of BlgC in the granule neurons did not affect the baseline fEPSPs at MPP-DG synapses in the hippocampal slices, nor the paired-pulse ratio suggesting no changes in the probability of release. Photostimulation of BlgC also did not change the fEPSPs during recording of the basal synaptic responses. We next evaluated DG LTP induced by tetanic activation of MPP fibers in the presence of the GABA_A-receptor blocker (10 μM bicuculline). Using this system, we demonstrated that the short blue light stimulation of photoactivatable BlgC coincident with the tetanic stimulation enhanced the level of evoked LTP. However, the presynaptic function was unaffected, indicating that when postsynaptic cGMP is elevated during high-frequency synaptic activity (tetanic stimulation), there is an enhanced postsynaptic potentiation of synaptic strength that persists for at least 90 min. This interesting feature of cGMP-dependent enhancement of LTP at MPP-DG indicates that cGMP has a selective impact during conditions that enable synaptic plasticity. Indeed, it has been previously shown that cGMP signaling pathways are involved in LTP in hippocampus CA1 neurons, where NO-GC-PKG signaling has been reported as a retrograde messenger system from post- to presynaptic terminal for the induction of LTP, suggesting a presynaptic cGMP function (O’Dell et al., 1991; Zhuo et al., 1994; Arancio et al., 1996). NO has been shown to be involved in LTP induction at the MPP-DG synapse but the signaling mechanism may not involve PKG, as determined by using pharmacological inhibition assays (Wu et al., 1997, 1998). Therefore, the LTP-enhancing pathways downstream of cGMP remain to be determined.

We employed a wireless LED with fiber optics implant system for photoactivation of BlgC in the DG neurons of freely behaving mice and simultaneously recorded the behavioral changes. BlgC photostimulation did not affect the general health condition or motor function of the animals.

In the behavioral test battery, the shorter behavioral tests of anxiety (10 min tests) such as LD and EP did not reveal any significant changes following photostimulation of BlgC. In the OF test, we photostimulated BlgC two times longer (10 min) than in other behavioral tests (LD and EP), but the mice did not show any increase in anxiety-like behavior during the initial 90 min of testing.

Upon exposure to the open field, compared to the BlgC group, the control group began to explore the center of the arena with time. The simplest interpretation is that control mice gradually overcome their fear or anxiety and begin to explore the most exposed part of the arena.

Interestingly, in the latter portion of the OF testing (90–120 min), BlgC photostimulated mice continued to avoid the center of the arena, which may reflect increased anxiety-like behavior. Both the NO/cGMP pathway and the hippocampus are implicated in stress responses and anxiety (Spolidorio et al., 2007; Calixto et al., 2010; Ghasemi et al., 2022; Shi et al., 2023). However, it must be noted that in contrast to the OF test, there were no differences in the time spent in the bright section of the light-dark arena, possibly because this latter test is much shorter in duration (120 vs. 15 min, respectively). Similarly, the 10-min EP test did not reveal any differences in time spent in the open arms. Additional longer testing may be able to clarify whether stress, anxiety, or possibly improved spatial learning, which is the reason underlying the open field results.

Notably, cGMP is known to decrease with aging (Domek-Lopacinska and Strosznajder, 2010), and aging impairs cognitive function as measured in the Barnes maze for mice (Shoji et al., 2016; Shoji and Miyakawa, 2019). At the same time, aging increases center time stay in the OF test, especially in the latter half of the period (Shoji et al., 2016; Shoji and Miyakawa, 2019). In the present study, cGMP was increased locally in the dentate gyrus, and the phenomena observed in both the BM and OF test were opposite to the phenotype of the aged mice, in which cGMP is expected to be decreased. Whether the anxiety-like behavior observed in OF is associated with spatial memory is not clear; however, our results may suggest that the dentate gyrus and cGMP are involved in both anxiety-like behavior and spatial memory.

The excitatory entorhinal cortical perforant projections to the dentate gyrus (EC-DG) neurons are tightly involved in social memory (Leung et al., 2018). In the SI test (10-min duration) following photostimulation of BlgC in DG neurons, there was a rapid effect for cGMP in enhancement of social behavior. It is not clear why cGMP and control groups performed similarly in the CSI test that compares interactions with either a caged stranger mouse or familiar mouse, or empty cage. The role of cGMP could be specific to direct social interaction between mice.

As postsynaptic cGMP elevation enhanced LTP at the MPP-DG synapses, we examined performance in several learning and memory tasks during photoactivation of BlgC in DG neurons. In the memory test battery, the BlgC group receiving photostimulation did not show any significant differences on tasks including the OLT, training period of the BM, and the TM test, suggesting no immediate effect of cGMP in the learning and memory. However, in the OLT, even the control mice showed a strong memory of the object novel location, which could have made it difficult to detect any further improvements in the cGMP group. Using an OLT training protocol that generates weak memory in control mice may help to further evaluate the more immediate function of cGMP.

Interestingly, 24 h after the training in the BM test, the cGMP mice showed significantly improved performance result compared to controls. This learning and memory improvement after training with photostimulation of BlgC could be due to cumulative cGMP-dependent enhancement of synaptic potentiation during the trials. When we assessed mice 2 weeks after the training trial, the controls already lost the reference memory, but the cGMP group showed a maintenance of the preference memory. Thus, cGMP in the DG neurons could serve to enhance the retrieval and long-term retention or reference memory.

The observation that cGMP elevation only affected a few of the behavioral tasks may relate to the relative cognitive role of the dentate in each case. Some specific roles of the dentate include pattern separation (novelty detection) and episodic memory encoding, though there is controversy on its role in memory retrieval or recall (Hainmueller and Bartos, 2020; Borzello et al., 2023). In terms of elevated cGMP, a PDE5 inhibitor has been shown to facilitate long-term retention of inhibitory avoidance memory (Baratti and Boccia, 1999). In addition, PDE2 inhibition has also been shown to improve memory in a cGMP-dependent manner (Lueptow et al., 2016). Interestingly, cGMP/PKG was found to mediate early memory consolidation and early-phase LTP, whereas cAMP/PKA mediated the late consolidation of LTP (Bollen et al., 2014); there is also evidence suggesting distinct changes involving increases in surface GluA1- and/or GluA2-containing AMPARs in the two signaling pathways with respect to acquisition versus consolidation phases of memory (Argyrousi et al., 2020). Any specific cGMP-mediated increase in GluA1 could potentially relate to the incorporation of calcium-permeable AMPARs that have been shown to be essential in cAMP/PKA-dependent forms of long-lasting LTP (Park et al., 2018; Park et al., 2021).

Presynaptic effects of elevating cGMP have also been described (Boulton et al., 1994; Fieblinger et al., 2022). We did not find any evidence for presynaptic actions, which may be because cGMP was not elevated in the perforant path inputs. In which case, additional roles of presynaptic alterations in cGMP signaling on dentate-dependent behaviors cannot be excluded.

In summary, our optogenetic approach using a light-sensitive BlgC actuator allowed us to directly assess the role of DG cGMP elevations on synaptic function by electrophysiology and behavioral functions that included cognition, sociability, and anxiety. The results demonstrate that postsynaptic cGMP elevation can facilitate LTP at MPP synapses onto DG granule cells. In behaving mice, DG cGMP increased sociability, had a long-term anxiogenic action, and improved reference memory. In future work, the optogenetic BlgC actuator could, in principle, be used to study spatiotemporal aspects of cGMP functions in synaptic plasticity and learning and memory. By extension, the results suggest that methods to elevate cGMP specifically within dentate granule neurons may provide a means to modify brain function in various brain disorders.