Creating vectors for expression of the thermosensitive human TRPV1 channel and the genetically encoded calcium sensor GCaMP6s

The plasmid pCAGGS-hTRPV1-p2A-tdTomato, carrying an open reading frame for expression of hTRPV1 under control of the cytomegalovirus enhancer fused to the chicken beta-actin promoter (CAG), was generated in two steps. First, p2A-tdTomato fusion from eolTRPA1-p2A-tdTomato construct [15] was cloned into the pCAG_PSD95.FingR-eGFP-CCR5TC vector (a gift from Don Arnold; Addgene plasmid #46295; http://n2t.net/addgene:46295; RRID: Addgene_46295) with SmaI and Bsp1407I restriction sites. The open reading frame of the human thermosensitive TRPV1 channel (Horizon Dicovery, # OHS5894-202501188) was used as a template in a polymerase chain reaction (PCR) with following primers: F1 5′- gaattcggtaccgcgggcccgccaccatgaagaaatggagcag-3′ and R1 5′- agttagtagctccgcttcccgccttctccccggaagcggc-3′. The PCR product was fused upstream the p2A-tdTomato using AQUA cloning technology according to conventional protocol [28].

The plasmid pAAV2-hSyn1-GCamP6s-p2A-tdTomato (hSyn1 = human Synapsin 1 promoter, p2A = self-cleaving peptide, and tdTomato = reporter red fluorescent protein tdTomato) was assembled with NEBuilder HiFi DNA Assembly Cloning Kit (NEB, #E5520S) using 25 bp overlapping primers for 4 PCR fragments obtained with Q5 High-Fidelity 2X Master Mix (NEB, #M0492L) from corresponding sources. The hSyn1 505 bp promoter (human synapsin I promoter; confers neuron-specific expression [29] was amplified from hSyn1-tdTomato plasmid (a gift from Viviana Gradinaru, Addgene plasmid #104060; http://n2t.net/addgene:104060; RRID:Addgene_104060) with hSyn1-rev 5′-gatgatgagaacccatggtggcgcctgcgctctcaggcacgac-3′ and hSyn1-frw 5′-ctattaccatggctcgagagtgcaagtgggttttaggaccag-3′ primers. An adeno-associated virus serotype 2 Amp resistant backbone of 4616 bp was amplified from plasmid pAAV2-HyPer2-DAO-NES [30] with AAV2-rev 5′- cttgcactctcgagccatggtaatagcgatgactaatacg -3′ and AAV2-frw 5′- acaagtaatgcctcgagcagcgctgctcgagagatctacgg-3′ primers. A GCaMP6s 1350 bp fragment was amplified from plasmid Tol2-elavl3-GCaMP6 (a gift from Misha Ahrens, Addgene plasmid #59531; http://n2t.net/addgene:59531; RRID: Addgene_59531) with GCaMP6s-frw 5′- cgccaccatgggttctcatcatcatcatcatcatggtatggctagc-3′ and GCaMP6s-rev 5′ gaagttagtagccttcgctgtcatcatttgtacaaact-3′ primers. A p2A-TdTomato 1488 bp insert was amplified from previously published caTRPA1-p2A-TdTomato construct [15] with p2A-frw59 5’-gatgacagcgaaggctactaacttcagcctgctgaagc -3′ and TdTomato-rev60 5′- ccgtagatctctcgagcagcgctgctcgaggcattacttgtacagctcgtccatgccg-3′ primers. Positive colonies were selected with colony PCR of resuspended bacterial clones with hSyn1-cl-frw 5′- tcagcactgaaggcgcgctg-3′ and pAAV2-sc-rev 5′- gggccaggagaggcact-3′, expanded, and used for maxiprep plasmid preparation with Plasmid Maxi Kit (25) (QIAGEN, #12163).

The plasmid pAAV2-CaMKIIα-hTRPV1-flag (CaMKIIα = Ca2 + /calmodulin-dependent protein kinase II alpha promoter; flag = FLAG epitope) was created using the pAAV2-hSyn1-GCamP6s-P2A-tdTomato plasmid as a backbone sequence source. The CaMKIIα promoter was amplified from pAAV-CaMKIIa-mCherry plasmid (a gift from Karl Deisseroth; Addgene plasmid #114469; http://n2t.net/addgene:114469; RRID: Addgene_114469) with F_CamkII_Mlu 5′-tttacgcgtttaacattatggccttaggtcacttc-3′ and R_CamkII_Nhe 5′-tttgctagcgctgcccccagaactagg-3′ primers. The open reading frame of the human thermosensitive TRPV1 channel (Horizon Discovery, # OHS5894-202501188) was used as a template in a PCR with following primers: hTRPV1_Fw 5′-tatatgctagcgccaccatgaagaaatggagcagcacagac-3′, which adds Kozak (protein translation initiation site) and an NheI restriction site, and hTRPV1_Rv 5′-atatagctcgaggaattcttacttatcgtcgtcgtccttgtaatccttctccccggaagcgg-3′, which adds FLAG-peptide for subsequent immunohistochemical visualization of the expressed channel, and EcoRI + PspIX restriction sites. Escherichia coli XL1-Blue strain (Evrogen, #CC001) was used for the cloning, maintenance, and propagation of plasmids.

DNA fragments extraction from agarose gel was carried out using a Cleanup Standard kit (Evrogen, #BC022S). Plasmid Midiprep 2.0 kit (Evrogen, #BC124) was used for plasmid DNA extraction. Ligation was conducted using T4-DNA ligase (Evrogen, #LK001).

The created constructs were verified by restriction analysis with appropriate restriction enzyme sets and sequencing. Midipreps were prepared using the Plasmid Midiprep 2.0 kit (Evrogen, #BC124) according to the manufacturer's instructions. Then, the constructs pAAV2-CaMKIIa-hTRPV1-flag and pAAV2-hSyn1-GCamP6s-P2A-tdTomato were packaged into adeno-associated viral (AAV) particles. The constructs were packaged in AAV serotype 9 viral particles. The virus titers were 4.4E + 12 viral genomes (VG)/mL for AAV2/9-CaMKIIa-hTRPV1-flag and 4.07E + 12 VG/mL for AAV2/9-hSyn1-GCaMP6s-p2A-tdTomato.

Culture and transfection of HEK293TN cell line

HEK293TN cells (gender: F, RRID: CVCL_UL49) were cultured in standard 25 cm2 flasks (SPL Life Sciences, #70025) in Dulbecco’s Modified Eagle Medium (DMEM) (Paneco, #C420E) supplemented with 10% fetal bovine serum (FBS) (Biosera, #FB-1001), GlutaMAX (1:100) (Gibco, #2009511) and a penicillin–streptomycin mixture (1:100) (Gibco, #15140-122) at 37 °C in the ambient atmosphere with 5% CO2. The HEK293TN cell culture was tested for mycoplasma contamination using a MycoReport kit (Evrogen, #MR001) according to manufacturer’s guidelines. For thermal stimulation and real-time calcium imaging, HEK293TN cells (60,000 cells/mL) were seeded into 35-mm plastic dishes (SPL Life Sciences, #20035) and cultured in 2 mL of DMEM supplemented with 10% FBS, GlutaMAX (1:100) and a penicillin–streptomycin mixture (1:100) at 37 °C in the ambient atmosphere with 5% CO2. Plastic dishes were preliminarily coated with polyethylenimine (1 mg/mL) (Sigma-Aldrich, #BCCC2235) for 30 min at room temperature, and then washed with distillated water three times. Cells were transfected with a mixture of 1.25 µg DNA and 2.5 µL FuGene HD reagent (Promega, #E2311) pre-dissolved in 150 µL Opti-MEM (Gibco, #15140-122) per dish. Experimental cells were co-transfected with pCAGGS-hTRPV1-p2A-tdTomato (750 ng per dish) and pCS2 + CMV-GCaMP6s [15] (500 ng per dish). Plasmid pCS2 + CMV-GCaMP6s carried an open reading frame for expression of genetically encoded calcium indicator GCaMP6s [31] under control of human cytomegalovirus immediate early enhancer and promoter (CMV). Control cells were transfected with pCS2 + CMV-GCaMP6s (1.25 µg per dish) only. Cells underwent simultaneous thermal stimulation and real-time calcium imaging 18–20 h after transfection.

Controlled heating and real-time calcium imaging of transfected HEK293TN cells

Prior to thermal stimulation and real-time calcium imaging, culture medium was replaced with 3 mL of the pre-heated Hanks’ balanced salt solution (HBSS) (PanEco, #P020) supplemented with 10 mM HEPES-Na (PanEco, #F134). Then, a 35 mm dish with experimental or control cells was placed onto a stage of a Nikon ECLIPSE Ti2-E epi-fluorescence microscope equipped with a SPECTRA X light engine and a Photometrics BSI camera. The fluorescent signals of GCamp6s and tdTomato (control cells expressed GCaMP6s only) were acquired using a Nikon CFI S Plan Fluor LWD 20X (NA 0.7) objective and Nikon NIS-Elements software. GCaMP6s was excited at 470 nm, and its fluorescence was acquired within the spectral range 500–530 nm. tdTomato was excited at 555 nm, and its fluorescence was acquired within the spectral range 580–610 nm. We performed time-lapse imaging with 1 s interframe intervals for the subsequent measurement of changes in fluorescence of GCaMP6s in experimental (tdTomato positive) and control cells.

Controlled heating of cells during real-time calcium imaging was carried out by a custom heater/temperature control setup connected to a computer (Suppl. Fig. S1A, B). A nichrome wire and a temperature sensor were placed into a plastic dish with experimental or control cells at the sites adjacent to the imaged field of view. This ensured that imaged cells are appropriately heated and temperature at the site of real-time calcium imaging is properly measured. The heater/temperature control setup was operated by a custom software that uses a proportional–integral–derivative (PID) controller algorithm (Suppl. Fig. S1C). The coefficients that determine how much each component of the PID controller contributes to the output heating current were adjusted empirically in preliminary experiments emulating conditions in a 35 mm dish such as liquid volume and positioning of a nichrome wire and a temperature sensor. In our experiments, optimal PID coefficients should provide fast heating of cells from the baseline temperature of 37 °C up to desired temperatures of 39, 41, and 43 °C with error not exceeding 0.5 °C.

Using the software for the heater/temperature control setup, cells in the dish were maintained at the temperature of 37 °C during 10 min prior to beginning of time-lapse imaging. Then, time-lapse imaging and controlled heating of cells got started simultaneously. Cells were heated according to the following scheme: 37 °C for 60 s, 39 °C for 30 s, 37 °C for 90 s, 41 °C for 30 s, 37 °C for 90 s, 43 °C for 30 s, and 37 °C for 90 s. Desired increases of temperature were achieved within several seconds. Cooling of cells to the basal temperature of 37 °C occurred passively. Temperature was measured with the frequency of 10 Hz and recorded in a Microsoft Excel file format (.csv).

Mice

Experiments were carried out using C57Bl/6 J mice (The Jackson Laboratory, #000664, RRID: IMSR_JAX:000664). Mice were maintained in groups (3–4 mice per cage) in the animal facility of Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry. Mice were housed at an ambient temperature 22–24 °C with a 12-h light/dark cycle and had ad libitum access to food and water. All manipulations with animals were conducted according to the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes (1986, ETS 123) and approved by the Institutional Animal Care and Use Committee (approvals no. 182 and 364).

Stereotaxic viral vector delivery and implantation of a fiber-optic interface

AAV vectors were delivered to the CA1 area of the hippocampus (4 to 6-week-old male mice) or CnF (8-week-old male mice) by standard stereotaxic surgery. Briefly, mice were deeply anesthetized with 3% isoflurane (Baxter, #5AGG9621) in an induction chamber using a low-flow anesthesia system (SomnoSuite, Kent Scientific). Then, mice were placed onto the heating pad of the stereotaxic apparatus (digital stereotaxic instrument, Stoelting Co). Anesthesia was maintained by delivery of 1.5% isoflurane via inhalation mask. Body temperature was controlled by a rectal thermometer. Prior to incision, 0.25% bupivacaine (Ozon, #LP-003590) was injected subcutaneously (the total dose was less than 8 mg/kg) at the site of incision for local analgesia. After the skin on the mouse head was incised, connective tissue was removed using sterile cotton buds, and the surface of the skull was dried to facilitate identification of the bregma. The skull surface was aligned in anteroposterior and mediolateral directions. AAV vectors were injected into the mouse brain using a Hamilton microsyringe 75RN 5 µL (Hamilton, #7634‑01) equipped with a 33-gauge needle via holes drilled in the skull at the following stereotaxic coordinates: − 3.3 mm anteroposterior, ± 3.4 mm mediolateral (both hemispheres), and − 3.9/− 3.6/− 3.3 mm dorsoventral relative to the bregma for the CA1 area of the hippocampus [32, 33] (Fig. 2A); and − 4.8 mm anteroposterior, 1.1 mm mediolateral (half of mice received an AAV injection into the left hemisphere, and the rest of mice received an AAV injection into the right hemisphere), and − 2.75 mm dorsoventral relative to the bregma [33, 34] for the CnF (Fig. 3B). The AAV vector solution was delivered at 200 nL/min. Experimental mice (hTPRV1+ mice) were injected with an AAV mixture (1:1) containing AAV2/9-CaMKIIa-hTRPV1-flag and AAV2/9-hSyn1-GCaMP6s-p2A-tdTomato. Control mice were injected with AAV2/9-hSyn1-GCaMP6s-p2A-tdTomato dissolved in sterile phosphate-buffered saline (1:1). In the hippocampus, 400 nL of an AAV vector solution was injected at each point in the dorsoventral axis. In CnF, 1000 nL of an AAV vector solution was injected at a single point in the dorsoventral axis. After the intrahippocampal injection, the holes in the skull were sealed by the dental cement, the wound was closed, and mice were placed into individual cages and allowed for recovering for at least two weeks prior to electrophysiological recordings. Immediately after an AAV vector solution injection into CnF, a fiber-optic interface with the diameter of 220 µm [35, 36] was implanted slightly above the site of the AAV delivery by slow deepening to the depth of 2.65 mm relatively to bregma. The ceramic pert of fiber-optic interface was fixed at the surface of the skull by the dental cement. After surgery, animals were placed into individual cages and allowed for recovering for at least two weeks prior thermogenetic stimulation and behavior recording.

For analgesia, mice were intraperitoneally injected with ketoprofen (5 mg/kg) (Sandoz, #N013942/01) prior to surgery, and again 24 and 48 h after surgery. The recovery period also enabled sufficient expression of hTRPV1 and GCaMP6s to be reached.

Electrophysiology and intracellular calcium imaging in acute brain slices

Mice that received AAV vector intrahippocampal injections were deeply anesthetized with 3% isoflurane and decapitated, and the brains were isolated from skulls. Transverse hippocampal 300 μm slices were prepared from the isolated brains using a Leica VT1200S vibrating blade microtome (Leica Microsystems). The slicing chamber contained an oxygenated ice-cold solution composed of (in mM): K-Gluconate, 140; N-(2-hydroxyethyl) piperazine-N′ -ethanesulfonic acid (HEPES), 10; Na-Gluconate, 15; ethylene glycol-bis (2-aminoethyl)-N, N, N′, N′-tetra-acetic acid, 0.2; and NaCl, 4; pH 7.2. Slices were incubated for 30 min at 35 °C before being stored at room temperature in ACSF containing (in mM): NaCl, 125; NaHCO3, 25; KCl, 2.5; NaH2PO4, 1.25; MgCl2, 1; CaCl2, 2; and D-glucose, 25. The ACSF was continuously bubbled with 95% O2 and 5% CO2.

For experiments, brain slices were placed in a perfusion chamber of a Scientifica SliceScopePro 2000 microscope (RRID:SCR_018405; Scientifica, UK) and were continuously perfused with the oxygenated ACSF. Patch electrodes were pulled from hard borosilicate capillary glass (Sutter P-97/PC Pipette Puller (RRID:SCR_018636)). Electrodes were filled with a solution consisting of (in mM): K-gluconate, 140; HEPES, 10; NaCl, 8; MgATP, 4; MgGTP, 0.3; and phosphocreatine, 10 (pH 7.3 with KOH). CA1 pyramidal cells were visually identified using the IR-video microscope Scientifica SliceScopePro 2000 in differential interference contrast mode.

Whole-cell recordings were conducted in ACSF with a constant temperature of 35 °C in voltage-clamp mode using a Heka Elektronik EPC 10 USB Patch Clamp Amplifier (RRID:SCR_018399) and the Patchmaster software (RRID:SCR_000034) with a sampling rate of 100 μs and filtered at 3 kHz. Patched neurons were held at − 70 mV. GCaMP6s fluorescence was acquired using a water immersion Olympus LUMPLFLN40 × W objective with numerical aperture (NA) of 0.8. Image acquisition was carried out using the Scientifica SliceScopePro 2000 microscope equipped with a CoolLED pE-300ultra (RRID:SCR_021972) light source and a Hamamatsu Orca Flash 4.0 CMOS monochrome digital camera (Hamamatsu Photonics). Fluorescence signals were acquired and analyzed with the uManager free software (RRID:SCR_000415) at 20 fps. The CoolLED pE-300ultra (RRID:SCR_021972) light source was synchronized with Heka Elektronik EPC 10 USB Patch Clamp Amplifier via BNC-TTL output. The GCaMP6s fluorescence was excited and detected via a 49054 CoolLED GFP filtercube.

To assess average elevations of membrane potential (∆Vm) during heating periods, patch-clamp traces were filtered using a low-pass cutoff frequency 1 Hz Butterworth filter in the pClamp software (RRID:SCR_011323). For counting numbers of action potentials generated during heating periods, we used Event detection function in the pClamp software.

The optical setup for thermal stimulation of principal hippocampal neurons in acute brain slices

A fiber coupled laser diode 4PN-117 (SemiNex) with an emission wavelength of 1375 nm and power up to 4.3 W was used as a source of the IR irradiation for heating brain slices (Fig. 2B). The laser diode was mounted onto a thermo-electrically controlled (TEC) plate “264 TEC HP LaserMount” (Arroyo Instruments). The TEC plate was operated by a TEC driver TECSource 5305 (A.I.). A laser diode driver LaserSource 4320 (A.I.) provided laser diode current stabilization and control. For targeting the IR laser to a desired area in an acute slice, we used a PL520 (Thorlabs) laser diode with an emission wavelength of 520 nm. The visible 520 nm laser was connected closely to the end of a 200-μm FG200UEA optical fiber with NA of 0.22 (Thorlabs) by optical glue. The IR and visible laser beams were combined into the same optical pathway using 90:10 splitters TM105R2F2B (Thorlabs) for fibers with a diameter of 105 µm, NA of 0.22, and reduced OH content. For efficient transmission of the IR laser irradiation, it was fed into the Signal channel with a 90% power transfer efficiency. The splitter’s auxiliary output was used to monitor IR laser power by a photodiode. The main channel was connected to a 2 m FG105LCA optical fiber (Thorlabs) with a diameter of 105 µm and NA of 0.22. An increased bandwidth of the fiber within the IR range provided a 100-fold reduction in energy losses. All optical elements of the fiber were securely coupled to each other through connectors. The overall IR power transfer efficiency in the optical setup was 65%. To synchronize IR laser treatments, electrophysiological recordings, and calcium imaging, the laser diode driver LaserSource 4320 used for the control of IR laser power and the CoolLED pE-300ultra light source used for excitation of the calcium biosensor GCaMP6s were connected to Heka Elektronik EPC 10 USB Patch Clamp Amplifier via a BNC analogue output and a BNC-TTL output, respectively.

The flat endface of the 2 m FG105LCA optical fiber was placed above the slice with some angle to the slice surface. The heating IR 1375 nm heating laser was targeted to a desired site in an acute brain slice using a spot of the visible 520 nm laser. The heating/cooling protocol included six sequential 20 s IR laser treatments with the 40 s intervals. IR laser power was gradually enhanced by increasing the diode current from 0.5 to 1.0 A with a 0.1 A step.

Calibration of IR laser power by determination of temperatures reached at certain IR laser diode currents was conducted by monitoring of currents flowing through open microelectrode as previously described [37]. Briefly, the ACSF in the perfusion chamber of the Scientifica SliceScopePro 2000 microscope was preheated up to 60 °C. Temperature of the ACSF was measured using a thermistor integrated in the Scientifica SliceScopePro 2000 microscope. The thermistor was placed into the perfusion chamber. Currents (I) flowing through open microelectrode at the constant voltage of 30 mV and corresponding temperatures (T) were measured while ACSF was passively cooled from 60 to 36 °C. Measured values were then used to build an Arrhenius plot. A linear approximation (y = mx + b, where y = ln(I) and x = T−1 (in K−1)) of these data by using the least squares method allowed for determination of an Arrhenius plot slope (m). Then, the calibrated microelectrode placed in the ACSF at the temperature of 35 °C (T0 = 308.15 K), and the baseline current (I0) was measured. The microelectrode underwent 20 s IR laser treatments with 40 s intervals. IR laser power was gradually enhanced by increasing the diode current from 0.5 to 1 A with a step of 0.1 A. Maximal currents (I) flowing through the open microelectrode were measured during periods when the IR laser was on. Temperatures reached at various IR laser power were then calculated by T = [1/T0 − R/Ea × ln(I/I0)]−1 as previously described [37]. Here, R is the gas constant, Ea is the activation energy of electrolyte, and T0 and I0 are the baseline temperature of the ACSF and the corresponding electrode current, respectively. The value of (-R/Ea) is equal to (1/m), where (m) is a slope of the Arrhenius plot.

Analysis of calcium imaging on cultured cells and acute brain slices

The image processing Fiji software (RRID: SCR_002285) was used to create quantitative datasets for determining the kinetics of the GCaMP6s signal in transfected HEK293TN cells and transduced hippocampal neurons.

For cultured cells, widefield image files in the Nikon nd2 format were imported into the Fiji software using Bio-formats importer. Files for experimental cultured cells expressing both hTRPV1-p2A-tdTomato and GCaMP6s contained images acquired in two channels: green—GCaMP6s, and red—tdTomato, whereas control cultured cells expressing GCaMP6s contained images acquired in one channel: green – GCaMP6s. The background was subtracted straightaway using the Sliding paraboloid item with a Rolling ball radius of 50.0 pixels. Cell bodies with GCaMP6s and tdTomato fluorescence were selected using the wand tool utility. The selections were used as regions of interest (ROIs) and added to the ROI manager of Fiji software. Then, channels were split, mean values of fluorescence intensity for selected ROIs were measured, and digitized fluorescence datasets for each ROI in the green channel were exported using the Multi measure tool. To normalize GCaMP6s fluorescence intensity data for a given ROI from time-lapse imaging, the minimal fluorescence intensity was determined within the time interval from 30 to 60 s prior to the cells underwent the first heating from 37 to 39 °C. Then, GCaMP6s fluorescence intensity values at every time-point for a given ROI were divided by this minimal fluorescence intensity to determine the kinetics of the GCaMP6s signal. Individual kinetic curves were averaged and combined with temperature measurements to provide a resultant response of experimental (hTRPV1⁺) or control cells to quick heating up to 39, 41, and 43 °C from the baseline temperature of 37 °C. Heat maps were created using the free web-enabled software Heatmapper [38].

For neurons, widefield image files in the uManager tiff format were opened in the Fiji software. A patched GCaMP6s⁺ neuron and a field of view without the fluorescent signal from neuronal cell bodies was manually selected as ROIs. The selected ROIs were added to the ROI manager of Fiji software. Then, mean values of fluorescence intensity for selected ROIs were measured, and digitized fluorescence datasets for each ROI were exported using the Multi measure tool. Signal of the background ROI was subtracted from signal of the patched GCaMP6s⁺ neuron. To determine the kinetics of the GCaMP6s signal for a given patched neuron, GCaMP6s fluorescence intensity values at every time-point were divided by its fluorescence intensity in the first image of the time stack. Individual kinetic curves were combined with electrophysiological traces and temperature measurements to provide a resultant response of hTRPV1⁺ or control neurons maintained at the baseline temperature of 35 °C to heating by IR laser treatments.

In vivo thermogenetic stimulation and fiber photometry

Prior to the first day of thermogenetic stimulation, mice were allowed for adaptation to the experimental conditions during two consecutive days (Fig. 3A). Freely moving mice were manually restrained, the long optical fiber coupled to the optical stimulation/photometry setup was connected to the implanted fiber-optic interface, and mice were then placed into an arena (40 cm × 40 cm) for a 5 min habituation (Fig. 3C). This optical stimulation/photometry setup provides simultaneous the IR laser irradiation and excitation and detection of the integral GCaMP6s signal (Fig. 3C).

hTPRV1+ and control mice connected to the optical stimulation/photometry setup via an optical fiber were placed in the arena and allowed for adaptation for 5 min prior to the beginning of the experiment. Then, we started video recording of mouse behavior and capturing of the integral GCaMP6s signal. Each mouse underwent three sessions of stimulation with different IR laser powers: 20, 40, and 60 mW (Fig. 3D). The stimulation was performed by continuous IR laser irradiation of the brain tissue. Each stimulation session consisted of three periods when the IR laser was on. The duration of the on-periods commonly was 20 s. If we observed that a mouse ran too fast in response to stimulation and there was a risk of getting tangled up in the optical fiber, we switched off the laser prior to a 20 s stimulation period over. The on-periods were separated by the 30-s off-periods. Experimental and control mice underwent to the same stimulation protocol for three consecutive days followed by euthanasia on the next day after the last stimulation. Their brains were collected for subsequent immunohistochemical analysis.

Locomotor activity of hTPRV1+ and control mice was evaluated in captured movies. Image processing was carried out using the Bonsai software (RRID:SCR_017218). An area containing the bottom of the arena was identified on each frame of a movie and cropped into a separate image. Then, it the image was smoothed with a spatial Gaussian filter with a kernel size corresponding to a size of a mouse. The latter allowed for blurring an optical fiber coupled to a mouse to eliminate its interference with an image of a mouse. Since a mouse is much darker than the arena, we used brightness threshold to find the position of the center of mass for each animal. Then, animal’s actual speed was determined by measuring changes in this position on sequential frames using the MATLAB software (RRID:SCR_001622). A final plot of animal’s actual speed was smoothed into a 1-s window by a moving average filter.

The time-course of the integral GCaMP6s signal in the CnF of hTPRV1+ and control mice was determined as follows. On each frame, an image of the optical fiber tip was cropped into a separate image, and the GCaMP6s signal was calculated by integration of signals from individual pixels. The mean value of the integral GCaMP6s signal F0 was determined for the time interval prior to the first round when the IR laser was on for a given thermal stimulation session. Then, GCaMP6s signals for a given thermal stimulation session were calculated as (F-F0)/F0, where F is an integral GCaMP6s signal for an individual time point, to determine the time-course of the integral GCaMP6s signal.

Mean values of animal’s actual speed and the integral GCaMP6s signal were calculated separately for the periods of time when the IR laser was on and the periods of time when the IR laser was off. Averaging of individual mean values of animal’s speed and the integral GCaMP6s signal provided resultant locomotor and calcium responses to the thermal stimulation with a given IR laser power. All calculations were conducted using the MATLAB software.

The relative speed increase was determined as the ratio of the average speed during periods when the IR laser was on to the average speed during periods when the IR laser was off for individual hTPRV1+ mice. Averaging of individual values enabled us to evaluate whether the stimulation effects of IR laser treatments remained stable accross at least three consecutive stimulation days.

To assess the potential anxiolytic effect of thermogenetic activation of CnF neurons, we determined the time spent by hTRPV1⁺ mice in the central area (20 cm × 20 cm) and near the walls of the arena for the periods when the IR laser was on and the periods when the IR laser was off. Then, we calculated ratios of the time spent by mice in the central area to the time spent by mice near the walls. Averaging of individual values allowed us to elucidate whether thermogenetic stimulation produced an anxiolytic effect in mice.

To verify the specificity of thermogenetic activation, separate cohorts of hTRPV1⁺ and control mice underwent a single stimulation session consisting of three 20 s periods when the IR laser was on (Fig. 6A). The laser power was 40 mW. hTRPV1⁺ and control mice were sacrificed 90 min after completion of the stimulation session for subsequent analysis of c-Fos expression.

The optical stimulation/photometry setup

The optical stimulation/photometry setup consisted of the following components (Fig. 3C): 1342 nm: IR continuous wave laser with wavelength 1342 nm (CNI, MIL-III-1342-1W); Arduino: Arduino controller; BP1 and BP2: band pass filter (Thorlabs, MF525-39 and Chroma, ET480/20x); C: collimator (Thorlabs, F950FC-A); Camera: CMOS camera (Thorlabs, CC505MU); CMOS: CMOS camera (The Imaging Source, DMK 33UX226); DM1 and DM2: dichroic mirror (Thorlabs, DMLP490 and DMSP900); Fiber: optical fiber with core diameter 200 μm and NA = 0.22 (Thorlabs, FG200LEA); HWP: half wave plate (Thorlabs, WPH05M-1310); L1, L2, and L3: spherical lenses (Thorlabs, LSA01); LED: fiber coupled light emitting diode with central wavelength 470 nm (Thorlabs, M470F4) with a fiber (Thorlabs, M134L02); M1 and M2—dielectric mirror (Thorlabs, BB1-E02 and BB1-E04); Obj: microscope objective (Mitutoyo,10X Mitutoyo Plan Apochromat Objective, 480–1800 nm, NA = 0.26, WD = 30.5 mm, MY10X-823); PBSC: Polarizing Beamsplitter Cube (Thorlabs, PBS124).

For thermal stimulation, we used an IR laser at a wavelength of 1342 nm. The power of radiation was tuned by rotating the HWP in front of the PBSC. The IR radiation was expanded by a telescope based on L2 and L3 lenses and coupled into the optical fiber using mirrors and a microscope objective. A fiber coupled LED at a wavelength of 470 nm with a collimator was used to excite the calcium sensor GCaMP6s. The diode radiation was filtered out by a BP2 bandpass interference filter and combined with IR radiation using M1 and DM1 mirrors. The combined radiation was coupled to a core of a multimode optical fiber with a step-index profile with NA 0.22 and a core diameter of 200 μm. Average power of the LED radiation at the output of the fiber was 6 μW. Fluorescence of the calcium sensor GCaMP6s in the mouse brain was collected by the same optical fiber and transmitted backwards. It was separated from the excitation radiation, additionally filtered by the BP1 bandpass filter, and focused by the L1 lens on the sensor of the cooled CMOS (main CMOS) camera. An additional CMOS camera was used for video recording of freely moving mice. Both cameras were synchronized at 30 frames per second. The stimulation was controlled by an Arduino microplate. Synchronization of the camera and the IR laser stimulation was implemented by the image of additional LED on the edge of an open field.

An implantable fiber-optic interface for in vivo thermogenetic stimulation

Implantable fiber-optic interfaces were prepared from step-index multimode fibers with a 200-μm pure-silica core (FG200LEA, Thorlabs, Newton, New Jersey). The fibers were finely processed for experiments using an in-house-built fiber-processing bench using in-house-tailored fiber components and fiberprocessing tools. As the first step of the fiber-preparation procedure, a factory-polished multimode patch cable was connected to a ceramic ferrule with a standard Lucent-connector-type ceramic sleeve. A short segment of bare optical fiber was then inserted inside the ceramic ferrule and fixed using a drop of UV-cured glue at the base of the ferrule. The length of each bare-fiber segment was chosen in such a way as to provide optical interrogation of a specific targeted area inside mouses brain. To this end, the length of each bare-fiber segment was adjusted with an accuracy of about 30 μm using a fiber cleaver (Thorlabs, XL411), driven by a computer-controlled motorized translation stage (Thorlabs, MT1/M-Z8) on an in-house-assembled fiber-cleavage frame. Transmission properties of each fiber probe were carefully characterized for calcium sensor pump radiation (470 nm LED), using a power meter (PM100D with an S120VC power sensor, Thorlabs). Only fiber probes with an overall transmission above 85% were used for implantation and subsequent measurements.

Stainless steel head plate installation

For in vivo thermometry, we installed stainless steel head plates onto exposed skulls of three mice by a standard stereotactic surgery (for details see the item “Stereotaxic viral vector delivery and implantation of a fiber-optic interface”). Stainless steel head plates were attached to the surface of aligned and dried skulls by the dental cement, leaving the point with stereotactic coordinates − 4.8 mm anteroposterior and 1.1 mm mediolateral from bregma easily available. This point was marked for subsequent identification. After surgery, mice were returned to their home cages and allowed for recovery during a week with appropriate treatment for analgesia.

In vivo thermometry

The optical setup for all-optical thermometry consisted of the following components (Fig. 7A): 1342 nm: IR continuous wave laser with the wavelength of 1342 nm (CNI, MIL-III-1342-1W); 473 nm: continuous wave laser with wavelength 473 nm (Cobolt Blues 50 473); AOM: acousto-optical modulator (ISOMET IMAD -P80L -1.5); DM1 and DM2–dichroic mirror (Thorlabs, DMLP490 and DMSP900); Fiber: dual cladding optical fiber with core diameter 25 μm and NA = 0.07, and inner cladding diameter 250 and NA = 0.48 Fiber (P-25/250DC-PM Thorlabs); HWP: half wave plate (Thorlabs, WPH05M-1310), PBSC—Polarizing Beamsplitter Cube (Thorlabs, PBS124); L1, L2 and L3 -spherical lenses (Thorlabs, LSA01); LP: low pass filter (Thorlabs, FELH0500); M1 and M2: dielectric mirror (Thorlabs, BB1-E02 and BB1-E04); Obj: microscope objective (Mitutoyo,10X Mitutoyo Plan Apochromat Objective, 480–1800 nm, 0.26 NA, 30.5 mm WD, MY10X-823); PG: pulse generator (DG645, Stanford Research Systems); SM: spectrometer (OceanOptics USB2000).

As in the experiment with thermogenetic modulation of mouse behavior, heating was carried out using continuous wave IR laser radiation with a wavelength of 1342 nm. The power of radiation was tuned by rotating the half-wave plate HWP in front of the polarizing beam splitter PBSC. The IR radiation expanded by a telescope was coupled into the inner cladding of a dual cladding optical fiber (Fig. 7B). For carrying out fully optical thermometry, a diamond microcrystal with NV0 centers was attached to the core at the output endface of the dual cladding optical fiber (P-25/250DC-PM Thorlabs) in the cyanoacrylate glue vapor. The microcrystal had a diameter of approximately 30 µm. The dual cladding optical fiber had the core diameter of 25 μm and the numerical aperture of 0.07, and the inner-cladding diameter of 250 µm and the numerical aperture of 0.48.

Mice with preliminary installed stainless steel head plates received an intraperitoneal injection of ketoprofen (5 mg/kg) for analgesia and then were anesthetized with isoflurane (3% for induction and 1.5% for maintenance). Their heads were secured in a head fixer of a custom airlifted platform [39]. We identified the earlier marked point on the skull, added a bupivacaine droplet, and drilled a hole in the skull. The dual cladding optical fiber with the attached diamond microcrystal was slowly deepened to the depth of 2.65 mm relatively to bregma. Then, anesthesia was interrupted, and mice with fixed heads were allowed for waking up. Awake mice were exposed to 20 s IR laser pulses with the laser power of 20, 40, and 60 mW (10 repeats per power value). Temperature kinetics for each IR laser power was determined by measuring the zero-phonon line ZPLNV⁰ parameters in the photoluminescence spectrum from diamond microcrystal. Immediately after in vivo thermometry, mice were deeply anesthetized by an intraperitoneal injection of a mixture of tiletamine/zolazepam/xylazine at doses of 40/40/20 mg/kg. The implanted fiber-optic probe was removed from the brain. Deeply anesthetized mice were sacrificed by decapitation. The dual cladding optical fiber with the attached diamond microcrystal was washed in saline and sterilized with 70% ethanol for the reuse. The repeated use of the same calibrated fiber-optic probe in several mice provided reproducibility of all-optical thermometry.

Changes of the zero-phonon line ZPLNV⁰ parameters in the photoluminescence spectrum from diamond microcrystal (Fig. 7D) allowed for determining relative alterations in temperature [40, 41]. The diamond microcrystal with NV0 centers was excited by a continuous-wave Nd:YAG laser with second-harmonic output at the wavelength of 473 nm. The radiation of the Nd:YAG laser was modulated by an acousto-optical modulator and coupled into the core of the dual cladding optical fiber by the mirror M1and the dichroic mirror DM1 for exciting the diamond microcrystal attached to the endface of the optical fiber (Fig. 7A). The power of the Nd:YAG laser was 50 mW. The diamond microcrystal photoluminescence was transmitted through the optical fiber back into the optical setup. The photoluminescence signal was then separated from the reflected laser beam by the dichroic mirror DM1 and the low pass filter LP and delivered to the spectrometer SM. The photoluminescence spectrum was approximated using the function

$$f\left(_,_,_,b,c,x\right)=_+_\left(x-c\right)+\frac_^}^+^},$$

(1)

where \(_\) and \(_\) are responsible for the background, \(_\) is the amplitude of the ZPLNV⁰, \(b\) is the full width at half maximum, and \(c\) is the ZPLNV⁰ position. The spectrum within the range from 565 to 585 nm was used for approximation. The dependence of the ZPLNV⁰ position from temperature for an individual diamond microcrystal was determined in a preliminary calibration experiment. Within the temperature range used for thermogenetic modulation of mouse behavior, this dependence was well described by a straight line with a slope of 0.01 nm/K.

Gating the IR laser diode current used for conducting heating of the brain tissue separately from all-optical thermometry in vivo. This allowed for eliminating the issues associated with charge state switching [42]. The operation cycle of the heating IR laser consisted of a 90 ms period when the laser was on followed by a 10 ms period when the laser was off. The Nd:YAG laser was turned on for 5 ms after the IR laser was turned off. The diamond microcrystal photoluminescence was recorded within the range from 1 to 4 ms after the Nd:YAG laser was turned on. Collected spectral data were used for calculating temperatures reached in the brain tissue. Operation cycles of both lasers and photoluminescence recording were synchronized by an external pulse generator. The photoluminescence recording delay after the IR laser was off was varied to ensure that we measure temperature changes in the brain tissue rather than temperature changes associated with IR-laser induced heating of the diamond crystal.

Theoretical model

The theoretical model is similar to the one used earlier to calculate the thermal penalty for optogenetic stimulation [43]. To analyze the heating in our thermogenetic stimulation protocol, we used mathematical modeling of IR radiation and heating transfer. The IR radiation transfer is analyzed in our model by means of Monte Carlo simulations with a Henyey–Greenstein scattering phase function [43, 44]:

$$p\left(\theta \right)=\frac\frac^}^-2gcos\left(\theta \right)\right)}^},$$

(2)

In our modeling to mimic brain we used anisotropy parameter g = 0.96 and scattering length ls = 4.8 mm−1 [44]. Light propagation in the brain tissue was modeled in cylindrical coordinates with absorbing ends. A cylinder had radius of 6 mm and thickness of 8 mm Cylindrical voxels were generate by space discretization with 5 µm steps. We also assumed that the optical fiber had a core diameter of 200 μm and numerical aperture of 0.2 and was placed to a depth of 1 mm from the brain surface. Then, photons propagated into the brain tissue at a random angle within the numerical aperture. When modeling, we used the absorption coefficient μa = 0.274 mm−1 [45, 46]. In the spectral range around the wavelength corresponding to the IR laser line (1342 nm), the absorption and scattering lengths of water are noticeably less than those of brain tissues. Therefore, the calculations are resistant to small variations in these parameters and correspond to the predictions of stimulation.

For temperature calculations, we used heat diffusion equation in a cylinder with radius of 6 mm and thickness of 8 mm and with absorbing boundaries. Heat transfer in the brain tissue was treated as a diffusion process using the well-known modification of the diffusion equation [43, 44, 47]:

$$\rho C\frac=\nabla k\nabla \text+___\left(_-T\right)+_+_,$$

(3)

where T is the local tissue temperature (37 °C in the absence of heat delivery); T0 is the temperature of the blood in the main arteries supplying the head, it was assumed to be constant and equal to 36.7 °C; and k = 0.527 mW m−1 °C−1, ρ = 1.04 10–6 kg mm−3, and c = 3.65 106 mJ kg−1 °C−1 are the thermal conductivity, density, and specific heat of the brain, respectively. Values for blood are given as ρb = 1.06 10–6 kg mm−3 and cb = 3.6 106 mJ kg−1 °C−1. The blood perfusion rate and the metabolic heat production in the tissue are wb = 8.5 10–3 s−1 and qm = 9.7 10–3 mW mm−3, respectively. I is intensity.

Immunohistochemistry and confocal imaging

Expression of hTPRV1 in acute brain slices and mouse brains, the inflammatory response of the brain tissue, and the appearance of a molecular marker of neuronal activity were evaluated by immunohistochemical staining and confocal microscopy. After electrophysiological recordings, acute brain slices were fixed for 24 h in 4% paraformaldehyde in 0.01 M phosphate buffered saline (PBS, pH 7.4) at 4 °C followed by three washings in PBS. Experimental and control mice subjected to in vivo thermal stimulation were deeply anesthetized by an intraperitoneal injection of a mixture of tiletamine/zolazepam/xylazine at doses of 40/40/20 mg/kg. Then, anesthetized mice underwent transcardiac perfusion with PBS followed by cold 4% paraformaldehyde in PBS (pH 7.4) according to standard procedures. Brains were extracted from skulls and postfixed overnight at 4 °C. On the following day, the brains were washed with PBS, and the hemispheres were sliced sagittally with a Leica VT1200S vibratome (Leica-Microsystems). Free floating slices (thickness of 50 μm) with an identified fiber-optic interface track surrounded by fluorescence of tdTomato were further processed for immunofluorescent staining with antibodies against FLAG-epitope (identification of TRPV1⁺ cells), GFP (identification of GCaMP6s⁺ cells), glial fibrillary acidic protein (GFAP) (identification of astrocytes), Iba-1 (identification of microglial cells), and c-Fos (a marker of neuronal activity).

Free floating brain slices collected after electrophysiological recordings and in vivo thermal stimulation were initially permeabilized in 5% Triton X-100 (Sigma-Aldrich, #T8787) in PBS for 1 h followed by blocking in 5% goat serum (SigmaAldrich, #G9023) in PBS containing 0.1% Triton X-100 for 1 h. Then, slices were incubated with the following primary antibodies overnight at room temperature: the rabbit anti-FLAG polyclonal antibody (1:500, Affinity, #T0053; RRID:AB_2843447); the chicken anti-GFP polyclonal antibody (1:500, Abcam, #ab13970; RRID:AB_300798), the rabbit anti-GFAP polyclonal antibody (1:1000, Abcam, #ab7260, RRID: AB_305808), the rabbit anti-Iba-1 monoclonal antibody (1:1000, Abcam, #ab178846, RRID:AB_2636859), and the guinea pig anti-c-Fos monoclonal antibody (Synaptic Systems, #226308, RRID:AB_2905595). Primary antibodies were diluted in the blocking solution overnight at room temperature. On the following day, slices were washed three times in PBS and stained with the following secondary antibodies for 2 h at room temperature: the goat anti-rabbit IgG (H + L) antibody conjugated with Alexa Fluor 647 (1:500, Thermo Fisher Scientific, #A-21245; RRID:AB_2535813), thegoat anti-chicken IgY antibody conjugated with Alexa Fluor 488 (1:500, Thermo Fisher Scientific, #A-11039; RRID:AB_2534096), the