An Ultra-Sensitive Step-Function Opsin for Minimally Invasive Optogenetic Stimulation in Mice and Macaques.

Resource Availability Lead Contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Guoping Feng ( fengg@mit.edu ). Materials Availability The Mouse line generated in this study have been deposited to Jackson Laboratory and named R26-LSL-SOUL-P2A-tdT mice, JAX Stock No. 032301.

Data and Code Availability

The custom written scripts supporting this study have not been deposited in a public repository because they require extensive user experience to navigate but are available from the corresponding author on request.

Experimental Model and Subject Details Hippocampal neuron cultures

Hippocampal neurons were prepared from postnatal day 0-3 Swiss Webster (Taconic) mice as previously described ( Chow et al., 2010 ; Klapoetke et al., 2014 ) but with the following modifications: dissected hippocampal tissue was digested with 100-200 units of papain (Worthington Biochem) for 8 min, and the digestion was stopped with ovomucoid trypsin inhibitor (Worthington Biochem). Cells were plated at a density of 52,000-64,000 per glass coverslip coated with Matrigel (BD Biosciences). Neurons were seeded in 75 ul Plating Medium containing MEM (500 mL, Life Technologies), glucose(2.5 g, Sigma), transferrin (50 mg, Sigma), Hepes (1.19 g, Sigma), glutagro (5 mL, 200 mM stock, Corning), insulin(1 mL, 12.5 mg/mL stock, Millipore), B27 supplement (10 mL, Gibco), heat inactivated fetal bovine serum (50 mL, Corning). After cell adhesion, additional 1 mL of Plating Medium was added. When glia density was 50-70% (about 2 days later), 1 mL of AraC Medium containing MEM (500 mL, Life Technologies), glucose (2.5 g, Sigma), transferrin (50 mg, Sigma), Hepes (1.19 g, Sigma), glutagro (1.25 mL, 200 mM stock, Corning), AraC (500 uL, 4 mM stock, Millipore), B27 supplement (10 mL, Gibco), heat inactivated fetal bovine serum (25 mL, Corning) was added. Neurons were grown at 37C degree and 5% CO2 in a humidified atmosphere. Cell plating density was determined using Countess II cell counter (Thermofisher).

Mouse lines

All husbandry and experimental procedures in this study were approved by the Committee for Animal Care of the Massachusetts Institute of Technology, conformed to the guidelines of the Division of Comparative Medicine, and were consistent with the Guide for Care and Use of Laboratory Animals (Eighth Edition, 2011). The following mouse lines were used to crossed with the SOUL knock-in mice generated in this study: Parvalbumin-IRES-Cre (PV-Cre, Stock No. 017320, The Jackson Laboratory), Drd1a-Cre (D1-Cre, GENSAT BAC transgenic EY217) or ChAT-IRES-Cre driver mice (ChAT-Cre, Stock No. 006410, The Jackson Laboratory), to generate PV-Cre:SOUL, D1-Cre:SOUL and ChAT-Cre:SOUL mice. C57Bl/6J mice (Stock no. 000664, The Jackson Laboratory) were used for virus injection and behavior test. Adult mice of both genders, between 12-20 weeks old were used. Mice were housed at 22-25 °C on a circadian cycle of 12-hour light and 12-hour dark with ad-libitum access to food and water, unless placed on the food restriction schedule. In selected experiments, mice were food deprived the night before and given ad libitum food during the experiment (fasting not to exceed 16 hours in total). Mice were housed with cage mates except after surgery when they were housed individually. Macaque All procedures were approved by the Committee for Animal Care of the Massachusetts Institute of Technology, conformed to the guidelines of the Division of Comparative Medicine, and in accordance with NIH guidelines for animal research. A 19-year old adult male rhesus macaque (Macaca mulatto) weighing 12.5 kg was used in the study. The animal was housed in a cage on a circadian cycle of 12-hour light and 12-hour dark, with ad-libitum access to food and water, as well as toys and other objects for environmental enrichment. Method Details Mouse Methods Cre-dependent SOUL knock-in mouse line The SOUL construct was created by introducing the T159C mutation into the SSFO (ChR2 C128S/D156A) cDNA by Quikchange mutagenesis. The SOUL cDNA was subcloned into the rAAV-hSyn1-P2A-tdTomato vector using NheI and HpaI restriction enzyme sites, thereby generating rAAV-hSyn1-SOUL-P2A-tdTomato construct. To create the Rosa26 targeting vector, first, the Ai9 targeting vector (a gift from Hongkui Zeng; Addgene vector #22799) was modified by replacing the FseI-tdTomato-FseI fragment with a new FseI-BstBI-EYFP-MluI-FseI fragment, yielding vector Ai9-EYFPmod. The SOUL-P2A-tdTomato expression cassette was then subcloned into Ai9-EYFPmod using BstBI and MluI restriction sites on the vector and BstBI and AscI restriction sites on the insert. The ligation reaction reconstituted the 5’ BstBI site but destroyed the 3’ MluI site in the resultant Ai9-SOUL-P2A-tdTomato targeting vector. Critical elements were verified by DNA sequencing. The targeting vector was linearized with KpnI and purified by phenol/choloroform extraction prior to electroporation into R1 mouse embryonic stem cells for homologous recombination. ES cells were grown under G418 selection, and correctly targeted ES cell clones were identified by PCR screening with primers R26SA-F5: 5’-TTGGTGCGTTTGCGGGGATG-3’ and CAG02B: 5’-GTTATGTAACGCGGAACTCC-3’ (~1.1 kb band), and subsequently injected into blastocysts derived from C57BL/6 donor mice. High percentage chimeric male mice were mated to wild-type C57BL/6 female mice for obtaining germline transmission of the Cre-inducible SOUL allele and establishment of the line. Subsequent genotyping was carried out by polymerase chain reaction using the following primers: Rosa01F 5’-CACTTGCTCTCCCAAAGTCG-3’, Rosa02B 5’-TAGTCTAACTCGCGACACTG-3’, and CAG-02B 5’-GTTATGTAACGCGGAACTCC-3’. The product from the endogenous Rosa26 allele is 550 bp and the product that is specific for the modified Rosa26 allele is 325 bp. The mice were backcrossed to C57B16/J mice for at least 11 generations prior to breeding animals for experiments. This SOUL knock-in mice will be available from The Jackson Laboratory as Stock No. 032301. In vitro recording Neurons were transduced at ~2 days in vitro with AAV encoding hSyn-SSFO-P2A-tdTomto (Boston Children’s Hospital Viral Core, 2 ul) or hSyn-SOUL-P2A-tdTomato (Boston Children’s Hospital Viral Core, 1.5 ul) per well. For the assessments of photocurrent and intrinsic properties on the cultured mouse primary hippocampal neurons, whole-cell patch clamp recordings were performed as described ( Nehme et al., 2018 ). Recording pipettes were pulled from thin-walled borosilicate glass capillary tubing (KG33, King Precision Glass, CA, USA) on a P-97 puller (Sutter Instrument, CA, USA) and had resistances of 3-5 MΩ when filled with internal solution (in mM: 128 K-gluconate, 10 HEPES, 10 phosphocreatine sodium salt, 1.1 EGTA, 5 ATP magnesium salt and 0.4 GTP sodium salt, pH=7.3, 300-305mOsm). The cultured cells were constantly perfused at a speed of 3 ml/min with the extracellular solution containing (in mM: 119 NaCl, 2.3 KCl, 2 CaCl2, 1 MgCl2, 15 HEPES, 5 glucose, pH=7.3-7.4, Osmolarity was adjusted to 325 mOsm with sucrose). All the experiments including the coverslips transferring and recordings were strictly performed at dark to avoid any activation by the light from the environments. Cells were visualized with a 40X water-immersion objective on an upright microscope (Olympus, Japan) equipped with IR-DIC. Recordings were made using a Multiclamp 700B amplifier (Molecular Devices, CA, USA) and Clampex 10.7 software (Molecular Devices, CA, USA). In voltage clamp mode, we delivered 5s 470 nm light pulses of different level of powers (3μW/mm 2 , 8μW/mm 2 , 20μW/mm 2 , 60μW/mm 2 , 1mW/mm 2 ) to photoactivate the cells, and used 5s 1mW/mm 2 589 nm orange light to deactivate the SSFO or SOUL. The membrane potential was held at −60 mV in the presence of 100 μM PTX and 50 μM NBQX to block GABAergic inhibitory and AMPA mediated excitatory synaptic transmission. Subsequent analysis was performed using Clampfit 10.7 software (Molecular Devices, CA, USA). The data were stored on a computer for subsequent off-line analysis. Any cells with Rs more than 20 MΩ at any time during the recordings were discarded.

Slice recording

Acute brain slices were prepared from 1.5 to 4-month-old mice. Animals were anesthetized by intraperitoneal injection of avertin (tribromoethanol, 20 mg/ml, 0.5 mg per g body weight) and perfused with ice-cold NMDG-based solution: 92 mM N-methyl-d-glucamine (NMDG), 2.5 mM KCl, 1.20 mM NH4PO4, 30 mM NaHCOs, 20 mM HEPES, 25 mM glucose, 2 mM thiourea, 5 mM Na-ascorbate, 3 mM Na-pyruvate, 0.5 mM CaCl 2 and 10 mM MgSO 4 (~300 mOsm, 7.2–7.4 pH). Following decapitation, brains were removed for coronal sectioning (300 pm) in the same NMDG-based solution using a Vibratome 1000 Plus, Leica Microsystems, USA. Slices were then recovered in carbonated regular aCSF: 119 mM NaCl, 2.5 mM KCl, 1.2 mM NaH2PO4, 24 mM NHCO3, 12.5 mM glucose, 2 mM MgSO4.7H2O, 2 mM CaCl2.2H2O (~300 mOsm, 7.2–7.4 pH) at 32–34°C for 10 min and transferred to room-temperature carbonated regular aCSF. Slices were allowed to recover for at least 1h prior to all recordings. Slices were transferred into a recording chamber (RC-27L, Warner Instruments) and constantly perfused at room temperature (20–24 °C) with carbonated regular aCSF at a rate of approximately 2 mL/min. Borosilicate glass recording microelectrodes (King Precision Glass) were pulled on a P-97 horizontal puller (Sutter Instruments) and backfilled with KGlu internal recording solution (145 mM K-Gluconate, 10 mM HEPES, 1 mM EGTA, 2 mM Mg-ATP, 0.3 mM Na 2 -GTP, and 2 mM MgCl 2 ). The internal pH was adjusted to ~7.3 with KOH and osmolarity adjusted to ~300 mOsm with K 2 SO4. The electrode tip resistance in the bath when filled with this internal solution was 3–5 MOhms. Recordings were performed in the dark after initially illuminating the slices with a custom mKate2 microscope filter (Ex max 590 nm, Em max 635 nm) and a mercury arc lamp. Cells were visually identified based on visualization of the tdTomato fluorophore. Recordings were initiated after seal rupture, initial stabilization, and equilibration of the whole-cell configuration for at least 3 min to allow dialysis of the internal recording solution. In order to measure the rheobase current, current clamp traces were recorded with 40 sweeps of current injections, each lasting 500 ms and starting from −150 pA with increments of 25 pA with (ON) or without (OFF) a 2s period of light pulse preceding the injection. Resting membrane potentials were determined in current clamp setting at baseline, after a 2s pulse of ON light (GFP filter), and after a 2s pulse of OFF light (mKate2). Photocurrents were measured in voltage clamp while holding the cell at −70V. The firing facilitation of SOUL was determined by examining the number of action potentials elicited upon current injection for several steps with or without prior 2s illumination with ON light (GFP filter) in voltage clamp configuration. For the membrane time constant, voltage-clamp trace recording followed a 2s ON light pulse (GFP filter) for 20-30 min, followed by a 2s OFF light pulse (mKate2) to return to baseline. The time constant was calculated by averaging the current for several time bins (each 1 min). The current of each bin was normalized to the peak current bin (first time bin). Tau (defined as the time when current is 1/e of max) was calculated from a mono-exponential fit. Data were acquired using a MultiClamp 700B amplifier and a Digidata 1440A. All analysis was performed using pCLAMP10 (Axon Instruments, Molecular Devices) and MATLAB.

Virus injection

Mice were primed for anesthesia in an induction chamber with a mix of 5% isofluorane/O 2 circulated at a rate of 1L/min. Deeply anesthetized mice were maintained under anesthesia with 1-1.5% isoflurane/O 2 and mounted on a stereotactic frame. The animals’ heads were shaved, and the remaining hair was removed with Nair. Body temperature was measured through a rectal probe and maintained using an electrical heating pad. Injections were performed with a glass micropipette (7-9 μm diameter). The injection speed was controlled at 100 nl/min with a micromanipulator (Quintessential Stereotaxic Injector, Stoelting). For in vivo electrophysiology experiment in mouse, titer-matched adeno-associated virus (AAV) encoding hSyn-SSFO-P2A-tdTomto (Boston Children’s Hospital Viral Core, total volume of 150 nl) or hSyn-SOUL-P2A-tdTomato (Boston Children’s Hospital Viral Core, total volume of 150 nl) was injected into the mediodorsal thalamus (MD) on opposite hemispheres of the same C57B1/6J mice at the following coordinate: anterior posterior (AP) −1.3 mm, medial lateral (ML) ± 0.6 mm, dorsoventral (DV) −3 mm. For the optical stimulation-evoked c-Fos staining experiments, AAV encoding hSyn-SOUL-P2A-tdTomato (Boston Children’s Hospital Viral Core, total volume of 120 nl) or mCherry (Penn Vector Core, total volume of 120 nl) was injected into the lateral hypothalamus (LH) of C57Bl/6J mice at the following coordinate: AP −1.6 mm, ML ±1.0 mm, DV −5.7 mm. For the feeding behavior test, AAV encoding Cre or mCherry under CaMKII promoter (both from Penn Vector Core, total volume of 120 nl) was injected into the bilateral LH at the following coordinate: AP −1.6 mm; ML ±1.0 mm; DV −5.7 mm. All viral vector titers were in the range of 2-9×10 13 genome copies (GC) per mL. Experiments were performed at least 2-4 weeks after virus injection. Implantation of sleeve and optical fiber(s) For optical stimulation of LH, a mating sleeve to be connected to the patch cable was implanted above the skull midline at AP = −1.6 mm. For in vivo comparison of SSFO and SOUL in MD, a 400 μm diameter optic fiber (Doric lenses, Quebec, Canada) was embedded centrally above the skull midline at AP −1.4 mm to deliver the transcranial stimulation and 200μm diameter fibers equipped with a 45 degree mirror tip were bilaterally implanted adjacent to the electrode arrays for direct stimulation of MD thalamus in each hemisphere separately. For the analysis of microglia activity in response to fiber implantation, an optic fiber was implanted into the cortex at the following coordinates: AP +1.6 mm, ML 1.0 mm, DV −0.5 mm. Mice had at least 2 weeks to recover after surgery. In vivo electrophysiology in mice For multi-electrode array construction and implantation, custom multi-electrode array scaffolds (drive bodies) were designed using 3D CAD software (SolidWorks) and printed in Accura 55 plastic (American Precision Prototyping) as described previously ( Brunetti et al., 2014 ; Liang et al., 2017 ). Prior to implantation, each array scaffold was loaded with 16–24 independently movable micro-drives carrying 12.5 μm nichrome (California Fine Wire Company) tetrodes. Electrodes were pinned to custom-designed 96- or 128-channel electrode interface boards (EIB, Sunstone Circuits) along with a common reference wire (A-M systems). For surgical implantation, after the preparation procedure similar to that of virus injection, an incision in the skin allowed access to the skull. Two ~1.2 x 1.6 mm craniotomies were drilled centered at (in mm from Bregma) AP −1.2, ML ±0.5. The dura was carefully removed, and the drive implant was lowered into the craniotomy using a stereotactic arm until the shortest tetrodes touched the cortical surface. Surgilube (Savage Laboratories) was applied around electrodes to guard against fixation through dental cement. Stainless-steel screws were implanted into the skull to provide electrical and mechanical stability and the entire array was secured to the skull using dental cement. The skin was subsequently closed with Vetbond and the animal was allowed to recover on a heating blanket. For electrophysiological recordings and spike sorting, signals were acquired using a Neuralynx multiplexing digital recording system (Neuralynx) through a combination of 32- and 64-channel digital multiplexing head stages plugged into the EIB of the implant. Signals from each electrode were amplified, filtered between 0.1 Hz and 9 kHz, and digitized at 30 kHz. Spike sorting was done automatically using MountainSort. Following sorting each cluster was manually inspected for quality. Clusters with spike waveforms that were symmetrical around their peak, indicative of an electrical noise signal, or showed inter-spike interval (ISI) distributions with more than 1% spikes