For experiments performed at The University of Tokyo, all methods for animal care and use were approved by the institutional review committees of School of Science, The University of Tokyo. For experiments at HHMI Janelia Research Campus, all surgical and experimental procedures were in accordance with protocols approved by the HHMI Janelia Research Campus Institutional Animal Care and Use Committee and Institutional Biosafety Committee. For experiments at University of Copenhagen, the procedures involving animal care, surgery, in vivo imaging, and sample preparation were approved by the local research ethics committee (Department of Experimental Medicine, University of Copenhagen) and conducted in accordance with the Danish Animal Experiments Inspectorate. All animal experiments at Kagoshima University were approved by the Experimental Animal Research Committee of Kagoshima University (Approval number: MD22086) and the Gene Recombination Experiment Safety Committee of Kagoshima University (Approval numbers: 22S018), and were performed in accordance with the ARRIVE guidelines, the National Institutes of Health (NIH) guide for the care and use of laboratory animals, and the Code of Ethics of the World Medical Association for animal experiments (http://ec.europa.eu/environment/chemicals/lab_animals/legislation_en.htm). All surgeries and experimental procedures at Stony Brook University were reviewed and approved by the Stony Brook University Animal Use Committee and followed the guidelines of NIH. The Stony Brook University IACUC-monitored Division of Laboratory Animal Resources maintained all animals. For experiments performed at University of Calgary, all methods for animal care and use were approved by the University of Calgary Animal Care and Use Committee and were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
General methods and materials
Synthetic DNA encoding LldR, CanlonicSF, Green Lindoblum, GEM-IL ver.3, and FiLa were purchased from Integrated DNA Technologies. A gene encoding Laconic was obtained from Addgene (Addgene plasmid #44238; http://n2t.net/addgene:44238). Phusion high-fidelity DNA polymerase (Thermo Fisher Scientific) was used for routine polymerase chain reaction (PCR) amplifications, and Taq DNA polymerase (New England Biolabs) was used for error-prone PCR. Restriction endonucleases, rapid DNA ligation kits, and GeneJET miniprep kits were purchased from Thermo Fisher Scientific. PCR products and products of restriction digests were purified using agarose gel electrophoresis and the GeneJET gel extraction kit (Thermo Fisher Scientific). DNA sequences were analyzed by DNA sequence service of Fasmac Co., Ltd. Fluorescence excitation and emission spectra were recorded on a Spark plate reader (Tecan).
Structural modeling of R-iLACCO1
Engineering of eLACCO2.1 and R-iLACCO1
A previously reported green extracellular l-lactate biosensor intermediate eLACCO0.9 (ref. 9) was subjected to an iterative process of library generation and screening in E. coli strain DH10B (Thermo Fisher Scientific) in LB media supplemented with 100 μg mL−1 ampicillin and 0.02% l-arabinose. Libraries were generated by error-prone PCR of the whole gene. Two rounds of the directed evolution led to the identification of eLACCO1.2. Cys340Ser mutation was introduced into eLACCO1.2. The resulting variant, designated eLACCO1.3, was subjected to an iterative process of library generation and screening in E. coli by error-prone PCR of the whole gene. For each round, ~200–400 fluorescent colonies were picked, cultured, and tested on 96-well plates under a plate reader. There were 6 rounds of screening before eLACCO2 was identified. Finally, Leu79Ile mutation was added to eLACCO2 to tune the l-lactate affinity. The resulting mutant was designated as eLACCO2.1. Asp444Asn mutation was added to eLACCO2 to abrogate the l-lactate binding. Three rounds of directed evolution were performed to improve the brightness. The resulting control mutant was designated as deLACCO1.
The gene encoding cpmApple with N- and C- terminal linkers (DW and EREG, respectively) was amplified using a red extracellular l-lactate biosensor R-eLACCO1 gene as template18, followed by insertion into the site between Leu109 and Val110 of LldR l-lactate binding protein in a pBAD (Invitrogen) by Gibson assembly (New England Biolabs). The resulting variant was designated as R-iLACCO0.1. N- and C-terminal linkers of R-iLACCO0.1 were deleted using Q5 high-fidelity DNA polymerase (New England Biolabs) to provide variants with different linker length. Variants were expressed in E. coli. Proteins were extracted using B-PER bacterial protein extraction reagent (Thermo Fisher Scientific) and tested for fluorescence brightness and l-lactate-dependent response. The most promising variant, designated as R-iLACCO0.2, was subjected to an iterative process of library generation and screening in E. coli. Libraries were generated by site-directed random mutagenesis or error-prone PCR of the whole gene. For each round, approximately 200–400 fluorescent colonies were picked, cultured, and tested on 96-well plates under a plate reader. There were 11 rounds of screening before R-iLACCO1, R-iLACCO1.1, and R-iLACCO1.2 were identified. Asp69Asn mutation was added to R-iLACCO1 to abrogate the l-lactate affinity. Two rounds of directed evolution were performed to improve the brightness. The resulting control mutant was designated as R-diLACCO1.
Protein purification and in vitro characterization
The gene encoding eLACCO2.1 and R-iLACCO variants, with a poly-histidine tag on the N-terminus, was expressed from the pBAD vector. Bacteria were lysed with a cell disruptor (Branson) and then centrifuged at 15,000 g for 20 min, and proteins were purified by Ni-NTA affinity chromatography (Agarose Bead Technologies). Absorption spectra of the samples were collected with a Lambda950 Spectrophotometer (PerkinElmer). To perform pH titrations for eLACCO2.1, protein solutions were diluted into buffers (pH from 4 to 11) containing 30 mM trisodium citrate, 30 mM sodium borate, 30 mM MOPS, 100 mM KCl, 1 mM CaCl2, and either no l-lactate or 100 mM l-lactate. To perform pH titrations for R-iLACCO variants, protein solutions were diluted into buffers (pH from 4 to 11) containing 30 mM trisodium citrate, 30 mM sodium borate, 30 mM MOPS, 100 mM KCl, and either no l-lactate or 100 mM l-lactate. Fluorescence intensities as a function of pH were then fitted by a sigmoidal binding function to determine the pKa. For l-lactate titration of eLACCO2.1, buffers were prepared by mixing an l-lactate (−) buffer (30 mM MOPS, 100 mM KCl, 1 mM CaCl2, pH 7.2) and an l-lactate (+) buffer (30 mM MOPS, 100 mM KCl, 1 mM CaCl2, 100 mM l-lactate, pH 7.2) to provide l-lactate concentrations ranging from 0 mM to 100 mM at 25 °C. For l-lactate titration of R-iLACCO variants, buffers were prepared by mixing an l-lactate (−, w/o Ca2+) buffer (30 mM MOPS, 100 mM KCl, pH 7.2) and an l-lactate (+, w/o Ca2+) buffer (30 mM MOPS, 100 mM KCl, 100 mM l-lactate, pH 7.2) to provide l-lactate concentrations ranging from 0 mM to 100 mM at 25 °C. Fluorescence intensities were plotted against l-lactate concentrations and fitted by a sigmoidal binding function to determine the Hill coefficient and apparent Kd. For Ca2+ titration, buffers were prepared by mixing a Ca2+ (−) buffer (30 mM MOPS, 100 mM KCl, 10 mM EGTA, 100 mM l-lactate, pH 7.2) and a Ca2+ (+) buffer (30 mM MOPS, 100 mM KCl, 10 mM CaEGTA, 100 mM l-lactate, pH 7.2) to provide Ca2+ concentrations ranging from 0 μM to 39 μM at 25 °C. Buffers with Ca2+ concentrations more than 39 μM was prepared by mixing a Ca2+ (−, w/o EGTA) buffer (30 mM MOPS, 100 mM KCl, 100 mM l-lactate, pH 7.2) and a Ca2+ (+, w/o EGTA) buffer (30 mM MOPS, 100 mM KCl, 100 mM CaCl2, 100 mM l-lactate, pH 7.2).
Two-photon excitation spectra and two-photon absorption cross sections of eLACCO2.1 and R-iLACCO1 were measured according to the general methods and protocols described in ref. 44. Briefly, a tunable femtosecond laser InSight DeepSee (Spectra-Physics, Santa Clara, CA) was used to excite fluorescence of the sample contained within a PC1 Spectrofluorometer (ISS, Champaign, IL). To measure the two-photon excitation spectral shapes, we used short-pass filters 633SP and 770SP in the emission channel. LDS 798 in 1:2 CHCl3:CDCl3 and Coumarin 540 A in DMSO were used as spectral standards. Quadratic power dependence of fluorescence intensity in the proteins and standards was checked at several wavelengths across the spectrum. The two-photon cross section (σ2) of the anionic form of the chromophore in eLACCO2.1 and R-iLACCO1 was measured as described in ref. 45. Rhodamine 6 G in methanol was used as a reference standard with excitation at 1060 nm (ref. 44). For one-photon excitation, we used a 532-nm line of diode laser (Coherent). A combination of filters 770SP and 561LP was used in the fluorescence channel for these measurements. Extinction coefficients were determined by alkaline denaturation as detailed in ref. 46. The two-photon absorption spectra were normalized to the measured σ2 values. To normalize to the total two-photon brightness (F2), the spectra were then multiplied by the quantum yield and the relative fraction of the respective form of the chromophore for which the σ2 was measured. The data is presented this way because eLACCO2.1 and R-iLACCO1 contain a mixture of the neutral and anionic forms of the chromophore. The method is described in further detail in refs. 46,47.
Construction of mammalian expression vectors
For cell surface expression, the genes encoding eLACCO2.1 were amplified by PCR followed by digestion with BglII and EcoRI, and then ligated into a pcDNA3.1 vector (Thermo Fisher Scientific) that contains N-terminal leader sequence and C-terminal anchor. N-terminal leader sequences were derived from ref. 18 and C-terminal anchors stemmed from refs. 18,20. To construct R-iLACCO variants for intracellular expression, the gene encoding R-iLACCO variants in the pBAD vector was digested with XhoI and HindIII, and then ligated into pcDNA3.1 vector. To construct plasmids for expression in primary neurons, astrocytes, ex vivo, and in vivo, the gene encoding R-iLACCO variants or eLACCO variants including the leader and anchor sequence in the pcDNA3.1 vector was first amplified by PCR followed by digestion with NheI and XhoI for eLACCO and BamHI and HindIII for R-iLACCO, and then ligated into a pAAV plasmid containing the human synapsin (hSyn) or gfaABC1D promoter. To construct plasmids for expression in principal cortical neurons in vivo, the gene encoding eLACCO variants including the leader and anchor sequence in the pcDNA3.1 vector was first amplified by PCR followed by digestion with BamHI and HindIII, and then ligated into a pAAV plasmid containing the human Ca2+/calmodulin-dependent protein kinase II α (CaMKIIα, 1.3 kbp) promoter.
Imaging of eLACCO and R-iLACCO variants in HeLa, HEK293T, and T98G cell lines
HeLa (American Type Culture Collection, ATCC, #CCL-2) and HEK293T (Thermo Fisher Scientific, #R70007) cells were maintained in Dulbecco’s modified Eagle medium (DMEM; Nakalai Tesque) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich) and 1% penicillin–streptomycin (Nakalai Tesque) at 37 °C and 5% CO2. T98G cells (ATCC, #CRL-1690) were maintained in minimum essential medium (Nakalai Tesque) supplemented with 10% FBS, 1% penicillin–streptomycin, 1% non-essential amino acid (Nakalai Tesque) and 1 mM sodium pyruvate (Nakalai Tesque) at 37 °C and 5% CO2. Cells were seeded in 35-mm glass-bottom cell-culture dishes (Iwaki) and transiently transfected with the constructed plasmid using polyethyleneimine (Polysciences). Transfected cells were imaged using a IX83 wide-field fluorescence microscopy (Olympus) equipped with a pE-300 LED light source (CoolLED), a 40× objective lens (numerical aperture (NA) = 1.3; oil), an ImagEM X2 EM-CCD camera (Hamamatsu), Cellsens software (Olympus), and an STR stage incubator (Tokai Hit). The filter sets used in live cell imaging had the following specification. eLACCO variants, iLACCO1, CanlonicSF, Green Lindoblum, FiLa (ex. 470 nm), and EGFP: excitation 470/20 nm, dichroic mirror 490-nm dclp, and emission 518/45 nm; R-iLACCO variants, mRFP1, and mApple: excitation 545/20 nm, dichroic mirror 565-nm dclp, and emission 598/55 nm; GEM-IL and Laconic (CFP): excitation 438/24 nm, dichroic mirror 458-nm dclp, and emission 483/32 nm; Laconic (FRET): excitation 438/24 nm, dichroic mirror 458-nm dclp, and emission 542/27 nm; FiLa (ex. 405 nm): excitation 405/20 nm, dichroic mirror 425-nm dclp, and emission 518/45 nm. Fluorescence images were analyzed with ImageJ software (National Institutes of Health).
For photostability test, HeLa cells transfected with respective genes were illuminated by excitation light at 100% intensity of LED (~10 mW cm−2 and ~4 mW cm−2 on the objective lens for eLACCO and R-iLACCO variants, respectively) and their fluorescence images were recorded for 4 min with the exposure time of 50 ms and no interval time.
For imaging of l-lactate-dependent fluorescence changes, HeLa cells seeded onto coverslips were transfected with plasmids encoding eLACCO or R-iLACCO variants. Forty-eight hours after transfection, the coverslips were transferred into Attofluor™ Cell Chamber (Thermo Fisher Scientific) with Hank’s balanced salt solution (HBSS(+); Nakalai Tesque) supplemented with 10 mM HEPES and 1 μM AR-C155858 (Tocris). Exchange of bath solutions during the image was performed in a remove-and-add manner using a homemade solution remover9.
For imaging of Ca2+-dependent fluorescence, HeLa cells on coverslips were transferred into AttofluorTM Cell Chamber with HBSS(−) buffer (Nakalai Tesque) supplemented with 10 mM HEPES and 10 mM l-lactate, 48 h after transfection of eLACCO variants. Other bath solutions were supplemented with Ca2+ of 100 nM, 1 μM, 10 μM, 100 μM, 300 μM, 1 mM, 3 mM, and 10 mM. Exchange of bath solutions during the image was performed in the same way as the imaging of l-lactate-dependent fluorescent changes.
Imaging of eLACCO and R-iLACCO variants in primary neurons and astrocytes
The neuron imaging was previously described9. In short, rat cortical/hippocampal primary cultures from the P0 pups (pooled tissues from males and females) were plated in glass-bottom 24-well plates where 0.5 × 106 cells were used for three wells. Cultures were nucleofected at time of plating, and imaged 14 days later. For the imaging of primary astrocytes, male and female pups were obtained from a single timed-pregnant Sprague Dawley rat (Charles River Laboratories, purchased from Japan SLC, Inc.). Experiments were performed with cortical/hippocampal primary cultures from the E21 (after C-section of the pregnant rat) plated in glass-bottom 24-well plates (Cellvis) where 0.5 × 106 cells were used for one well. Cultures were nucleofected at time of plating with Nucleofector 4D (Lonza), and imaged 5 days later. Astrocytes were cultured in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin at 37 °C and 5% CO2. Culture media were replaced with 1 mL of imaging buffer (145 mM NaCl, 2.5 mM KCl, 10 mM d-glucose, 10 mM HEPES, 2 mM CaCl2, 1 mM MgCl2, pH 7.4) for imaging.
Imaging of R-iLACCO1.2 in myotubes
Mouse C2C12 myoblasts (#CRL-1772, ATCC) were grown in DMEM supplemented with 10% FBS before induction of differentiation. After reaching confluence 2 days post-seeding, the culture medium was replaced with a differentiation medium (DMEM supplemented with 1% horse serum (Gibco)), and the medium was changed every 2 days until 6 days after differentiation. Retroviruses were used to express R-iLACCO1.2 and R-diLACCO1 in C2C12 myotubes. To construct retrovirus vectors, the genes encoding R-iLACCO1.2 and R-diLACCO1 were amplified by PCR followed by digestion with BamHI and EcoRI and then ligated into a pMX vector. Retrovirus was produced using plat-E packaging cells (Cell Biolabs) which were transfected with pMX plasmids using TransIT-LT1 reagent (TakaraBio). Retrovirus-containing supernatants were collected 60 h after transfection. To transduce retrovirus, undifferentiated C2C12 myoblasts were plated in a 35-mm dish at 40% confluency. Retrovirus was added to the medium three times at intervals of 6 h, and the medium was changed to culture medium 12 h after the final infection. For the observation of glucose-dependent fluorescence in myotubes, undifferentiated myoblasts expressing R-iLACCO1.2 or R-diLACCO1 were seeded to eight-well μ-Slide (ibidi) and differentiated into myotubes. Culture media was changed to imaging buffer without glucose (Krebs-Ringer phosphate buffer; 5 mM KH2PO4, 136 mM NaCl, 4.7 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 25 mM NaHCO3, 20 mM HEPES) 2 h before observation. Myotubes were imaged using an FV-1200 laser-scanning confocal fluorescence microscope (Olympus) equipped with a UPLXAPO 10× air/dry objective (NA 0.40), and 559-nm laser for R-iLACCO1.2 and R-diLACCO1 excitation. Myotubes were maintained at 37 °C with a stage top incubator (STX, Tokai Hit).
Clustering analyses were performed using R (R version 4.2.3; R Core Team). For clustering analysis, the traces of ΔF/F from glucose-stimulated cells were used. Among 42 cells analyzed, 21 cells were pretreated with MCT inhibitor, AR-C155858. To group cells that showed similar fluorescence traces upon glucose addition, hierarchical clustering was performed using Euclidean distance by Ward’s method. The resulting dendrogram and a heatmap were generated using the R package “ggplot2”.
Transgenic line generation in Drosophila melanogaster
The coding sequence of the corresponding biosensor was amplified from pcDNA3.1 by PCR and subsequently inserted into EcoRI/XbaI site of pUAST-attB vector using In-Fusion Snap Assembly Master Mix (Takara). The transgenic lines were generated via Φ integrase-mediated recombination into the fly genome at landing site attp2 (Rainbow Transgenic Flies Inc).
Live imaging of ex vivo Drosophila adult brains
Live imaging of ex vivo Drosophila adult brains (3–5 day old females) were performed at room temperature with LSM800 confocal microscope (Zeiss). Whole brain images were captured with a 20× Plan-Apochromat objective (NA 0.8) in single plane mode (scan speed 8; 1024 × 1024-pixel; 8 bits per pixel; averaging number 16; pinhole 1–1.5 AU). Time series images were taken every 60 s for 5 min with a 63× Plan-Apochromat Oil DIC objective (NA 1.4) in z-stack mode (scan speed 8; 1024 × 1024-pixel; 8 bits per pixel; averaging number 4; pinhole 1–1.5 AU; z-stack 5 slices). Adult brains were quickly dissected in Schneider’s Drosophila Medium (Gibco) and then put on a poly-lysine coated glass bottom dish (Matsunami Glass Ind. Ltd.) filled with HL3 buffer (70 mM NaCl, 5 mM KCl, 20 mM MgCl2, 10 mM NaHCO3, 115 mM sucrose, 5 mM HEPES; pH 7.2) supplemented with 5 mM d-glucose, 1 mM l-lactate, and 0.5 mM pyruvate11. The dish was placed on the microscope stage, and the medium in the dish was replaced with an HL3 buffer containing 6 mM oxamate. After 5 min, l-lactate was added to a final concentration of 10 mM. Samples were excited at 561 nm, and the emission was recorded at 566–700 nm.
Packaging and purification of adeno-associated viruses (AAVs)
AAVs were generated in HEK293T17 cells by a triple transfection of a helper plasmid pXX680, a plasmid encoding either the AAV2/5 or AAV2/9 rep/cap, and a pAAV plasmid encoding the corresponding LACCO biosensor. Forty-eight hours post transfection, the cells were harvested by gentle scraping and pelleted by centrifugation at 2300 × g. Viral particles were released by four freeze-thaw cycles on dry-ice/ethanol and free DNA was digested with benzonase. The AAV particles were purified by a discontinuous gradient of iodixanol (15-25-40-60%) and ultracentrifugation (462,000 × g for 70 min at 16 °C). The viral preparation was then washed and concentrated through a 100 kD amicon filter. During the last concentration step, the buffer was adjusted to 5% d-sorbitol and 0.001% pluronic acid for storage and experimentation. Titration was performed by TaqMan digital droplet-PCR using primers specific for AAV2 inverted terminal repeat (ITR).
Preparation of mouse acute cortical brain slice
At 4 weeks after the injection of AAV encoding hSyn-HA-eLACCO2.1-NGR or gfaABC1D-HA-eLACCO2.1-NGR, mice (C57Bl6, male, P28, housed at 21 °C with 47% humidity) were anaesthetized with gaseous isofluorane (5%) and then decapitated. The brain was removed, then submerged for 2 min in ice-cold slicing solution containing: 119.9 mM N-methyl-d-glucamine, 2.5 mM KCl, 25 mM NaHCO3, 1.0 mM CaCl2·2H2O, 6.9 mM MgCl2·6H2O, 1.4 mM NaH2PO4·H2O, and 20 mM d-glucose. The brain was then Krazy Glued onto a vibratome tray (Leica Instruments, VT1200S) and then resubmerged in ice-cold slicing solution. Acute coronal slices were prepared from the somatosensory cortex (400 μm thick) using a vibratome. The slices were incubated for 45 min at 33 °C in a recovery chamber filled with artificial cerebrospinal fluid (ACSF) containing: 126 mM NaCl, 2.5 mM KCl, 25 mM NaHCO3, 1.3 mM CaCl2·2H2O, 1.2 mM MgCl2·6H2O, 1.25 mM NaH2PO4·H2O, and 10 mM d-glucose. After recovery, slices were stained with astrocyte marker sulforhodamine 101 (SR101, 10 μM for 20 min, recovery for 1 h) to visualize astrocyte processes to maintain focal plane during the imaging experiments. Throughout, brain slices were continuously supplied with carbogen (95% oxygen, 5% CO2).
Two-photon microscopy of eLACCO2.1 in acute cortical brain slice
Brain slices were imaged using a custom built two-photon microscope fed by a Ti:Sapphire laser source (Coherent Ultra II, ~4 W average output at 800 nm, ~80 MHz). Image data were acquired using MATLAB (2013) running the open source scanning microscope control software ScanImage (version 3.81, HHMI/Janelia Research Campus). Imaging was performed at an excitation wavelength of 940 nm. The microscope was equipped with a primary dichroic mirror at 695 nm and green and red fluorescence was split and filtered using a secondary dichroic at 560 nm and two bandpass emission filters: 525–540 nm and 605–670 nm (Chroma Technologies). Time series images were acquired at 0.98 Hz with a pixel density of 512 by 512 and a field of view size of ~292 μm. Imaging used a ×40 water dipping objective lens (NA 1.0, WD 2.5 mm, Zeiss). Imaging was performed at room temperature.
For aglycemia experiments in acute brain slices, glucose was not added to the ACSF and in its place 5 mM extra NaCl was supplemented to maintain osmolarity (10 mosmols). For electrical afferent stimulation in acute slices, a Grass S88X stimulator, a voltage isolation unit, and a concentric bipolar electrode (FHC) assembly were used. The electrode was positioned onto the surface of the slice by a micro-manipulator ~300 µm away from the region of interest. The tissue was stimulated with 1 ms long mono-phasic pulses at 1.2 V using a theta burst pattern (100 Hz for 50 ms, repeated at 4 Hz). Theta burst was applied for 5, 15 and 30 s in separate experiments. The different lengths of stimulation produced similar results and the data was pooled.
In vivo validation of eLACCO2.1
Mice (C57BL6/J, male, 8–9 weeks, housed at 23 °C with 40–70% humidity) were anesthetized by intraperitoneal (i.p.) injection of the mixture consisting of 0.75 mg kg−1 medetomidine (Domitor® Nippon Zenyaku Kogyo Co., Ltd., Tokyo, Japan), 4.0 mg kg−1 midazolam (Dormicum®, Astellas Pharma Inc., Tokyo, Japan), and 5.0 mg kg−1 butorphanol (Vetorphale®, Meiji Seika Pharma, Co., Ltd., Tokyo, Japan). Holes were drilled on the skull. Glass needles (Ringcaps, Hirschmann Laborgeräte GmbH & Co. Eberstadt, Germany) were pulled by a puller (PE-22, Narishige, Tokyo, Japan) and stereotaxically introduced into the visual cortex. Half mL of AAV9-CaMKIIα-HA-eLACCO2.1-NGR or AAV9-CaMKIIα-HA-deLACCO1-NGR (4 × 1013 viral genome (vg) mL−1 each) was unilaterally injected, and needles were settled at the injected places for 10 min to defuse the viral vectors. After the viral injection, a fiber optic cannula (Lucir, Tsukuba, Japan) was implanted into the same visual cortex. For the intracerebroventricular (i.c.v.) injection, a guide cannula (Eicom, Kyoto, Japan) was placed into the lateral ventricle. The coordinates of specific brain regions were as follows: the visual cortex (lateral, +2.7 mm; posterior, −2.8 mm from the bregma and ventral, −0.4 mm from the brain surface), and the lateral ventricle (lateral, −1.0 mm; posterior, −0.2 mm from the bregma and ventral, −1.8 mm from the brain surface) according to the mouse brain atlas. A fiber optic cannula and an injection cannula were fixed on the mice’s skulls using dental composite resin. Subsequently, mice were injected with 0.75 mg kg−1 atipamezole (i.p., FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) to awaken them and 0.8 mg per mouse of penicillin G (s.c., FUJIFILM Wako Pure Chemical Corporation) to prevent infections. Three weeks later, the mice were subjected to i.c.v. l-lactate injection.
On the day of the experiment, an optical fiber was connected to a fiber optical cannula, and an injection cannula was inserted into a guide cannula. l-Lactate (300 nmol and 1 μmol) or saline was administered intracerebroventricularly via an injection cannula (1 mL min−1). eLACCO fluorescence was acquired by the fiber photometry system (COME2-FTR/GFP, Lucir, Tsukuba, Japan). A fiber-coupled LED (LEDD1B/M470F3, Thorlabs, Newton, NJ) produced excitation blue light (470 nm) of power 0.1 mW at the tip of the silica fiber; the blue light to pass through the excitation bandpass filter (passband 475 ± 12.5 nm), was reflected by a dichroic mirror-1, and joined the single silica fiber (diameter 400 µm; NA 0.6). The blue light emitted from the tip of the fiber optical cannula reflected the eLACCO fluorescent proteins; the reflected green fluorescence signals were transmitted to the same silica fiber. The signals to pass through the dichroic mirror, were reflected by dichroic mirror-2, and passed through the bandpass emission filter (passband: 510 ± 12.5 nm). Finally, the signals were guided to a photomultiplier (PMTH-S1-1P28, Zolix Instruments, Beijing, China). The signals were digitized by an A/D converter (PowerLab2/25, AD Instruments Inc., Dunedin, New Zealand) and recorded with the LabChart version-7 software (AD Instruments Inc.). Data was acquired at 1 Hz. Baseline fluorescence intensity (F0) was calculated as the average of the 10 s before l-lactate administration, and fluorescence intensity (F) at each time point was used to calculate ΔF/F = (F–F0)/F0. The area under the curve (AUC) was calculated by adding the ΔF/F over 5 min after l-lactate or saline administration.
In vivo eLACCO2.1 imaging
All mice were housed in a 12-h light/dark cycle condition at 18–26 °C with 30–70% humidity. All behavioral experiments were performed during the light cycle. Mice were provided ad libitum access to food and water. Surgeries were conducted using 8–10 week-old male mice (C57BL/6, Charles River Laboratories). Mice were anesthetized with isoflurane at 4% for induction and 1.5–2% during surgery with 0.4 L min−1 O2, while the depth of anesthesia was monitored by breathing rate. AAV9-hSyn-HA-eLACCO2.1-NGR or AAV9-hSyn-HA-deLACCO1-NGR was stereotaxically injected into the dorsal dentate gyrus (AP: −2.0, ML: 1.5, DV: 2.5). After about 1 h after viral injection, an integrated GRIN lens (outer diameter: 1.0 mm, length: 4.0 mm) was implanted approximately 200 mm above the dentate granule cell (DGC) layer. Mice were allowed to have 4 weeks for viral expression and recovery from surgery. All in vivo imaging videos were recorded from freely moving animals (4–5 weeks after surgery) in their home cage by using a miniature microscope and the nVue data acquisition system (Inscopix, Palo Alto, CA) at a frame rate of 20 Hz. For each recording, animals were allowed 10 min to familiarize the recording setup, and a 10 min baseline recording was performed before insulin treatment (i.p. injection, 0.75 U kg−1). The same volume of saline injection with some recording paradigm was set as a control experiment. The order for treatment and control recordings was randomly assigned. The fluorescence intensity of the region of interest of images was analyzed in the Inscopix Imaging Processing (ver. 1.8.0). MATLAB (R2022) was used for further analysis and the final data plot.
In vivo R-iLACCO imaging
Mice (C57BL/6 J, male, 8 − 15 weeks, housed at 21 °C with 40 − 60% humidity) were anesthetized by 3–4% isoflurane for induction and the anesthesia was maintained at 1–1.5% isoflurane while they were mounted to the stereotaxic frame. The body temperature was maintained at 37 °C throughout the surgery. The skull was exposed after applying local analgesia (lidocaine, 0.2 mg mL−1) by making an incision to the scalp, and a metal frame (head plate) was then attached to the skull using dental cement (Super Bond C&B, Sun Medical, Shiga, Japan). A 4-mm diameter craniotomy above the somatosensory cortex was made and the dura mater was surgically removed. AAVs were injected into the somatosensory cortex at depths of 300 μm (1.3–1.5 × 1012 vg mL−1 in PBS, 500 nL) using a glass pipette attached to a micromanipulator and a microinjector (FemtoJet, Eppendorf). A 4-mm diameter autoclaved cover slip was carefully mounted to cover the brain and then sealed by dental cement. Mice received a subcutaneous administration of carprofen (5 mg kg−1) for systemic analgesia after the surgery and were recovered for at least three weeks in their home cage. To probe the performance of l-lactate biosensors in response to sensory stimuli, shallowly anesthetized (70 mg kg−1 ketamine + 10 mg kg−1 xylazine) mice were rigidly fixed under the fluorescence macroscope (self built, Thorlabs). Sample was excited to collect the fluorescence using an LED (pE4000, CoolLED), a 488/561 dual band filter set (59904, Chroma), and an EM-CCD camera (Andor iXon 888, frame rate 10 Hz). For sensory stimulus, air puffs (50 ms, 20 psi, 5 Hz, Picospritzer, Parker) were presented to the mouse’s whisker pad for 30 s. The sensory stimulation session was repeated for 8 times including 30 s baseline and 120–180 s post-stimulation periods.
Statistics and reproducibility
All data are expressed as mean ± s.d. or mean ± s.e.m., as specified in figure legends. Box plots are used for Figs. 3c, f, g, j, k, m, n and 4b, d–j. In these box plots, the horizontal line is the median; the top and bottom horizontal lines are the 25th and 75th percentiles for the data; and the whiskers extend one standard deviation range from the mean represented as black filled circle. Sample sizes (n) are listed with each experiment. No samples were excluded from analysis and all experiments were reproducible. In the representative fluorescence images in Figs. 3a, b, h, i and 6b, e, h, similar results were observed in more than five cells. Statistical analysis was performed using two-tailed Student’s t test or one-way analysis of variance (ANOVA) with the Dunnett’s post hoc tests (GraphPad Prism 9), as specified in figure legends. Microsoft Excel software was used to plot for Fig. 1d, e, g, h.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.