| It has become increasingly popular for people to take some neurostimulant drugs as nootropics enhance cognition and learning beyond the initially approved that are expected to therapeutic purposes,such as curing attention deficits.Likewise,brain stimulation techniques,including,e.g.,transcranial magnetic stimulation and transcranial direct-current stimulation,have also been extensively practiced in patients with similar expectations.However,while those drugs or stimulation techniques could help alleviating the relevant deficits,their effects in diseased conditions do not necessarily imply that healthy subjects could benefit from them in boosting their normal learning capabilities.Learning is an intricate process that involves highly specific patterns of neuronal activation,and the neocortex is known to be functionally relevant for the associative learning process.In the past decade,specific manipulations of neuronal activities have been achieved by using the optogenetic technique,advancing the understanding of learning and memory formation.Although having become highly popular for animal experiments,the optogenetics technique shows little potential for applications in healthy humans due to the requirement of introducing exogenous genes in the brain.Here,we present a fundamentally different energy stimulation technique,mid-infrared modulation（MIM）,which delivers mid-infrared（MIR）light energy through the opened skull or non-invasively through the thinned intact skull to the brain and can significantly elevate neuronal firing rates in the targeted brain region.Notably,mid-infrared modulation induces neuronal firing in complete absence of any exogenous gene.As a striking example,we demonstrate that MIM application in the auditory cortex of healthy adult mice during a sound-licking associative learning task boosts learning speed by～50%,which will provide a new method for the application of mid-infrared modulation in clinical treatment and rehabilitation.Objectives:1.To explore the effect of MIM on cortical neurons in vivo.2.To explore the temporal and spatial distribution of the effects of MIM.3.To explore the mechanism of MIM on cortical neurons in vivo.4.To explore the effects of MIM on associative learning in mice.Methods:1.Mid-infrared light source.A quantum cascade mid-infrared（MIR）laser was used for this study.The laser contains four modules for continuous wavelength tuning but it was set to work constantly at 5.6μm wavelength for this study.The output was coupled with a MIR fiber（IRF-Se-100）with a core diameter of 100μm and a numerical aperture（NA）of 0.27.For all experiments in this study,the average power at the fiber tip（measured in the air）was at a stable level of 9±0.5m W with the following configurations:pulse duration 300 ns,repetition rate 100 k Hz（equivalent to 300 m W peak power at a duty cycle of 3%）.This MIR pulsing laser was initially designed for the purpose of infrared spectroscopy;when we used it for this study,all the parameters were configured to maximize the total average output power at the 5.6μm wavelength（spectral-wise as well,the laser delivers maximum output power at this wavelength）.2.Visible light source.As a control for the c-Fos imaging experiment,we used standard mono-wavelength visible（VIS）lasers with an optical fiber of the same geometry as that of the MIR fiber.The laser models were MBL-III-473-50m W and MBL-III-594-100m W with working wavelengths473 nm and 594 nm,respectively.These VIS lasers were delivered by an optical fiber that had the same geometric parameters as that for MIR light（core diameter 100μm,NA=0.27）.The output power at the fiber tip was adjusted and the exposure time was set to be the same as those of the MIR light（9 m W,20 s）.For c-Fos imaging experiments,we did not find a significant difference between the results obtained by using the two different visible wavelengths,and thus these data were pooled together.For temperature measuring,we used only the 594 nm one as the VIS light source.3.Immunohistochemistry.For the c-Fos experiments,the animal was first head-fixed under general anesthesia with isoflurane（1.5%）.The fiber tip was placed either above the brain surface via an open craniotomy or over the thinned intact skull.After the MIM application,the animal was kept under anesthesia for 90 min for c-Fos expression.The animal was then transcardially perfused with 4%paraformaldehyde（PFA）and the brain was fixed in 15%sucrose in 4%PFA and refrigerated overnight at 4°C.Coronal sections（thickness:30μm）encompassing the fiber target spot were made by a freezing microtome.The c-Fos immunostaining was performed with the primary antibody:Anti-c-Fos（ABE457,Millipore rabbit polyclonal antibody,Lot#3221531,1:500 dilution）,followed by the secondary antibody:Alexa Fluor 594 goat anti-rabbit（A11012,Invitrogen,and Lot#2119134,1:500 dilution）.DAPI（4’,6-diamidino-2-phenylindole,D9564,Sigma-Aldrich,1:1000 dilution）was used to stain cell nuclei for all c-Fos+cell counting experiments.In an independent set of control experiments,antibody Anti-Neu N（MAB377,Millipore,mouse polyclonal antibody,1:200 dilution）was used to label neurons,followed by the secondary antibody:Alexa Fluor 488 donkey anti-mouse（A21202,Invitrogen,Lot#1644644,1:500 dilution）.Sections were mounted on slides with coverslips and imaged using a scanning confocal microscope（TCS SP5,Leica）.Cells exhibiting both c-Fos and DAPI were defined as c-Fos positive cells,and cells exhibiting DAPI alone were defined as c-Fos negative cells.In the control experiments with c-Fos and Neu N co-labeling,cells exhibiting Neu N were defined as neurons.4.Cortical tissue temperature measurement.We performed the tissue temperature measurement protocols with an ADINSTRUMENT acquisition system（Power Lab 4/35）coupled to a T-type hypodermic thermocouple（MT 29/5,Physitemp）.The animal was first head-fixed under general anesthesia with isoflurane（1.5%）and the MIR（or VIS in an independent set of control measurements）fiber was placed above the cortical surface in the same configuration as in the c-Fos imaging experiments.The brain surface was immersed with ACSF solution that was circulated and controlled at 37°C.The thermo sensor was inserted into the cortex from a horizontal orientation by a fine micro-manipulator to adjust the depth and lateral position of the sensor tip.The dura mater under the craniotomy was removed for these measurements to expose the cortical tissue with a small bulge,allowing the sensor to be inserted from a horizontal orientation.We expected that the actual cortical tissue temperature change with intact dura in the other physiological experiments（c-Fos,loose-patch,two-photon imaging）would be less than what we measured in absence of dura.From the onset time of MIR or VIS light irradiation,the sensor readout reached a plateau in less than 10 s and we took this plateau value as the measured value.5.Loose-patch recording and TTX local application.For loose-patch recordings in cortical neurons in vivo,we used the“shadow-patching”procedure,except that we did not rupture the membrane of targeted cells to maintain a loose-patch configuration.Cell-attached recordings were performed with an EPC10 amplifier（HEKA Elektronik,Germany）.The glass electrode was filled wit h normal ACSF（with～10μM OGB1-6K dye to be visualized under a two-photon microscope）and had a tip resistance of 5–8 MΩ.The electrode penetrated the dura with a positive pressure of～100 mbar and was then reduced to～20–30 mbar to obtain good images of both electrode and shadow neurons under two-photon imaging.Raw signals were filtered at 10 k Hz and sampled at 20 k Hz using Patchmaster software（HEKA Elektronik,Germany）.In some experiments（6 neurons）,a second micropipette（the tip resistance of 5–8 MΩ）was inserted to be close to the recorded neuron,containing tetrodotoxin（TTX,10μM）which was injected for 10 s at a pressure of 50mbar.6.Two-photon Ca2+imaging.For acute two-photon Ca2+imaging experiments,we exposed the right auditory cortex of the mouse.In brief,the animal was anaesthetized by isoflurane and kept on a warm plate（37.5°C）.The skin and muscles over the Au1 were removed after local lidocaine injection.A custom-made plastic chamber was glued to the skull with cyanoacrylate glue（UHU）,followed by a small craniotomy（～2 mm×2 mm）（the center point:Bregma-3.0 mm,4.5mm lateral to the midline）.These stereotaxic coordinates correspond to a larger region of the auditory cortex,including the primary auditory cortex（Au1）,the dorsal and ventral secondary auditory cortex（Au D and Au V）as well as a part of the adjacent temporal association cortex（Te A）.After performing the craniotomy,the animal was transferred to the imaging rig with a head-fixation chamber.The chamber was perfused with normal artificial cerebral spinal fluid（ACSF）containing 125 m M Na Cl,4.5 m M KCl,26 m M Na HCO3,1.25 m M Na H2PO4,2 m M Ca Cl2,1 m M Mg Cl2 and 20 m M glucose（p H 7.4 when bubbled with 95%oxygen and 5%CO2）.The Ca2+dye,Cal-520 AM（AAT Bioquest）,was dissolved in DMSO with 20%Pluronic F-127 to a final concentration of 567μM for bolus loading.The loading procedure was conducted to label L2/3 neurons in Au1.The pipette containing dye was inserted into the cortex up to 200μm deep from the pial surface.About 2 h after dye injection to allow sufficient cellular uptake,two-photon imaging was performed with a custom-built two-photon microscope system based on a 12.0 k Hz resonant scanner（model“Lotos Scan 1.0”,Suzhou Institute of Biomedical Engineering and Technology）.Two-photon excitation light was delivered by a mode-locked Ti:Sa laser（model“Mai-Tai Deep See”,Spectra Physics）at a near-infrared（NIR）wavelength of 920 nm.A 40×/0.8 NA（Nikon）water-immersion objective was used for imaging,and this objective had a long working distance of 3.5 mm and a large access angle of～40 degree to allow micropipette manipulations from the side.We placed the fiber tip from an oblique angle to the cortical surface and protected it with a glass micropipette to avoid excess contact with ACSF.The typical size of the imaging field-of-view（FOV）was～200μm×200μm.In a typical experiment,time-lapse imaging recording at different focal depths could be performed sequentially.The average power of the output laser（under the objective）was in the range of 30–120 m W,depending on the depth of imaging.7.Associative training.For the sound-licking associative training experiments,we adapted the training protocol from our previous studies.Sound stimuli were delivered by an ED1 electrostatic speaker driver and a free-field ES1 speaker（both from Tucker Davis Technologies）.The distance from the speaker to the mouse ear（contralateral to the imaged A1）was～6 cm.The sound stimulus was produced by a custom-written,Lab VIEW-based program（Lab VIEW 2012,National Instruments）and transformed to analog voltage through a PCI6731 card（National Instruments）.Sound levels tested with a microphone placed～6 cm away from the speaker were calibrated by a pre-polarized condenser microphone（377A01 microphone,PCB Piezotronics Inc.）.For broadband noise（BBN,bandwidth 0–50 k Hz）,the sound level was～65 d B sound pressure level（SPL）.A waveform segment of BBN was first generated,and the same waveform segment was used for all experiments（i.e.,a“frozen noise”）.The duration of a sound stimulus was 50 ms.Before training,the animal was implanted with the headpost under isoflurane anesthesia and then allowed to recover for 5 days.After one night（20:00–08:00）of water restriction,the mouse was habituated for head fixation for 2–3 days.During habituation,the mouse received water in the behavioral setup exclusively.For animals in the“MIM-opened”or“MIM-thinned”group,the fiber tip was fixed above the cortical surface or the thinned section of the skull,respectively,before the beginning of each session from session#2（and was removed after the end of session）.The stereotaxic coordinates used for the placement of the fiber tip were the same as those for two-photon Ca2+imaging experiments in the auditory cortex.Careful management of water consumption was taken to ensure that animals in all groups received similar volume of water and had a similar level of motivation throughout all sessions and days.During the training sessions,the animal was head-fixed to the training rig.A droplet of water was formed at a spout by automatically controlled pumping（pumping duration,20 ms）at 100 ms after the end of the sound stimulus（in total,50+100=150 ms from the stimulus onset）.Water droplets remained at the spout after being delivered so that the animal could always obtain water if ever it voluntarily made a licking action at any time after water was delivered.If the animal had not licked before the next trial occurred,a new droplet would replace the previous one at the spout.The spout was positioned at a distance of approximately 3–4 mm from the animal mouth（and with no visible ambient light）such that the animal had to voluntarily stretch out its tongue to probe and acquire water droplets on the spout.Licking actions were monitored by a camera（frame rate 30 Hz）under NIR illumination that was invisible to the animals.There was no cue,stimulus or rewarding/punishing object.Beyond the sound and the water.We did not apply any punishment for incorrect licking timing.One training session contained 100 sound stimulation events.These stimuli were delivered in 25 discrete engagement time windows,each consisting of four events with random inter-trial intervals（in the range of 5–10 s,longer than the duration of a licking action）.The rationale for using a randomized inter-trial interval setting was to avoid the possible effect of the rhythmic predictive responses that have been known to exist in mice.Each engagement time window was followed by a 60 s pause with neither sound nor water being present.For animals in MIM group,the MIR irradiation was turned on（by a sound-free electronic shutter）during the engagement time window（which was 20–35 s depending on the random inter-trial interval,similar to the MIM applications in the two-photon imaging experiments）.A successful sound-evoked licking event was defined as an event in which the animal initiated a licking action within 500 ms from sound stimulus onset,otherwise the event was defined as a miss.For a session,the success rate was defined as the number of successful events divided by the total number of events.For analyzing the licking response latency,the behavior monitoring videos were first inspected by humans and the video frame of the licking action onset was marked for each trial.The response latency was then defined as the duration from the sound stimulation onset to the first video frame in which the animal stretched out its tongue.Results:1.MIM evokes the expression of c-Fos of neurons.We first removed the skull above the visual cortex and then used the mid-infrared laser introduced above to stimulate the cortical cells for 20 s.After c-Fos immunohistochemistry,it was found that some cells in the stimulated region were activated.The activated region was similar to a bullet head,about 400μm in horizontal and vertical distance,basically located in the 2/3 of the cortex.We carried out co-labeling of neuron marker Neu N again,and the results showed that there were about 89%co-labeling cells,indicating that the majority of activated cells were neurons.To test the irradiation time dependence,we tested the irradiation time for5 s,10 s,20 s and 60 s respectively,and then performed c-Fos immunohistochemical staining.We found that the longer the irradiation time,the more neurons were activated.In the previous experiment,the skull of the mice in the irradiation area was removed.Next,we tried to thin the skull to 50μm,and c-Fos immunohistochemical staining was performed after mid-infrared modulation for 20 s.It was found that some neurons were still activated,but the activation efficiency was reduced compared to the 20 s direct irradiation on the cerebral cortex.Mid-infrared modulation can activate neurons directly without the need for additional protein mediation,which has the advantage of being non-invasive compared to optogenetics,making it easier to operate,and has the potential to be applied to more advanced animals and even humans.2.MIM increases the firing rate of neurons in vivo.We used cell-attachment electrophysiological recording to record the action potential of neurons when they were irradiated by mid-infrared light.It was found that the firing rate of neurons increased in the time window of mid-infrared light,but normal physiological characteristics of neurons were not affected,such as the full width at half maxima of action potential.After irradiation,the neurons recovered the firing rate with before,indicating that the activation effect of mid-infrared irradiation is reversible.Next,we used in vivo two-photon imaging to study the temporal and spatial distribution of cortical neurons irradiated in mid-infrared.Firstly,Cal-520 AM,a calcium indicator,was used to label neurons in layer 2/3 of the primary auditory cortex,and the changes in calcium activity of neurons were observed by MIM during a two-photon calcium imaging experiment.We found that the calcium activity of some neurons increased significantly in the time window of MIM.After the irradiation,the calcium activity of neurons returned to the level before the irradiation,which also verified the previous electrophysiological results and confirmed the reversibility of mid-infrared irradiation.After data analysis,it was found that the proportion of neurons activated by mid-infrared modulation was about 10%,and these neurons were scattered in the cerebral cortex,without a clear spatial location relationship.3.The mechanism of activation by MIM in vivo.MIM can cause an increase in cortical temperature,but the increase is not high,only 0.6℃on average,which is not enough to cause changes in neuron excitability.Therefore,the increase in temperature cannot explain the activation mechanism of neurons,and the specific activation mechanism needs further experiments.4.MIM accelerates learning during an associative training task.According to the aforementioned sensory-motor behavioral paradigm,mice take about2-3 days to learn the task,and then we design the two experimental groups,removed skull and thinned skull group.In the process of training,the experimental group will be given the mid-infrared modulation,the control group did not give any stimulus.Through analysis we found that,the learning speed of mice in the two experimental groups was significantly accelerated compared to the control group,confirming that the learning speed of animals can be improved by mid-infrared modulation.There was no significant difference between the two experimental groups.Conclusions:1.MIM activates cortical neurons in vivo.2.MIM activates～10%of cortical neuronal population in vivo.3.MIM activates cortical neurons do not due to the temperature elevation.4.MIM accelerates learning during an associative training task. |
PDF Full Text Request