Activation of mechanoluminescent nanotransducers by focused ultrasound enables light delivery to deep-seated tissue in vivo

Light is used extensively in biological and medical research for optogenetic neuromodulation, fluorescence imaging, photoactivatable gene editing and light-based therapies. The major challenge to the in vivo implementation of light-based methods in deep-seated structures of the brain or of internal organs is the limited penetration of photons in biological tissue. The presence of light scattering and absorption has resulted in the development of invasive techniques such as the implantation of optical fibers, the insertion of endoscopes and the surgical removal of overlying tissues to overcome light attenuation and deliver it deep into the body. However, these procedures are highly invasive and make it difficult to reposition and adjust the illuminated area in each animal. Here, we detail a noninvasive approach to deliver light (termed ‘deLight’) in deep tissue via systemically injected mechanoluminescent nanotransducers that can be gated by using focused ultrasound. This approach achieves localized light emission with sub-millimeter resolution and millisecond response times in any vascularized organ of living mice without requiring invasive implantation of light-emitting devices. For example, deLight enables optogenetic neuromodulation in live mice without a craniotomy or brain implants. deLight provides a generalized method for applications that require a light source in deep tissues in vivo, such as deep-brain fluorescence imaging and photoactivatable genome editing. The implementation of the entire protocol for an in vivo application takes ~1–2 weeks. This protocol describes the synthesis and characterization of mechanoluminescent nanotransducers covering the entire visible spectrum, their characterization in response to ultrasound in tissue-mimicking phantoms and in artificial circulatory systems and their use in mouse behavioral and immunohistochemical assays. Alternative deep-tissue light-delivery approaches include implanted optical fibers or micro light-emitting diodes, intracranially injected upconversion nanoparticles and wavefront-shaping methods based on spatial light modulators. This protocol describes the synthesis and characterization of mechanoluminescent nanotransducers covering the entire visible spectrum, their characterization in response to ultrasound in tissue-mimicking phantoms and in artificial circulatory systems and their use in mouse behavioral and immunohistochemical assays. Alternative deep-tissue light-delivery approaches include implanted optical fibers or micro light-emitting diodes, intracranially injected upconversion nanoparticles and wavefront-shaping methods based on spatial light modulators. This protocol describes the use of noninvasively triggered light in deep tissue via focused ultrasound-activated and systemically injected mechanoluminescent nanotransducers to achieve localized emissions with submillimeter resolution and millisecond response times.


Introduction
Many forms of energy, such as light, electric fields, magnetic fields, and ultrasound waves, have been used to interact with biological systems [1][2][3][4][5][6] .Among them, light has widespread applications in bidirectional communication with living matter, such as imaging, neuromodulation and phototherapies [7][8][9] .However, one common challenge for these applications in vivo arises from the difficulty of efficiently delivering photons deep inside the body, as a result of severe attenuation of light by the highly scattering biological tissues 9 .Conventional approaches to address this challenge usually involve invasive implantation of external devices, such as optical fibers for optogenetic neuromodulation 10 and microendoscopes for neural activity imaging in the deep brain 11,12 .These implants induce acute tissue damage and chronic immune responses around the implantation site, which substantially alter the local physiological environment.As a result, current approaches for in vivo light delivery face the tradeoff between penetration depth and invasiveness, hampering their utility and capability in both research and clinical settings.
To address this challenge, several innovative methods for in vivo light delivery have been demonstrated.For example, implantable micro light-emitting diodes (LEDs) powered by radio frequency waves have been used as wireless light sources for deep-tissue optogenetic neuromodulation in both the central and peripheral nervous systems of freely moving animals [13][14][15] .This method replaces the physical connection between the animal and the external light source with free-space radio frequency waves, thus effectively freeing the subject from tethering.However, the requirement for chronic implants still inevitably induces tissue damage.In addition, upconversion nanoparticles (UCNPs), which convert tissue-penetrating near-infrared (NIR) light into visible light, have also been used as nanoscopic light sources for deep-brain optogenetics 16 .Although this method eliminates chronic tissue implants, the delivery of UCNPs into the brain still requires invasive brain surgeries.Furthermore, although NIR light is less scattered by biological tissues than visible light, its tissue penetration depth is still limited to a few millimeters before a noticeable thermal effect is generated in the superficial tissue 17 .Therefore, although systemic delivery may represent a less invasive route for delivering UCNPs, their effectiveness is still constrained by the limited penetration depth of NIR light and the inability to confine the stimulation area precisely.Furthermore, another method to overcome photon scattering inside the tissue is through wavefront shaping.Specifically, a spatial light modulator (SLM) was used to decode the wavefront change of an ultrasound guidestar and encode the reverse wavefront onto the incident light field, thus allowing tight focusing of 532-nm light to a spot of <30 μm at 2-mm depth inside acute brain slices 18 .The wavefront shaping method for deep-tissue light delivery does not require any implants and is thus noninvasive.However, the penetration depth of this method is limited only to a few millimeters by the efficiency of the ultrasound guidestar.
Despite these recent advances in deep-tissue light-delivery methods, the tradeoff between tissue penetration depth and invasiveness remains largely unresolved.To address this challenge, we developed an ultrasound-mediated, noninvasive, deep-tissue light source, named 'deLight', based on circulation-delivered mechanoluminescent nanotransducers (MLNTs) [19][20][21][22][23][24] .Specifically, MLNTs are luminescent colloidal materials with engineered defects.The bandgap of MLNTs allows them to absorb UV light, which excites the electron in the valence band to the conduction band (process 1; Fig. 1a).The excited electron is then trapped by the defects, where the photoexcitation energy is stored without emission (process 2; Fig. 1a).Upon mechanical perturbation (e.g., ultrasound), the trapped electron is detrapped, returning to the conduction band (process 3; Fig. 1a).Then, this electron undergoes a radiative relaxation process to transition to the valence band with concomitant light emission (process 4; Fig. 1a).Furthermore, these MLNTs can be systemically delivered into the circulatory system in vivo, generating localized light emission upon noninvasive stimulation of tissue-penetrant focused ultrasound (FUS; Fig. 1b).Comprehensive studies examining the biodistribution, excretion and toxicology of MLNTs administered systemically in mice suggest that MLNTs exhibit excellent biocompatibility and minimal adverse effects 22,23 .We demonstrated that deLight can Protocol noninvasively activate channelrhodopsin-2 (ChR2)-expressing neurons in the mouse brain, producing significant changes in the behavior of the mouse and immunostained c-fos activity via a specific neurotechnology termed 'sono-optogenetics'.Here, we provide a comprehensive protocol for deLight, including the synthesis of the MLNTs with emission wavelengths covering the entire visible spectrum, optical characterizations of MLNTs in polydimethylsiloxane (PDMS) phantoms and artificial circulatory systems under FUS and a specific demonstration of deLight for noninvasive optogenetic neuromodulation.

Development of the protocol
The deLight method takes advantage of tissue-penetrant FUS to gate light emission in deep tissue from circulating MLNTs, which can store the photoexcitation energy inside their crystal lattice and release it in the form of light upon localized mechanical stimuli 19,20 .We recently developed a biomineral-inspired suppressed dissolution approach to produce multicolor water-dispersible MLNTs from their micron-sized solid-state precursors [22][23][24] .The produced MLNTs form stable suspensions in aqueous solutions and can be repeatedly 'charged' and 'discharged' with continuous photoexcitation and FUS pulses, respectively.
To enable reproducible light emission upon FUS stimulation, deLight uses the endogenous circulatory system to implement the recharge and discharge cycles after the systemic delivery of MLNTs.First, excitation light can be applied to superficial vessels near the skin to recharge circulating MLNTs therein.Second, charged MLNTs migrate to deep tissue via the continuous in vivo vascular network, emit light locally at the focus of ultrasound and thus become discharged.Third, discharged MLNTs keep circulating inside the body and become recharged again when they return to superficial vessels under continuous photoexcitation.Therefore, the endogenous circulatory system is effectively turned into an 'optical flow battery', enabling deep-tissue light emission at any location or depth on demand without any invasive implants or surgical removal of overlying tissues.We have validated the feasibility of deLight both in an ex vivo artificial circulatory system and in an in vivo mouse model.In addition, we have also demonstrated the in vivo utility of deLight in a special application of noninvasive neuromodulation in the live mouse brain.

Applications of the protocol
The deLight method provides a versatile approach to noninvasively produce a light source deep inside biological tissue, thus enabling any applications that require light in vivo.Specifically, a representative application of this protocol, as demonstrated in our recent works [19][20][21][22] , is noninvasive optogenetic neuromodulation in live rodent brains.Using deLight to modulate opsin-expressing neurons in vivo enables the dissection of complex neural circuitry and potential treatment of neurological disorders, such as Parkinson's and Alzheimer's diseases.Apart from optogenetics, deLight can also be used to provide an excitation light source for deep-tissue fluorescence imaging, such as deep-brain calcium and voltage imaging through the intact scalp and skull 23 .Furthermore, deLight offers a noninvasive method to control other light-gated systems in vivo, such as photoactivatable CRISPR-Cas9 systems for genome modification and photosensitizers for cancer therapy in deep tissues 25,26 .We envision that the wide adoption of deLight will transform the field of biophotonics by significantly expanding the utilities of any biotechniques that require light in vivo.

Comparison with other methods
Compared with other deep-tissue light-delivery methods, such as implanted light sources based on optical fibers or micro LEDs, intracranially injected light sources based on UCNPs and wavefront-shaping methods based on SLMs, the advantages of deLight are at least threefold.First, because of the deep penetration of ultrasound in biological tissue, deLight can produce light emission inside the tissue at a depth >1 cm without the surgical removal of any overlying tissues (e.g., craniotomy) or invasive implantation of light sources.In contrast, both implanted and injected light sources inevitably cause tissue damage and immune responses around the surgical and implantation sites 27,28 , while the penetration depths of UCNP-and SLM-based methods are usually restricted to a few millimeters 16,18 .Second, the light emission spot of deLight can be easily relocated throughout the body by simply scanning the focal spot of the ultrasound in all three dimensions.The ability to relocate the illuminated region is particularly challenging for implanted or injected light sources, which exhibit fixed illumination volumes near the site of implantation or injection.Third, deLight can access organs that are generally refractory to fiber implantation or invasive surgeries, such as the lung, heart and gastrointestinal tract, because of the absence of a physical implant in vivo to overcome structural and functional constraints 29,30 .

Expertise needed to implement the protocol
The synthesis of the MLNTs requires facilities such as a fume hood and special equipment such as a tube furnace, a centrifuge and a ball-mill machine.The implementation of these synthesis procedures requires general wet laboratory training.The following procedures, such as optical and spectral characterizations of the MLNTs under FUS, require an optical table and a complete FUS system.Furthermore, in vivo mouse optogenetic studies require specific skills including retro-orbital injection, tail vein injection, transcardial perfusion and immunohistochemistry staining.For researchers unfamiliar with these animal procedures, we recommend practicing by following online tutorials and attending hands-on workshops before attempting the deLight application.An animal husbandry facility is required for relevant optogenetic experiments in live rodents.

Limitations
Although deLight provides a new platform for deep-tissue light delivery in live animals, the relatively short lifetime (half-life up to 30 min) of systemically delivered MLNTs represents a potential constraint for in vivo applications at a longer timescale.To mitigate this challenge, additional administrations of MLNTs are required if a consistent light source with duration longer than 30 min is needed.Furthermore, at the current stage, the need for an ultrasound transducer to deliver ultrasound stimuli in vivo requires head fixation of animals.However, we envision that the engineering of a wearable ultrasound transducer will enable deLight in freely moving animals 31 .Moreover, because the application of ultrasound may cause nonspecific activation of the peripheral auditory pathway [32][33][34] , experiments must be carefully designed to rule out potential confounding effects besides FUS-mediated light emission.

Protocol
Specifically, when deLight is used for sono-optogenetic neuromodulation, it is recommended that the rectangular envelope of the applied ultrasound waves be smoothed out to remove any undesirable auditory responses in the mouse brain 35 .In addition, when deLight is used for sono-optogenetics, FUS alone may activate the widely expressed mechanoreceptors inside the brain, thus masking the effect of light emission in the specific activation of opsin-expressing neurons.Therefore, restricting the ultrasound intensity below the threshold intensities that have been reported to stimulate neurons by FUS alone at certain frequencies 36 will ensure the desirable effect of neuromodulation by deLight.Lastly, although the current deLight protocol disallows concurrent fluorescent imaging with FUS stimulation, simultaneous calcium or voltage imaging of deLight-stimulated brain regions can be achieved with advanced engineering in both the FUS transducer and the imaging system 37 .

Experimental design
In this protocol, we include detailed procedures for producing MLNTs with their emission wavelengths covering the entire visible spectrum (Procedure 1).In addition, we also include detailed procedures for characterizing the ultrasound responses of MLNTs in tissue-mimicking phantoms (Procedure 2) and artificial circulatory systems (Procedure 3).Procedures 2 and 3 are required for examining the optical and spectral properties of the as-synthesized MLNTs for the implementation of deLight in vivo.Furthermore, we include a representative demonstration of deLight in two assays: a mouse behavioral assay and an immunohistochemical assay of c-fos (Procedure 4).Researchers can choose from these two assays on the basis of their specific experimental design.

Subjects
This protocol primarily describes the use of deLight to activate ChR2-expressing neurons in B6.Cg-Tg(Thy1-COP4/EYFP)18Gfng/J (Thy1-ChR2-YFP) mice of both sexes (6-12 weeks) 38 .In addition, researchers can choose from the four MLNTs introduced in this protocol for optogenetic neuromodulation with other wavelengths of visible light, such as activation of halorhodopsin for neural inhibition and red-shifted channelrhodopsin for neural activation.

Controls
Proper controls are crucial in optogenetic studies.When using deLight for noninvasive optogenetics, we recommend the following control groups wherever the procedure permits.The first control group, wild-type C57Bl/6J mice, which are recommended by the Jackson Laboratory as the proper control on the same genetic background for Thy1-ChR2-YFP mice, should be systemically delivered with MLNTs and stimulated with the same FUS protocol in the same brain region.The second control group is Thy1-ChR2-YFP mice systemically injected with 1× PBS carrier (sham injection) and stimulated with the same FUS protocol.This control can help rule out the possibility of nonspecific and direct neuromodulation with FUS via the activation of the auditory pathway or endogenous temperature-sensitive and mechanosensitive ion channels.This control becomes more crucial when using deLight on target organs and tissues outside the brain, considering the numerous mechanoreceptors (e.g., Piezo1 and TRPA1) found in these tissues 39,40 .The third control group, Thy1-ChR2-YFP mice systemically delivered with MLNTs but without FUS stimulation, should be used as another control to help account for potential confounding effects caused by circulating MLNTs.The fourth control group, an internal control, can be achieved by analyzing neural activity in the non-FUS-stimulated hemisphere of the same Thy1-ChR2-YFP mouse brain.

Regulatory approvals
All procedures involving living animals must be approved by the Institutional Animal Care and Use Committee and comply with national guidelines regarding the use of animals in research.All procedures performed on mice here were approved by Stanford University's Administrative Panel on Laboratory Animal Care.

Reagent setup
Copper stock solution Prepare a solution of 0.01 M copper acetylacetonate in chloroform.It may be stored at room temperature (25 °C) indefinitely.

Sodium citrate buffer
Prepare a solution of 0.08 M sodium citrate in Milli-Q water.It may be stored at room temperature indefinitely.

Anesthetic drugs
Prepare the anesthetic drug (16 mg/ml ketamine and 0.2 mg/ml Dexdomitor) by mixing 0.8 ml of stock ketamine solution, 2 ml of stock Dexdomitor solution and 2.2 ml of 1× PBS.Store at room temperature for ≤1 month.

70% (vol/vol) ethanol
Dilute anhydrous ethanol in Milli-Q water.Store at room temperature for ≤1 month.

Antisedan drug
Prepare the antisedan drug (0.5 mg/ml) by diluting the stock antisedan drug (5 mg/ml) 10 times in 1× PBS.Store at room temperature for ≤1 month.

PFA solution (4% (wt/vol))
Prepare the PFA solution (4% (wt/vol)) by diluting the stock PFA solution (16% (wt/vol)) with 1× PBS.PFA solution should be prepared fresh before use.▲ CAuTIoN PFA is toxic if inhaled or swallowed or if it comes in contact with the skin, causing severe skin and eye damage.It may cause cancer.Handle with care under a fume hood and wear appropriate personal protective equipment (e.g., gloves, mask and laboratory coat).

M phosphate buffer (PB)
For making 200 ml of PB solution, mix 81 ml of Na 2 HPO 4 stock solution with 19 ml of NaH 2 PO 4 stock solution.Then, dilute to a total volume of 200 ml by using Milli-Q water.

Sucrose solution
Prepare a solution of 30% (wt/vol) sucrose in 0.1 M PB solution.Store at room temperature for ≤1 month.

PBS Triton-X100
Prepare a solution of 0.3% (vol/vol) Triton-X100 in 1× PBS.It may be stored at room temperature indefinitely.

Blocking buffer
Prepare a solution of 5% (vol/vol) normal donkey serum in 1× PBS Triton-X100 solution.Blocking buffer should be prepared fresh before use.

Washing solution
Prepare a solution of 0.05% (vol/vol) Triton-X100 in 1× PBS.It may be stored at room temperature indefinitely.

Primary antibody solution
Dilute the primary antibodies in the blocking solution with an appropriate ratio (Table 1).The primary antibody solution should be prepared fresh before use.

Secondary antibody solution
Dilute the secondary antibodies with an appropriate ratio (Table 1) in 1× PBS with 5% (vol/vol) normal donkey serum and 0.1% (vol/vol) Triton-X100.The secondary antibody solution should be prepared fresh before use.▲ CAuTIoN Secondary antibodies are light sensitive.Cover with aluminum foil after dilution.

Ball-mill machine
Seek expert guidance from the facility manager regarding the use of this equipment.General guidelines for setting up the ball-mill machine can be found at https://www.spexsampleprep.com/knowledge-base/resources/manuals/8000d%20mixermill%20manual%20100714%20abridged.pdf.The duration of a single session of the ball-milling process here is 10 min.

Camera
The operation of the camera in this protocol requires the installation of the software ThorCam.General information regarding the software ThorCam can be found at https://www.thorlabs.com/software_pages/ViewSoftwarePage.cfm?Code=ThorCam.

FuS system
The operation of the FUS system is customized by Image Guided Therapeutics.FUS transducers with center frequencies of 0.65-3.5 MHz have been demonstrated to produce intense emission from the MLNTs described in this protocol.The representative transducer used in the protocol is a 1.5-MHz transducer with a diameter of 38.2 mm, a weight of 40 g and a lateral focal spot size of 1.0 mm.The maximum power supplied to the transducer used in this protocol is 50 W.In addition, the actual input power can be determined as the product of the drive amplitude (in percentage) and the maximum input power (50 W).The operation of the FUS system requires the installation of the software BBBop.Contact the vendor for further information (https:// www.igtradiology.com/services/ultrasound/).

ocean optics spectrophotometer
The operation of the ocean optics spectrophotometer in this protocol requires the installation of the software OceanView.General information regarding the software OceanView can be found at https://www.oceaninsight.com/globalassets/catalog-blocks-and-images/manuals--instruction-ocean-optics/software/oceanviewio.pdf.

Perfusion pump
Select the syringe volume of the perfusion syringes used in the procedure and set the flow rate to 5 ml/min.

Cryostat
General information regarding the operation of the cryostat can be found at https://www.leicabiosystems.com/us/histology-equipment/cryostats/leica-cm3050-s/.In this protocol, the Protocol temperature of the operation chamber was set to −20 °C, the blade temperature was set to −18 °C and the sectioning thickness was set to 40 μm.

Microscope
Seek expert guidance from your local imaging facility to set up the microscope.General guidelines for the ZEISS LSM 980 with Airyscan 2 can be found at https://www.zeiss.com/microscopy/en/products/light-microscopes/confocal-microscopes/lsm-980-with-airyscan-2.html.

LabVIEW
LabVIEW software is required for acquiring the optical power measurements of the mechanoluminescence emission from the artificial circulatory system.General information regarding LabVIEW can be found at https://www.ni.com/en-us/support/downloads/ software-products/download.labview.html#477380.

MATLAB
For synchronizing the control of the FUS transducer, the UV-LED and the camera, the installation of the software MATLAB is required in this protocol.General information on MATLAB can be found at https://www.mathworks.com/products/matlab.html.

ImageJ
For analyzing the line profile of the mechanoluminescence emission and the confocal imaging of the stained brain slices, installation of the software ImageJ is used in this protocol.General information on ImageJ can be found at https://imagej.net/software/fiji/.

Synthesis of bulk mechanoluminescent materials via solid-state reactions
• TIMING 2 d ▲ CRITICAL The detailed reagent compositions for producing each bulk mechanoluminescent material are listed in Table 2. 1. Prepare all reagents needed for synthesis.2. Weigh all reagents by using a balance and transfer all reagents to an agate mortar.3. Grind the reagents by using an agate pestle for 1 h (Fig. 2a).Transfer the ground reagents into an alumina crucible (Fig. 2b) and place the crucible on an alumina combustion boat.▲ CRITICAL STEP Do not use porcelain crucibles and combustion boats, because both can be used only under 1,050 °C.By contrast, alumina crucibles and combustion boats can withstand temperatures up to 1,750 °C. 5. Carefully place the combustion boat into the tube furnace (Fig. 2c) with a hook (Fig. 2d) and make sure that the combustion boat is right in the heating zone of the tube furnace.Tightly seal the end, which has the tubing connected to the gas cylinder.Connect the other end of the furnace with another piece of tubing and immerse the open end of the tubing in a beaker filled with water inside the fume hood.6. (Optional) Turn on the gas cylinder containing 5% H 2 in Ar and make sure that the gas flow is stable.▲ CRITICAL STEP Only Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al and ZnS:Mn require 5% H 2 in Ar during synthesis.CaTiO 3 :Pr needs to be synthesized in air.When gas flow is needed, the open end of the tubing is placed inside a water-containing beaker for monitoring the gas flow rate via the production of gas bubbles.7. Run the pre-set program of the furnace.After the program is complete, let the furnace cool down to room temperature.▲ CRITICAL STEP The detailed temperature settings for producing each bulk mechanoluminescent material are listed in Table 2. 8. (Optional) Turn off the gas when the program is finished.9.After the furnace is cooled down to room temperature, carefully take out the combustion boat from the furnace with the hook.▲ CAuTIoN Beware hot surfaces!Make sure that the furnace has cooled down to room temperature.10.Transfer the as-synthesized bulk mechanoluminescent materials out of the crucible into the agate mortar and roughly grind the materials with a pestle for 1 min.■ PAuSE PoINT The materials can be stored in air at room temperature for >1 year.13.Load a tungsten carbide grinding vial with the materials from Step 11 and one 7/16-inch (11.2-mm) tungsten carbide ball (Fig. 4a).14.Seal the vial with a screw-on tungsten carbide-lined cap and place it in a clamp of the ball-mill equipment (Fig. 4b).15.Tighten the clamp with the knob and lock the clamp with the locking tab.Close the lid of the ball-mill machine and fasten it with the manual latch.▲ CRITICAL STEP When mounting the grinding vials in the ball-mill equipment, note that two clamps must be balanced with a load.If only one sample is loaded, place an empty vial in the other clamp.16.Start the ball-milling process with three 10-min sessions and a 5-min interval between the sessions.17.Take the vial out of the clamp and collect the ball-milled material.

Table 2 | Synthesis conditions and peak wavelengths of mechanoluminescent materials
▲ CRITICAL STEP Usually, there will be a residue of ball-milled materials stuck to the grinding vial and the grinding ball.To remove this residue and clean the grinding setup, pour a small amount of sand into the grinding vial and start a ball-milling process for 10 min.
■ PAuSE PoINT The ball-milled materials can be stored at room temperature for >1 year.

Synthesis of MLNTs via a biomineral-inspired suppressed dissolution approach
• TIMING 3 d 18. Load the flask with 2 g of the material in Step 17 and 240 ml of the sodium citrate buffer (0.08 mol/l).19.Set the oil bath to 80 o C. 20.Immerse the flask in the oil bath and connect a condenser to the flask (Fig. 4c).
▲ CAuTIoN Beware hot surfaces!Handle with care.▲ CRITICAL STEP Make sure that the level of the oil is higher than that of the solution inside the flask.21.Turn on the magnetic stirrer and start a timer for 72 h for the suppressed dissolution process.
■ PAuSE PoINT The as-etched materials can be stored at room temperature for >1 year.

Surface modification of biocompatible MLNTs
• TIMING 2 d

22.
After the suppressed dissolution process, transfer all materials from their original container (i.e., the reaction flask or storage container) to centrifuge tubes.23.Start the centrifugation process at 1,000 rpm (116 relative centrifugal force (rcf)) for 10 min and collect the supernatant containing colloidal MLNTs (Fig. 4d).24.Transfer the colloidal solution into a beaker and add sodium chloride (3 g/10 ml) to the solution to facilitate the collection of the colloid at a relatively low centrifugation speed for its purification.25.Stir the colloidal solution with a glass rod to facilitate the dissolution process of sodium chloride and wait for 30 min (Fig. 4e).

Transfer the solution from
Step 25 into centrifuge tubes and start the centrifugation process at 5,000 rpm (2,907 rcf) for 10 min.▲ CRITICAL STEP Avoid using high centrifugation speed over 5,000 rpm (2,907 rcf), which may cause irreversible agglomeration of MLNTs during surface modification.27.Remove the supernatant from all centrifuge tubes.28.Disperse the pellets in Milli-Q water to a total volume of 5 ml.29.Transfer the colloidal solution from Step 28 into a clamped cellulose dialysis tubing with a molecular weight cutoff of 300 kDa.30.Hang the dialysis tubing on a glass rod via a rubber band and place the glass rod on top of a 2-liter glass beaker containing Milli-Q water (Fig. 4f).31.Turn on the magnetic stirrer and dialyze for 48 h to remove excess ions, such as citrate, Na + , Cl − and etched ions, from the colloidal solution.The presence of these ions may interfere with the following measurements for determining the concentration of MLNTs.▲ CRITICAL STEP Change the bath with fresh Milli-Q water with a total of six water changes during dialysis.32.Collect the colloidal solution from the dialysis bag in a glass vial.sonicator for 1 h of sonication at room temperature (Fig. 4g).39.Remove the glass vial from the clamp and transfer the solution into a centrifuge tube.
Wash the colloids with Milli-Q water three times via a centrifugation-resuspension process.The centrifuge is set to 4,000 rpm (1,860 rcf) for 10 min, and the supernatant is removed after each wash before resuspending the pellet in 20 ml of Milli-Q water.40.Wash the colloid with anhydrous DMF three times via a centrifugation-resuspension process.The centrifuge is set to 4,000 rpm (1,860 rcf) for 10 min, and the supernatant is removed after each wash before resuspending the pellet in 20 ml of DMF.41.Disperse the colloid from the last wash into 10 ml of anhydrous DMF with sonication.46.Wash the colloids with anhydrous DMF three times via a centrifugation-resuspension process.The centrifuge is set to 4,000 rpm (1,860 rcf) for 10 min, and the supernatant solution is removed after each wash before resuspending the pellet in 20 ml of DMF.47. Wash the colloid with Milli-Q water three times via a centrifugation-resuspension process.The centrifuge is set to 4,000 rpm (1,860 rcf) for 10 min, and the supernatant solution is removed after each wash before resuspending the pellet in 20 ml of Milli-Q water.48.Disperse the PEGylated MLNTs from the previous step in 1 ml of Milli-Q water and transfer it to a glass vial.▲ CRITICAL STEP Examine the color of each colloidal solution of MLNTs under room light and its afterglow in the dark.Each MLNT colloid should exhibit the afterglow of its corresponding color as shown in Fig. 4h-k  4. Mix all the materials until the MLNTs are uniformly distributed in the mixture.5. Place the glass Petri dish in a vacuum desiccator and vacuum for 30 min to remove air bubbles.6. Remove the glass Petri dish from the vacuum desiccator and place it in a 70 °C oven for 30 min.▲ CAuTIoN The glass Petri dish will be hot.Handle with care.7.After the PDMS phantom is cured, remove the Petri dish from the oven and separate the phantom from the glass Petri dish.
■ PAuSE PoINT The phantom may be stored at room temperature indefinitely.◆ TRouBLESHooTING

(Optional) Calibration of FUS transducers with a needle hydrophone
• TIMING 1 d ▲ CRITICAL The calibration of an FUS transducer, which entails the derivation of the relationship between the drive amplitude and ultrasound peak pressure, is required for every new FUS transducer before use.Figure 5a shows a schematic illustration of the calibration setup described in Steps 8-12.
8. Connect the FUS system to a computer with a USB-to-USB-Type-B cable.9. Connect the needle hydrophone to the DC coupler.10.Connect the DC coupler to an oscilloscope with a BNC-to-BNC cable and a 50-Ω impedance adapter.▲ CRITICAL STEP If the oscilloscope has an input impedance of 50 Ω, then the 50-Ω impedance adapter is not needed in this step.11.Mount the FUS transducer on a stereotaxic manipulator.

Protocol
12. Immerse the FUS transducer and the needle hydrophone into a water-filled 2.5-liter glass tank (Fig. 5b).Align the transducer plane perpendicular to the axis of the needle hydrophone (Fig. 5c).▲ CRITICAL STEP Aligning the transducer and the needle hydrophone is crucial for the calibration.If the axis of the needle hydrophone is not perpendicular to the transducer plane, the calibration results will not be accurate.pressure map by moving the transducer in desirable dimensions.Specifically, first move the transducer from its starting position in the y direction within the range of ±5 mm with an interval of 0.2 mm and repeat Step 20 with a constant FUS amplitude (30%) at each step.▲ CRITICAL The starting position (x 0 = 0, y 0 = 0, z 0 = 0) of the FUS transducer is determined in Step 16. 24.Next, move the transducer back to its starting position before moving it in the x direction within the range of ±5 mm with an interval of 0.2 mm.Repeat Step 23 for each x position.25.Finally, to evaluate the pressure with respect to the distance between the tip of the needle hydrophone and the transducer (i.e., in the z direction), first move the transducer back to its starting position, then move the transducer axially to the needle hydrophone within the range of ±10 mm with an interval of 0.2 mm and repeat Step 20 with a constant FUS amplitude (30%) at each step.26.Move the transducer back to its starting position before moving it in the x direction within the range of ±5 mm with an interval of 0.2 mm.Repeat Step 25 for each x position.27.Plot the 2D pressure map in the x-y plane (at z 0 = 0) on the basis of the measurements in Steps 23 and 24.Plot the 2D pressure map in the x-z plane (at y 0 = 0) on the basis of the measurements in Steps 25 and 26.28.Save the spatial mapping plots of the FUS transducer for future reference.Representative pressure mapping results of a 1.5-MHz FUS transducer in the x-y plane (at z 0 = 0) and in the x-z plane (at y 0 = 0) can be found in Extended Data Fig. 3a,b, respectively.

Spectral characterizations of ultrasound-mediated light emission in the phantom
• TIMING 2 h ▲ CRITICAL Figure 5d shows a schematic illustration of the spectral characterization setup described in Steps 29-33.29.Fill the transducer housing with ultrasound gel until the gel level surpasses the height of the housing.Place the phantom prepared from Step 7 on the ultrasound gel.▲ CRITICAL STEP Make sure that there are no air bubbles in the ultrasound gel between the FUS transducer and the phantom.30.Couple one end of the optical fiber to the spectrometer and the other end with a collimating lens.31.Place a vertical post near the transducer and place a post clamp on it.32.Mount the collimating lens to a horizontal post with a post clamp and connect the horizontal post to a vertical post.Make sure that the bottom of the collimating lens is close to the PDMS phantom.▲ CRITICAL STEP The joint between the horizontal post and the vertical post should be loose to facilitate the rotation of the horizontal post.33.Place a 'stop' post to confine the rotation angle of the horizontal post such that the center of the transducer and that of the collimating lens are vertically aligned when the two posts are in contact.A complete experimental setup can be found in Fig. 5e, where we define the current position of the lens-coupled fiber as Position 1 (Fig. 5f).34.Open the OceanView software and set the software for time-series spectrum acquisition.35.In the View mode, click the 'Configure graph saving' button and select 'Time Series (column data)' as the file format.36.In the Acquisition Group Window, check the 'Burst Mode' and set the integration time to 100 ms and the number of 'Spectrum back to back' to 50-75, which corresponds to a total acquisition time of 5-7.5 s.

Protocol
37. Connect the FUS system to the computer with a USB-to-USB-Type-B cable for FUS generation.38.Turn on the FUS hardware and open the BBBop software.Load the FUS setting as follows: amplitude, 30%; pulse duration, 200 ms; and delay, 800 ms.39.Move the lens-coupled fiber away from the center of the transducer to Position 2 (Fig. 5g).40.Verify that the charging light is aimed at the PDMS phantom.Adjust the position and the angle of the UV-LED lamp to ensure uniform light illumination.41.Measure the power of the UV-LED lamp with an optical power meter and adjust the controller to set the incident power density on the PDMS phantom to 0.5 mW/mm 2 .42. Turn off the room light and start charging with the UV-LED lamp for 10 s.
▲ CRITICAL STEP It is important to move the fiber to Position 2 during charging to avoid saturation or damage to the spectrometer during charging.43.After charging, immediately move the lens-coupled fiber over the PDMS phantom (Position 1) and start time-series spectrum acquisition.44.Start FUS pulses.The time-series spectra will be saved automatically.45.Compute the mechanoluminescence spectrum by subtracting the baseline spectrum, which corresponds to the spectrum taken immediately before the FUS pulse, from the peak spectrum, which corresponds to the spectrum taken during maximum emission under an FUS pulse.Normalized mechanoluminescence spectra of MLNTs of Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al, ZnS:Mn and CaTiO 3 :Pr can be found in Extended Data Fig. 4.

◆ TRouBLESHooTING ■ PAuSE
PoINT Researchers may pause here and continue the following steps later.

Spatial characterizations of ultrasound-mediated light emission in the phantom
• TIMING 1 h ▲ CRITICAL Figure 5h shows a schematic illustration of the spatial characterizations setup described in Steps 46-49.46.Place the phantom on the FUS transducer as described in Step 29.47.Mount a camera above the phantom.Make sure that the camera is horizontal with a spirit level (Fig. 5i).48.Connect the camera to the computer by using a USB-to-USB-B micro superspeed cable for image acquisition.49.Open the ThorCam software and adjust the relative position between the phantom and the camera to make sure that a clear phantom image forms in the center of the camera's field of view.50.Set the imaging acquisition parameters of the camera as follows: exposure time, 50 ms; gain, 0; binning factor, 4. 51.Check whether the camera blocks the charging light of the UV-LED lamp.Adjust the position and the angle of the UV-LED lamp if necessary.52.Set up the FUS system as described in Steps 37 and 38.53.Turn off the room light and start the UV-LED lamp charging for 10 s. 54.After charging, immediately start time-series image acquisition and FUS pulses.▲ CRITICAL STEP When acquiring the images, make sure that no pixels are saturated on the camera.In addition, to obtain the optimal imaging result, it is important to start FUS pulses immediately after charging, because most of the stored energy will gradually be released via a thermal process.55.Save the time-series images as a multipage tiff file.56.Obtain the line profile of the emission spot by using the ImageJ software.57.Fit the emission line profile with a Gaussian function.The spatial resolution of FUS-mediated light emission is defined as the full-width-at-half-maximum of this fitted Gaussian function.An example of acquiring the emission line profile and Gaussian fitting can be found in Extended Data Fig. 5. ■ PAuSE PoINT Researchers may pause here and continue the following steps whenever needed.

Rechargeability measurements of ultrasound-mediated light emission in the phantom
• TIMING 1 h 58. Place the phantom on the FUS transducer as described in Step 29.59.Set up the camera over the phantom as described in Steps 47-49.60.Set up the FUS system as described in Steps 37 and 38.61.Connect the analog output ports of the multifunction I/O device to the UV-LED lamp controller and FUS hardware ('Trigger-in' for both) via BNC-to-BNC cables.▲ CRITICAL STEP Port 0 should be connected to the UV-LED lamp, and Port 1 should be connected to the FUS system, consistent with the MATLAB code.62. Open the BBBop software, switch 'all measurements' to 'no measurement' and set the FUS system to 'trigger pulse'.63.Run a test trial by setting the cycle number in the MATLAB code to 1 and run the code.
A single time-series tiff file will be saved automatically.A link to a representative tiff file can be found in Data availability.▲ CRITICAL STEP In the custom-written MATLAB code, each cycle consists of 5-s charging and 7.5-s image acquisition (49.966-ms exposure time, 20-Hz frame rate, 150 frames in total) with a 2-s interval in between them.FUS pulses start at 2 s after the start of image acquisition.The interval between consecutive cycles is 5.5 s. ◆ TRouBLESHooTING 64.Verify that emission can be clearly seen and that no pixels are saturated in the image.65.Change the cycle number in the MATLAB code to 50 and run the code.Representative rechargeability curves of MLNTs composed of Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al, ZnS:Mn and CaTiO 3 :Pr over 20 cycles can be found in Extended Data Fig. 6. 66. (Optional) Because this procedure takes ~30 min, researchers may leave the room, leaving the rechargeability test operating automatically.Make sure that the room is kept in the dark during the entire experiment.■ PAuSE PoINT Researchers may pause here and continue the following steps later.

Procedure 3: characterizations of MLNTs in an artificial circulatory system
Construction of the artificial circulatory system • TIMING 30 min 1. (Figure 6a shows a schematic illustration of the artificial circulatory system.)Fill the transducer housing with ultrasound gel and 3D-print a holder that can be tightly mounted on the transducer housing while leaving two openings on the opposite sides of the circular frame.An STL file of the holder can be found in Data availability.2. Insert the tubing, which has an i.d. of 1.59 mm, an o.d. of 3.18 mm and a total length of 35 cm, through the 3D-printed holder.Mount the holder on top of the transducer (Fig. 6b) and make sure that there is no air between the transducer and the tubing.3. Assemble the tubing to the peristaltic pump and set the flow rate of the pump to 11.3 ml/min.4. Prepare 1.5 ml of an MLNT colloidal solution at a concentration of 40 mg/ml in a 2-ml glass vial.Add a stir bar in the vial. 5. Place the glass vial on a magnetic stirrer and secure it with a flask clamp.6. Insert the two ends of the tubing into the glass vial and make sure that the openings of the tubing are fully immersed in the colloidal solution (Fig. 6c).

Temporal dynamics of ultrasound-mediated light emission in the artificial circulatory system
• TIMING 2 h 7. Connect the UV-LED with a DC power supply.8. Mount the UV-LED on the optical table and adjust the controller to set the power density to 3.8 mW/mm 2 .

Protocol
9. Place the tubing on top of the LED chip.Secure the tubing in place with tape (Fig. 6d).10.Mount a camera above the 3D-printed holder with tubing inside and make sure that the camera is horizontal with a spirit level (Fig. 6e).11.Connect the camera to the computer via a USB-to-USB-B micro superspeed cable for image acquisition.12. Open the ThorCam software and adjust the relative position between the tubing and the camera to make sure that a clear tubing image forms in the center of the camera's field of view.13.Set the image-acquisition parameters of the camera: exposure time, 5 ms; gain, 0; binning factor, 4. ▲ CRITICAL STEP To investigate the temporal dynamics of the ultrasound-mediated light emission in the artificial circulatory system, change the FUS pulse width from 10 to 50 ms.The amplitude (30%) and the repetition rate (1 Hz) remain constant.In our experiments, we find that varying the FUS pulse width from 10 and 20 to 50 ms is sufficient to study the temporal dynamics of ultrasound-mediated light emission in the artificial circulatory system.17.Turn on the magnetic stirrer and the peristaltic pump.Make sure that the flow of MLNT colloidal solution is continuous inside the tubing.18. Seal the light leakage from the UV-LED and the illuminated segment of the tubing with black aluminum foil (Fig. 6f).19.Turn off the room light, turn on the UV-LED and make sure that the camera does not capture any stray light.20.Start image acquisition and FUS pulses.
▲ CRITICAL STEP The tubing filled with the MLNT colloidal solution under FUS stimuli should exhibit bright emission at the focus of ultrasound as shown in Fig. 6g-j.21.After the first session of FUS, adjust the FUS pulse width accordingly.22.For each experimental condition, repeat the measurements several times and average the onset and offset times of FUS-mediated light emission, which correspond to the time delay of light emission after the FUS pulse is applied and removed, respectively.Expected temporal kinetics of FUS-mediated light emission from mechanoluminescent fluids composed of Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al, ZnS:Mn and CaTiO 3 :Pr can be found in Extended Data Fig. 7. ▲ CRITICAL STEP The onset time is defined as the time between the start of FUS pulse (t 1 ) and the time (t 2 ) when the light intensity (I) has increased to three times the standard deviation of I before t 1 .The offset time is defined as the time between the end of the FUS pulse (t 3 ) and the time (t 4 ) when I has dropped below three times the standard deviation of I at infinite time.23.After the test, turn off the magnetic stirrer, the peristaltic pump and the UV-LED and turn on the room light.

Absolute emission intensity measurement of the ultrasound-mediated light source in the artificial circulatory system
• TIMING 1 h 24.Determine the radius of the emission spot, r, in the artificial circulatory system by repeating Procedure 2, Steps 56 and 57.25.Replace the camera with a photodiode that is connected to the photodiode amplifier.26.To calculate the tubing's transmittance T, perform the following steps.Set up a stable light source, such as an LED, that matches the wavelength of the tested MLNTs and position the photodiode in front of this light source to measure the light intensity I 1 .Next, slice the tubing lengthwise and flatten it so it can cover the photodiode's sensing area.Measure the light intensity of the light source, I 2 , with the tubing covering the photodiode.Finally, calculate the tubing's transmittance T at this specific wavelength by using the formula T = I 2 I 1 .27. Connect the output from the photodiode amplifier to an analog input port (AI 0) of the multifunction I/O device with a BNC-to-BNC cable.28.Load the FUS settings and set the pulse width to 200 ms, amplitude to 30% and the repetition rate to 1 Hz.29.Open the LabVIEW software and open the file 'PD.vi'.Set the optical power acquisition parameter as follows: sampling frequency, 1,000; binning factor, 10. 30.Repeat Steps 17 and 18 to set up the artificial circulatory system.31.Turn off the room light, turn on the UV-LED and make sure that the readings from the photodiode remain unchanged after the UV-LED is turned on.This step is to ensure that the photodiode does not capture any light leakage from the UV-LED.

Protocol
32. Start the optical power measurement of light emission from the artificial circulatory system and start FUS pulses.33.Save the power measurements in Excel files.34.Calculate the optical power density (I 0 ) according to TAr 2 P m, where d is the distance between the emission spot and the sensor of the photodiode, T is the transmittance of the tubing at the peak wavelength of the tested MLNTs (Table 2), A is the sensing area of the photodiode, r is the radius of the emission spot in the artificial circulatory system and P m is the measured power of light emission by the photodiode.For reference, the calculated I 0 of the mechanoluminescent fluid composed of Sr 2 MgSi 2 O 7 :Eu,Dy at a concentration of 25 mg/ml is 0.7 mW/cm 2 with a variation of 6.8% between repeated measurements.

Pressure-to-luminance transfer functions of ultrasound-mediated light emission in the artificial circulatory system
• TIMING 2 h 35.Keep the magnetic stirrer, the peristaltic pump and the UV-LED on.Keep the room light off.36.Set the FUS pulse width as 200 ms and the repetition rate as 1 Hz.37. Set the drive amplitude of the FUS as 5%, which corresponds to a certain peak pressure obtained from the calibration plot from Procedure 2, Step 22. 38.Start image acquisition and apply FUS pulses.39.Gradually increase the FUS amplitude and repeat Step 37 for each drive amplitude.40.Determine the threshold pressure for the MLNTs to produce light emission by plotting the measured emission intensity from each captured image versus its corresponding FUS pressure.Expected dependence of the emission intensity from mechanoluminescent fluids composed of Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al, ZnS:Mn and CaTiO 3 :Pr on FUS pressure can be found in Extended Data Fig. 8.

Procedure 4: using deLight for in vivo sono-optogenetic neuromodulation
▲ CRITICAL MLNTs, which exhibit strong emission under FUS and consistent emission under repeated charging and discharging cycles, must be prepared before any in vivo applications of deLight.Researchers can choose either tail vein injection (option 1) or retro-orbital injection (option 2) for systemic delivery of the MLNT solution (Fig. 7a).

Using deLight for in vivo optogenetic neuromodulation
• TIMING 1-5 h 1. Prepare and sterilize the procedure area and surgical tools (Fig. 7b).▲ CAuTIoN All procedures involving live animals must be approved by the Institutional Animal Care and Use Committee.2. (Optional) Mount a video-recording device for monitoring mouse limb movements.Add a lens tube to the UV-LED lamp for better spatial confinement of the recharging area (Fig. 7c).▲ CRITICAL STEP This step is required only for the behavioral assay.3. Anesthetize a mouse by intraperitoneally injecting the anesthetic mixture.The anesthesia level of the mouse should be checked by assessing the paw withdrawal reflex in response to a toe pinch.(A) Behavioral assay (i) Administer the anesthetic drug at a dose of 1 μl/g.The animal should show the paw withdrawal reflex in response to a toe pinch.(B) c-fos expression evaluation (i) Administer the anesthetic drug at a dose of 5 μl/g and a maximum dosage of 150 μl regardless of the animal's weight.The animal should show no paw withdrawal reflex in response to a toe pinch.▲ CRITICAL STEP Follow option A if a behavioral assay will be performed.Follow option B if brain c-fos expression levels will be evaluated.Protocol 5. Apply a small amount of eye lubricant to each eye to prevent drying during this procedure.6. (Optional) Mark the joints of both hindlimbs with four different colors (Fig. 7d).▲ CRITICAL STEP This is required only for the behavioral assay.Try to mark those regions as symmetrically as possible.7. Place the animal on a heating pad set to 37 °C to prevent hypothermia.▲ CRITICAL STEP If a behavioral assay is performed, skip Step 8 and proceed to Step 9.
If researchers intend to evaluate c-fos expression, proceed to Step 8. 8. (Optional) Wait for 2 h before proceeding to the next step.During the wait, apply additional eye lubricant as needed.▲ CRITICAL STEP This step is crucial for reducing and even eliminating the baseline c-fos expression in the targeted brain region.In our experience, 2 h is sufficient for producing a clean c-fos background in the absence of neural stimulation.Researchers can adjust the time as needed.9. Mount the mouse head to the head holder by placing the mouse's front teeth into the inner hole of the head holder and tightening the placement with the screw.10.Place the ear bars into the mouse's ear canal and tighten the ear bars with screws in place.
▲ CRITICAL STEP This step is essential for setting the right coordinates for later procedures.11.Adjust the position and the angle of the UV-LED lamp to make sure part of the mouse's back is illuminated according to the following instructions.▲ CRITICAL STEP If a behavioral assay is performed, adjust the relative position of the UV-LED lamp and the video-recording device to make sure that the lower body of the animal (i.e., the area that is captured by the video-recording device) is not illuminated.The illumination area in the behavioral assay should be the upper back of the animal, thus justifying the removal of fur on the entire back in Step 4. If only post-mortem immunostaining of brain sections will be performed, the shaved lower back of the animal should be illuminated.12. Measure the power of the UV-LED lamp with an optical power meter and adjust the controller to set its output power to 1 mW/mm 2 .13. Connect the FUS system to the computer with a USB-to-USB-Type-B cable for FUS stimulation.14.Turn on the FUS hardware, open the BBBop software and connect the stereotaxic manipulator with the mounted FUS transducer to control the movement of the transducer.15.Lift the scalp with forceps and make an incision with scissors.16.Clean the exposed skull with 1× PBS and mark the bregma with a sharpie (Fig. 7e).The incision of the scalp and exposure of the skull are intended for generating deLight only at precise stereotaxic coordinates with respect to the bregma in the skull.FUS can penetrate through the intact mouse scalp and skull with minimal attenuation, thus not requiring their removal for producing light emission in the brain.17.Use a spirit level to make sure that the transducer is horizontal (Fig. 7f).18. Place a small piece of Sr 2 MgSi 2 O 7 :Eu,Dy-doped PDMS phantom (refer to Procedure 2) on the FUS transducer housing, which is filled with ultrasound gel, and charge it with a UV flashlight.19.Turn off the room light and turn on FUS.Mark the bright point (Fig. 7g) in the phantom with a Sharpie as the focus of the FUS transducer (Fig. 7h,i).◆ TRouBLESHooTING 20.Place an L-shaped aligner consisting of one post and one needle to align the lateral position of focus of the transducer with one arm (i.e., attached needle) of the aligner (Fig. 7j).21.Rotate the transducer together with the aligner from an upward-facing position to a downward-facing one.22. Align the marked bregma (from Step 16) of the mouse with the needle tip (Fig. 7k) by moving the transducer laterally until the needle tip aligns with the mouse's bregma.This approach ensures that the FUS transducer's focus shares the AP and ML coordinates with the bregma.▲ CRITICAL STEP Make sure that after the rotation the transducer stays horizontal with a spirit level.This step aligns the focus of the transducer with the bregma of the mouse brain, and in this way, the alignment achieves its highest possible accuracy.

Protocol
23. Remove the aligner and the phantom.24.Apply extra ultrasound gel to the transducer and the animal's exposed skull (Fig. 7l).25.Move the transducer with the stereotaxic manipulator to the targeted coordinates with respect to the bregma in both x and y directions.▲ CRITICAL STEP The reference coordinates can be found at Mouse Brain Atlas (http://labs.gaidi.ca/mouse-brain-atlas/)and Reference Atlas (https://mouse.brain-map.org/static/atlas).Researchers should choose the stereotaxic coordinates according to their specific studies.The mouse skull exhibits a negligible effect on shifting the focus of ultrasound (within the frequency range of 1.5-6 MHz) 41 , thus enabling the use of deLight without significant off-target effect in the mouse brain without adjusting stimulation coordinates.26.Slowly lower the transducer by adjusting its height until the 'ring' of the transducer housing is ( f − t − DV) mm away from the exposed skull, where f corresponds to the the focal length of the FUS transducer, t the thickness of the housing and DV the dorsal-ventral coordinate of the target brain region (Fig. 7m).A representative side view of the relative positions of the transducer and the mouse head is shown in Fig. 7n.▲ CAuTIoN Avoid causing damage to the animal when lowering the transducer.▲ CRITICAL STEP When lowering the transducer, researchers should use a caliper to measure the distance between the edge of the transducer housing (i.e., the 'ring') and the surface of the skull to match the desired value of ( f − t − DV) mm.27.(Optional) Lift the transducer by 10 cm via the stereotaxic manipulator of the FUS system.
▲ CRITICAL STEP This is required only if the retro-orbital injection is performed in Step 30.28.Prepare 200 μl of the MLNT colloidal solution in 1× PBS (30 mg/ml) and load the solution into a 29-gauge insulin syringe.29.(Optional) Adjust the position of the video-recording device to make sure that the lower body of the mouse with the markers can be clearly seen before starting video recording.▲ CRITICAL STEP This is required only for the behavioral assay.30.Charge the MLNT solution with a UV flashlight and perform systemic administration.
Researchers can either perform tail vein injection (Fig. 7o) or retro-orbital injection (Fig. 7p).31.After the injection, immediately turn on the UV-LED lamp for recharging on the animal's skin.32.(Optional) Lower the transducer by 10 cm to reach the preset brain coordinates for the targeted region with the stereotaxic manipulator (Fig. 7q).▲ CRITICAL STEP This is required only if retro-orbital injection is performed in Step 30.▲ CRITICAL STEP This is required only for the behavioral assay.36.Remove the mouse head from the head holder and close the wound with Vetbond (Fig. 7r).
▲  ▲ CRITICAL STEP Avoid causing damage to inner organs, especially the heart.41.Insert a 27-gauge needle into the left ventricle and make a small incision on the right atrium.
Perfuse the animal with 15 ml of 1× PBS through the inserted needle at a rate of 5 ml/min.42.After 1× PBS perfusion, perfuse the animal with 25 ml of 4% (wt/vol) PFA.43.Collect the brain from the skull and post-fix it in 4% (wt/vol) PFA at 4 °C for 24 h.44.Remove the brain from PFA and rinse it with 1× PBS.Place the brain in the brain slicer (Fig. 8a).Use razor blades to collect coronal sections with ~3-mm thickness around the region of interest (Fig. 8b).▲ CRITICAL STEP Make sure that the ventral side of the brain faces upward.This orientation ensures that the cutting plane is roughly parallel to the standard brain atlas.Before the cut, reference the brain atlas and determine the locations of the cut.Increase the brain section thickness to avoid missing the region of interest if necessary.45.After collecting the desired brain section, cut a small piece of brain tissue from the bottom of the non-FUS-stimulated cerebral hemisphere.▲ CRITICAL STEP This step marks the side of the mouse brain not illuminated with deLight to facilitate immunostaining and imaging of c-fos later.46.Put the brain section of interest in 10 ml of 0.1 M PB solution containing 30% (wt/vol) sucrose at 4 °C until the tissue block sinks to the bottom.▲ CRITICAL STEP The brain tissue will float in the sucrose solution when first immersed (Fig. 8c), before sinking to the bottom overnight (Fig. 8d).If the tissue section is thicker Protocol than 3 mm, a longer time is needed before the tissue block sinks to the bottom.Researchers may replace the sucrose solution with freshly made solutions to expedite the diffusion process.
Frozen embedding and sectioning of brain tissue

Troubleshooting
Troubleshooting advice can be found in Table 3.The PDMS phantom is misaligned from the ultrasound focus, or the light emission spot is misaligned from the lens-coupled fiber Make sure that the focus of the FUS transducer is positioned within the PDMS phantom according to Extended Data Fig. 3 Make sure that there are no air bubbles in the ultrasound gel between the transducer and the PDMS phantom Adjust the position of the lens-coupled fiber to make sure that it is vertically aligned with the ultrasound focus for efficient light collection

Anticipated results
This protocol provides a detailed description of the implementation of a noninvasive deep-tissue light-delivery method, termed 'deLight'.After the completion of this protocol, researchers will be able to synthesize MLNTs with emissions spanning the entire visible spectrum under FUS, characterize the properties of MLNTs and deLight in tissue-mimicking phantoms and artificial circulatory systems and apply deLight for any applications that need light in deep tissue in live animals.Specifically, successfully produced MLNTs should have bright emissions and good colloidal stability when dispersed in an aqueous solution.Before applying deLight in vivo, researchers can use this protocol to characterize the emission spectra, spatial resolution and temporal dynamics of synthesized MLNTs in tissue-mimicking phantoms and in an artificial circulatory system.Furthermore, in vivo implementation of deLight takes advantage of the endogenous circulatory system that allows these MLNTs to be recharged in the superficial vessels near the skin and discharged to produce local light emission at the ultrasound focus.As a representative example, this protocol describes a procedure for applying deLight for in vivo sono-optogenetic neuromodulation in live mice, followed by a behavioral assay (Fig. 9a) and an immunohistochemical assay (Fig. 9b) to validate efficacious light emission.Besides the demonstrated in vivo optogenetic application of deLight described in Procedure 4, we envision that this unique approach for systemic light delivery can facilitate minimally invasive optogenetic neuromodulation in deeper brain regions in larger animals and enable any biological applications that require light in the deep body.emission can be found at https://doi.org/10.6084/m9.figshare.23691312.Further data are available from the corresponding author upon request.Characterized and quality-controlled MLNTs are available to other research laboratories upon request.

Fig. 1 |
Fig. 1 | Principles of deLight and its application in vivo.a, Mechanism of light emission from charged MLNTs upon FUS stimulation.b, A schematic illustration of the charging (photoexcitation) and discharging (photoemission) processes of MLNTs during circulation.FUS, focused ultrasound.Panel b reproduced from ref. 20 under the Creative Commons license CC BY 4.0.

Na 2 NaH 2
HPo 4 stock solution Prepare a solution of 28.4 g/l Na 2 HPO 4 in Milli-Q water.It may be stored at room temperature indefinitely.Po 4 stock solution Prepare a solution of 24.2 g/l NaH 2 PO 4 in Milli-Q water.It may be stored at room temperature indefinitely.

Fig. 2 |
Fig. 2 | Synthesis of bulk mechanoluminescent materials via solid-state reactions.a, Grind the reagents with an agate mortar and a pestle.b, Transfer the ground reagents into an alumina crucible.c, The tube furnace used for solid-state reactions.d, Place the combustion boat in the tube furnace with a hook.

Fig. 4 |
Fig. 4 | Synthesis process of biocompatible MLNTs from bulk mechanoluminescent materials.a,b, Ball-milling process of the bulk mechanoluminescent material: a tungsten carbide grinding vial loaded with the as-synthesized bulk mechanoluminescent material and a tungsten carbide grinding ball (a); the ball-mill machine used in this protocol (b).c, The setup used for the biomineral-inspired suppressed dissolution approach.d, Colloidal 33. Acquire TEM images of colloidal MLNTs after dialysis to make sure that the diameters of MLNTs are in the range of 10-110 nm.Representative TEM images of colloidal MLNTs composed of Sr 2 MgSi 2 O 7 :Eu,Dy, ZnS:Cu,Al, ZnS:Mn and CaTiO 3 :Pr can be found in Extended Data Fig. 1. 34.Dilute the solution prepared in Step 32 into three different concentrations and measure their corresponding UV-visible spectra over a wavelength range from 400 to 700 nm with a UV-visible spectrophotometer.▲ CRITICAL STEP We recommend making 100×, 200× and 300× dilutions of the solution from Step 32.Make sure that the measured absorbance values are between 0.1 and 1 for an optical path length of 1 cm in the entire spectral range of measurements.35.Measure the mass concentrations of the solutions tested in Step 34 via inductively coupled plasma mass spectrometry.36.Determine the extinction coefficient Ɛ of the colloidal solution per milligram of the colloid per milliliter via linear fitting of the measured absorbance at 400 nm at three concentrations (from Step 34) versus their mass concentrations determined by inductively coupled plasma mass spectrometry (from Step 35) for each MLNT material.▲ CRITICAL STEP The concentration of each colloid is calculated on the basis of the Beer-Lambert law (A = Ɛlc), in which A is the absorbance, Ɛ is the extinction coefficient per milligram of the colloid per millilter, l is the optical path length (centimeters) and c is the concentration (milligrams per milliliter) of the colloid.37. Transfer 20 ml of the colloidal solution diluted to a concentration of 6 mg/ml from Step 32 into a glass vial and add 40 μl of sodium hydroxide solution (10 M). ◆ TRouBLESHooTING 38.Mount the glass vial with a flask clamp and immerse the glass vial into a water-filled 42. Transfer the suspended solution into a glass vial.43.Add 40 mg of mPEG-silane (20 kDa) to the colloid in DMF.44.Mount the glass vial with a flask clamp and immerse the glass vial into a water-filled sonicator for 4 h of sonication at 50 °C.◆ TRouBLESHooTING Protocol 45.Remove the glass vial from the clamp and transfer the solution into a centrifuge tube.

13 .
Turn on the FUS hardware and open the BBBop software.14.Set the FUS pulse amplitude to 5%, pulse duration to 200 ms and delay to 800 ms. 15.Turn on the DC coupler and the oscilloscope and start FUS pulses.16.Monitor the voltage readout on the oscilloscope and adjust the position of the transducer in three axes by using the stereotaxic manipulator until the voltage of the corresponding FUS pulses reaches the maximum.17.Repeat the measurements three times and record the maximum voltage V. 18. Increase the FUS pulse amplitude to 50% in 5% increments and repeat Step 17 in each setting.19.Find the sensitivity M of the needle hydrophone at the frequency of the FUS transducer from the hydrophone calibration data provided by the manufacturer.20.Calculate the peak pressure P of each measurement with the formula P = V/M.21.Plot the calculated peak pressure as a function of the drive amplitude of FUS and fit the values with linear regression.

Fig. 5 |Protocol 22 .
Fig. 5 | Calibration of the FUS system and characterizations of MLNTs in a tissue-mimicking phantom.a, A schematic illustration of the FUS calibration process.b, Experimental setup of a. c, A zoom-in photograph showing the alignment of the transducer and the needle hydrophone.d, A schematic illustration of the setup for measuring the mechanoluminescence spectrum of the PDMS phantom under FUS.e, Experimental setup of d. f, A top view of

Fig. 6 |Protocol 14 .
Fig. 6 | Characterizations of MLNTs in an artificial circulatory system.a, A schematic illustration of the artificial circulatory system.b, A 3D-printed holder that can be tightly mounted on the transducer where the inserted tubing lies at the focus of the FUS transducer.c, The construction of the artificial circulatory system.d, The UV-LED with a segment of the tubing fixed on top of the LED chip.e, The complete setup for the characterizations of MLNTs in the

Protocol 4 .
Remove the fur of the mouse with hair-removal lotion.▲ CRITICAL STEP If a behavioral assay is performed, remove the fur on the head, entire back, hindlimbs and hindpaws of the mouse.If only post-mortem immunostaining of brain sections will be performed, remove the fur on the head and the lower back of the mouse.

FUSFig. 7 |
Fig. 7 | Application of deLight in vivo.a, A schematic illustration of the application of deLight in vivo.b, The setup for applying deLight in optogenetic neuromodulation followed by c-fos immunostaining.c, The setup for applying deLight in optogenetic neuromodulation with simultaneous behavioral monitoring.d, Photograph showing the marked locations on the mouse's back.e, A head-mounted animal with its skull exposed and part of the fur removed.f, A side view of the FUS transducer.g, PDMS phantom showing bright emission at the focus of applied ultrasound.h, A marked dot on the PDMS phantom indicating the focus of the FUS transducer.i, A side view of the transducer with the PDMS phantom.j, Alignment of the needle of the L-shaped aligner with the marked dot on the PDMS phantom in the z-direction.k, Rotation of the transducer with the aligner and alignment of the needle with the marked bregma of the mouse skull.l, Removal of the aligner and the PDMS phantom and application of extra ultrasound gel.m, Movement of the transducer to the desired brain coordinates and lowering of the transducer.n, A side view of the transducer and the mouse head, showing their relative positions.o, Tail-vein injection of the loaded MLNT solution charged by a UV flashlight.p, Retro-orbital injection of the loaded MLNT solution charged by a UV flashlight.q, After injection of the MLNT solution, start of FUS pulses and UV recharging on the back.r, Glueing the scalp back at the end of the deLight stimulation session.Panel a adapted with permission from ref. 22, ACS.

33 .
Start the first FUS pulse train for 5 min (amplitude 25%, pulse duration 200 ms and delay 800 ms).▲ CRITICAL STEP Because of the circulation half-life of systemically administered MLNTs, starting the first FUS pulse train immediately after the administration of the MLNT fluids is recommended.34.Start a timer of 30 s after the first session is finished and start the second 5-min session (amplitude 25%, pulse duration 200 ms and delay 800 ms) at the end of the 30-s interval.▲ CRITICAL STEP The UV-LED lamp should be turned off during the 30-s interval and turned back on when the second FUS session starts.If the researchers plan to change the duration of the FUS stimulation, ensure that the total duration combining Steps 32-34 is within 30 min after MLNT administration.35.(Optional) Stop video recording.

Protocol 38 .
(Optional) Wait for 90 min for c-fos expression.▲ CRITICAL STEP During the wait, prepare the procedure area and tools for transcardial perfusion.Apply eye lubricant as needed.Perfusion, tissue collection and fixation• TIMING 2 d ▲ CRITICAL The following Steps 39-82 are for c-fos evaluation only.39.Place the animal on the dissection stage inside a fume hood.◆TRouBLESHooTING 40.Perform a midline laparotomy to expose the heart.

Fig. 8 |
Fig. 8 | Dissection and staining of brain sections to evaluate the efficacy of deLight.a, Determination of the region of interest from the ventral side of the brain.b, Collection of tissue blocks with razor blades.c, A brain tissue block floating in the sucrose solution upon immersion.d, A brain tissue block sinking

63 3 23Procedure 4 19Procedure 1 Steps 1 - 2 Steps 1 - 7 , 3 Steps 1 - 6 , 4 Steps 1 -
Anticipated image files are not saved The hardware (i.e., FUS system and UV-LED lamp) are not connected to the correct ports of the I/O device Adjust the connection between the hardware and the I/O device and make sure that the analog output ports of the I/O device are connected as described in the protocol Procedure No emission is observed MLNT colloidal solution is not flowing inside the tubing, or the charging segment of the tubing falls off the LED chip Check the ends of the tubing in the glass vial to ensure a continuous flow of the colloidal solution.Check the tubing on the UV-LED chip to ensure continuous recharging of the colloidal solution No bright spot is observed in the PDMS phantom under FUS There may be air bubbles in the ultrasound gel between the FUS transducer and the PDMS phantom to hinder ultrasound transmission Check the interface between the FUS transducer and the PDMS phantom and remove air bubbles.Add additional ultrasound gel if needed39  The animal responds to toe pinch The animal is gradually waking up from the initial dosage of the anesthetic drug An additional dose of the anesthetic drug may be needed until the animal shows no responses to a toe pinch.In our experience, a supplemental anesthetic (25% of the initial dose) can be given at ~70 min after deLight application 73 The image is too dim The imaging setting is not optimized Adjust the laser intensity, pinhole size and gain to improve the image quality 74 No detectable c-fos expression is evident in the brain sections within the hemisphere where FUS was applied The FUS transducer may have been misaligned with respect to the mouse bregma, thus yielding off-target stimulation in the brain Expand the search of positive c-fos signals in neighboring coronal sections.This expanded search can be done by staining multiple brain sections, extending their range to include AP coordinates ±1 mm from the original AP coordinate.The sections should be taken at intervals of 200 µm, and the process should continue until positive c-fos signals are identified Protocol Timing 17, synthesis of bulk mechanoluminescent materials via solid-state reactions: 2 d Steps 18-21, synthesis of MLNTs via a biomineral-inspired suppressed dissolution approach: 3 d Steps 22-48, surface modification of biocompatible MLNTs: 2 d Procedure preparation of the tissue-mimicking phantom uniformly doped with MLNTs: 1 h Steps 8-28, calibration of FUS transducers with a needle hydrophone: 1 d Steps 29-45, spectral characterizations of ultrasound-mediated light emission in the phantom: 2 h Steps 46-57, spatial characterizations of ultrasound-mediated light emission in the phantom: 1 h Steps 58-66, rechargeability measurements of ultrasound-mediated light emission in the phantom: 1 h Procedure construction of the artificial circulatory system: 30 min Steps 7-23, temporal dynamics of ultrasound-mediated light emission in the artificial circulatory system: 2 h Steps 24-34, absolute emission intensity measurement of the ultrasound-mediated light source in the artificial circulatory system: 1 h Steps 35-40, pressure-to-luminance transfer functions of ultrasound-mediated light emission in the artificial circulatory system: 2 h Procedure 38, using deLight for in vivo optogenetic neuromodulation: 1-5 h Steps 39-46, perfusion, tissue collection and fixation: 2 d Steps 47-55, frozen embedding and sectioning of brain tissue: 2 h Steps 56-74, immunostaining, mounting and imaging: 2-3 d Steps 75-82, evaluating the volume of brain tissue optogenetically activated by deLight on the basis of immunostaining of c-fos: 1 d

Fig. 9 |Extended Data Fig. 4 |
Fig. 9 | Representative results of in vivo sono-optogenetic neuromodulation with deLight.a, Tracking of mouse limb movements produced by optogenetic neuromodulation via deLight in the brain.b, Immunostaining of c-fos in mouse brain sections after neuromodulation with deLight (left), along with the statistical analysis of the density of c-fos + cells (right) for the Thy1-ChR2-YFP

ProtocolExtended Data Fig. 5 |
Figure adapted with permission from ref. 22, ACS.An example mechanoluminescence image (a) and its corresponding emission line profile fitted to a Gaussian function (b) to obtain the full-width-at-half-maximum (FWHM).Protocol Extended Data Fig. 7 | Temporal kinetics of FUS-induced light emission from colloidal solutions of MLNTs.a−d, Temporal kinetics of MLNT solutions composed of Sr 2 MgSi 2 O 7 :Eu,Dy (a), ZnS:Cu,Al (b), ZnS:Mn (c) and CaTiO 3 :Pr (d) upon 10-, 20-and 50-ms FUS pulses in the artificial circulatory system with an imaging frame rate of 200 Hz.e−h, Onset and offset times of mechanoluminescence emission measured from a-d.All data are presented as the mean ± s.d. of 10 independent measurements.ML, mechanoluminescence.Figure reproduced with permission from ref. 22, ACS.

Table 1 | Antibodies Primary antibody Secondary antibody Name Dilution Company name and cat. no. Name Dilution Company name and cat. no.
The time required to obtain permissions varies between months and 1 year, depending on the local committee.Institutional Material Transfer Agreements are also required if other research laboratoriess request MLNTs from Protocol our laboratory.The submission for the Administrative Panel on Laboratory Animal Care's approval and the establishment of Material Transfer Agreements represent straightforward and standard procedures that are inherent to the research process.Importantly, these steps do not necessitate any supplementary financial expenditure.

characterizations of MLNTs in a tissue-mimicking phantom Preparation of the tissue-mimicking phantom uniformly doped with MLNTs
in the dark.When examining the color of the MLNT colloidal solutions, note that the afterglow intensity naturally decays without further charging.Specifically, it takes ~3 min to thermally dissipate all stored energy in MLNTs of Sr 2 MgSi 2 O 7 :Eu,Dy to 1%.All four MLNTs are ranked in terms of their power and energy storage capability as follows: Sr 2 MgSi 2 O 7 :Eu,Dy > ZnS:Cu,Al > ZnS:Mn > CaTiO 3 :Pr.■ PAuSE PoINT The PEGylated MLNTs can be stored at room temperature for ≤1 week.Dry MLNTs prepared from Step 32 in Procedure 1 in air and grind them roughly with an agate mortar and a pestle.2. Prepare 1.54 g of PDMS base, 0.15 g of PDMS curing agent and 112 mg of dried MLNTs.3. Add the PDMS base, PDMS curing agent and MLNTs into a glass Petri dish.
CRITICAL STEP If a behavioral assay is performed, proceed to Step 37.If researchers intend to evaluate c-fos expression, proceed to Step 38.37. (Optional) Inject a dose of an anti-sedative drug and monitor the animal's health condition according to the approved animal protocol.Analyze the recorded limb motion with a customized MATLAB code.▲ CRITICAL STEP This is the last step if the deLight behavioral assay is performed.If researchers intend to evaluate the c-fos expression level, proceed to Step 38.
Prepare the materials and tools needed for tissue sectioning.First, take the brain tissue out from the sucrose solution and wipe off the remaining solution from the surface with Kimwipes.48.With blunt forceps, transfer the brain tissue into a plastic mold prefilled with a thin layer of O.C.T. compound.Add extra O.C.T. compound to fill the mold and put the plastic mold on ice for 30 min.▲CRITICALSTEPToavoiddrying of the O.C.T. compound, it is preferable to place the mold inside a styrofoam container.Also avoid the formation of air bubbles.If there are any air bubbles, remove them with blunt forceps.Be careful not to damage the tissue.49.During the 30-min wait, place a 10-cm Petri dish on top of a foam box (which provides thermal insulation) and prepare a slurry composed of dry ice and methanol in the Petri dish.50.Repeat Step 48 to transfer the brain tissue to a second plastic mold.51.Use long forceps to grasp the mold and place it into the dry ice-methanol slurry.After all the O.C.T. compound has turned white, transfer the mold to dry ice.▲CRITICALSTEPMakesure that the mold floats on the surface of the dry ice-methanol slurry, so that methanol is not in contact with O.C.T. compound or tissue.O.C.T. compound should turn white when it freezes within 1-2 min (Fig.8e).■PAuSEPoINTThetissueblockcanbestored at −80 °C for ≤6 months.52.Transfer the tissue block to the precooled cryostat (Fig.8f).▲CRITICALSTEPDuring the transfer, always keep the tissue block on dry ice.Do not thaw the tissue block at room temperature.53.Remove the tissue block from the plastic mold and glue the tissue block to the sample holder with O.C.T. compound.Wait until the sample equilibrates to the chamber temperature.▲CRITICALSTEP If the tissue block is freshly frozen-embedded, wait for 30 min.If the tissue was stored in the −80 °C freezer, wait for ≥1 h.54.Mount the sample and the blade to the cryostat.Adjust the angle of the tissue block and the blade as needed.▲CAuTIoNPay extra attention to the sharp blades to avoid injuries.55.Cut the tissue block into sections with a thickness of 40 μm and collect the sections of interest (Fig.8g) into a 48-well plate filled with pre-cooled 1× PBS if the sections will be stained within 2 d. ■ PAuSE PoINT Replace the 1× PBS with cryoprotectant for long-term storage.Sections can be stored at −20 °C for ≤6 months.Put the shaker inside a 4 °C refrigerator.Rinse the brain sections in 1× PBS for 10 min on the shaker at a speed of 450 rpm.Repeat three times.▲ CRITICAL STEP After each rinse, transfer the brain sections to a new well filled with fresh solution or discard the old solution with a pipette.This step applies to Steps 58-63.59.Block each brain section in 200 μl of the blocking solution for 1 h at room temperature on a shaker at a speed of 150 rpm.60.Incubate each brain section in 200 μl of the primary antibody solution at 4 °C overnight on a shaker at a speed of 150 rpm.