Delivering DNA Aptamers Across the Blood–Brain Barrier Reveals Heterogeneous Decreased ATP in Different Brain Regions of Alzheimer’s Disease Mouse Models

DNA aptamers have been developed as sensors to detect metabolites with high sensitivity, selectivity, and biocompatibility. While they are effective in sensing important targets in the brain, the lack of methods for their efficient delivery across the blood–brain barrier (BBB) has significantly hindered their applications in brain research. To address this issue, we herein report the development of brain cell-derived exosomes as endogenous BBB delivery vehicles to deliver an ATP-responsive aptamer across the BBB of live mice for noninvasive live brain imaging. We found that the system uses endosome recycling to transfer the sensors between the delivered exosomes and native recycling endosomes, resulting in high delivery efficiencies. Using this system, we observed unique signal distributions for ATP across different brain regions, with significant accumulation in the subiculum and cortex in healthy mice. In an Alzheimer’s disease transgenic mouse model, ATP levels decreased in the subiculum and cortex, demonstrating this method’s capability to determine metabolite location and relative abundance with high spatial resolution in vivo. Since DNA aptamers have been obtained for many other targets, the method developed in this work can be applied to deliver sensors across the BBB to image a wide range of other brain-related metabolites.

PBS buffer.Run times of 1 minute, with the camera level set to 8 at room temperature and screen gain set to 1, were used to collect size and purity data.Data tracking and analysis was automatically performed by the Nanoparticle Tracking Analysis software on the Nanosight.The Center for Electrochemistry used calibration beads (Malvern) to calibrate the instrument before each run.

Sensor hybridization
For the ATP aptamer sensor, 10 µM stocks of annealed sensor was prepared by mixing the aptamer to quencher strand at a 1:1.1 ratio in PBS + 2 mM MgCl2.The strands were annealed following a thermocycler method of heating at 90°C for 5 minutes and then cooled to 25°C over 40 cycles.Unless otherwise noted, all cellular and in vivo experiments used the hybridized sensors, noted as the ATP aptamer sensor.Control strands (scrambled sequence for the ATP aptamer) were hybridized with the same protocol).For these controls, all cellular and in vivo experiments used the hybridized strands and were labeled as inactive control.
Loading efficiency measurements of exosomes with the ATP aptamer Multiple methods for internalizing the sensor within the exosomes were attempted.For all methods, 1.25 µM ATP aptamer modified with a 5′ Rhodamine Green fluorophore was incubated in a 250 µL solution of 1 X 10 6 exosomes (particle count based upon Nanosight measurements) at room temperature for 30 minutes.Any excess volume required was made up with PBS + 2 mM MgCl2.For the incubation loading technique, no further steps were performed.For freeze-thaw, the samples were then flash frozen in liquid nitrogen and allowed to warm in a heated water bath.The freeze-thaw process was performed for a total of 6 cycles.For sonication, the samples were sonicated for 6 cycles at 100% power, 37 kHz of 20 second sonication, with 5 second rest periods between each sonication pulse.For electroporation, the samples were electroporated at 200 mV for 6 cycles, with 5 second rest periods between each pulse.For all methods, the unloaded aptamer was then removed from loaded exosomes with 50 kDa MWCO Amicon ultrafiltration tubes by 3X centrifugation at 4500g for 10 minutes.The removal of unloaded sensor was visually evaluated by the presence of a light yellow solution in the flow-through.Fluorescence signal post loading was determined by incubating the loaded exosomes with 0.1% Triton-100 for 10 minutes before fluorescence emission analysis by fluorimetry (Horiba Fluoromax, excitation: 480 nm, emission: 520 nm, slit widths 5 mm).To generate a standard curve, we prepared a dilution range of 200 µL solutions of 50 nM to 1 µM Rhodamine Green-labeled ATP aptamer in 50 mM Tris, 100 mM NaCl, 5 mM MgCl2 pH 7.4 at 37°C.We measured these fluorescence emissions by a Horiba Fluoromax (excitation: 480, emission: 520, slit widths 5 mm) to generate a standard curve of fluorescence intensity versus known Rhodamine Green-labeled ATP aptamer concentration.To determine the loading efficiencies, we used the standard curve's linear best fit equation with fluorescence signal from the loaded exosome to calculate the concentration of aptamer in each exosome solution.

Cell viability assay
The cytotoxicity of the ATP aptamer sensor loaded exosome systems was evaluated by the standard MTT assay.Cells (bEnd.3 and SY5Y) were grown until 90% confluency in 96 well plates.Stock solutions of exosomes were prepared in cell culture grade water, with gradient dilutions to prepare different concentrations of native and surface modified exosomes from both SY5Y cells and bEnd.3 cells.On the day of the experiment, media was replaced with fresh Opti-MEM.A negative control of only Opti-MEM was used.A positive control was 200 µL of fresh 30% hydrogen peroxide.The serial dilution of exosomes was added to each well, in triplicate, and incubated at 37°C for 4 hours in the cell growth chamber.A solution of MTT (stock 5 mg/mL in PBS) was diluted in Opti-MEM to a final concentration at 0.25 mg/mL and 100 µL was added to each well after aspirating exosome containing solutions.The plates were then incubated for 5 hours at 37°C.The MTT solution was then aspirated and replaced with 200 µL DMSO, and the absorbance was measured at 570 nm using a BioTek plate reader.
Monocultured cell delivery assays Cells (bEnd.3 or SY5Y) were plated in ibidi 8 well plates (for imaging).Exosome solutions were normalized based on BCA assays, in which the total probe-loaded exosome solution was ~5.7 µg total protein per 0.5 mL cell culture media, corresponding to ~200 nM functional-DNA loaded.As a comparison to commercially available transfection agents, the manufacturer's instructions were followed for liposome complex formation with Lipofectamine 3000 and Turbofect and 400 nM DNA.Briefly, for Lipofectamine 3000, two separate tubes were prepared: 1, 25 µL Opti-MEM with 1 µL Lipofectamine 3000 (vortex 3 seconds) and 2, 25 µL Opti-MEM + 200 nM DNA).Tubes 1 and 2 were mixed and incubated at room temperature for 3 minutes.Then 50 µL Opti-MEM was added and the solution was incubated for 12 minutes at room temperature before being added to the cell culture media.For Turbofect, 50 µL of Opti-MEM were mixed with 1 µL Turbofect and 200 nM DNA at room temperature for 15 minutes before being added to the cell culture media.Exosome solutions were allowed to incubate with the cells for 4 hours in the humidified cell culture incubator at 37°C.Before imaging, cells were washed 3X with PBS.200 µL fresh DMEM without phenol red was added to each well.Lasers of 405 nm (collect 410 nm to 480 nm), 488 nm (collect 490 to 550 nm), and 561 nm (collect 570 to 630 nm) were used.The pinhole was set to 1 Airy unit.All microscope settings were kept constant.All images shown use the same minimum and maximum settings.
Exosome mediated delivery in the Transwell BBB model Transwell membrane inserts (24 mm, 0.4 µm pores, polyester TC-treated, Corning) were prewarmed with 1 mL bEnd.3 cell culture media (DMEM), in which the apical side was prewarmed with 2 mL SY5Y cell culture media (EMEM/Ham's F12).bEnd.3 cells (0.3 X 10 6 ) were seeded in the apical side of the insert and SY5Y (0.3 X 10 6 ) were seeded in the basolateral side.The cells were cultured at 37°C in 5% CO2 for 10 days, or until 100% confluency for the bEnd.3cell layer.On the day of the delivery experiment, the media was replaced with fresh Opti-MEM and Rhodamine Green labeled ATP aptamer sensor loaded exosome solution (5.7 µg total protein per 0.5 mL cell culture media, corresponding to ~200 nM sensor loaded.)was added to each the apical side.Native and surface modified SY5Y and bEnd.3 exosomes were used.For imaging, after 4 hours of incubation in a humidified cell culture incubator at 37°C, 75 nM of Lysotracker Red was added to each well and allowed to incubate for 1 hour.After 1 additional hour of total incubation in a humidified cell culture incubator at 37°C, the cells were washed 3X with PBS and replaced with 1 mL fresh DMEM without phenol red and 1 µg/mL Hoescht 33342.Positive controls include adding the probe loaded exosome solution to the basolateral neuroblastoma cell media directly and Lipofectamine 3000-sensor to the basolateral side directly.For Transwell transportation efficiency assays measured over time, the exosomes were allowed to incubate over 5 hours.Each Transwell BBB integrity was verified with Lucifer Yellow penetration.A stock solution of 60 µM Lucifer Yellow in Hanks Balanced Salt Solution with 1% DMSO was prepared.Lucifer Yellow was added to the bEnd.3cell layer at a final concentration of 0.1 mg/mL.At each time point, 150 µL of the cell culture media from the SY5Y layer was transferred to a clear bottom 96 well plate (Corning).The plate was read in a BioTek plate reader with excitation: 430 nm, emission 535 nm.Controls of Hank's Balanced Salt Solution buffer and 0.1 mg/mL Lucifer Yellow were used.The percent permeability was calculated as (sampleblank)/(lucifer yellow-blank) * 100.Lucifer Yellow assays repeatedly showed less than 2% permeabilization across the brain endothelial cells, which is in the acceptable window of endothelial cell leakiness.All imaging studies were performed with live cells, except for the colocalization studies with the recycling endosome pathway.Lasers of 405 nm (collect 410 nm to 480 nm), 488 nm (collect 490 to 550 nm), and 561 nm (collect 570 to 630 nm) were used.The pinhole was set to 1 Airy unit.All microscope settings were kept constant.Images could only be collected with the 10X objective, due to the working distance limitation and large Transwell depth.All images shown use the same minimum and maximum settings.

Recycling endosome pathway modulation
In the Transwell BBB model, both bEnd.3 and SY5Y cells were treated with 0.1 μM Bafilomycin A1 for 30 min in DMEM, 1 μg/mL U18666A for 30 minutes in DMEM, or in DMEM alone for 30 minutes in a cell culture incubator at 37°C.After this initial incubation, sensors were added to the Transwell, as described in exosome-mediated delivery across the Transwell BBB model.Deliveries across the Transwell were performed for 2 hours, 5 hours, and 8 hours in a cell culture incubator at 37°C.Subsequently, the cell fixation for antibody staining along the recycling endosome pathway protocol was followed.
Cell fixation for antibody staining along the recycling endosome pathway After the exosome mediated delivery in the Transwell BBB model protocol, both the bEnd.3 and SY5Y cells were washed 1X with PBS.Cells were then fixed for 10 minutes with 4% paraformaldehyde and washed 1X with PBS.Both cell layers were permeabilized with 0.2% Triton X-100 for 10 minutes and washed 1X with PBS.Cells were then blocked with 1% BSA in PBS at room temperature for 45 minutes.Cells were then washed 1X with PBS and cells were incubated with 1:100 dilution of Rab11a antibody in PBS (Invitrogen Catalog #71-5300) overnight at 4°C, washed 3X with PBS, and then stained with a 1:1000 dilution of secondary donkey anti-rabbit Alexa Fluor 647 in PBS (Invitrogen Catalog #A-31573) in 1% BSA in PBS blocking solution for 45 minutes.Before imaging, cells were washed 1X with PBS, with fresh PBS added as the imaging solution.Lasers of 405 nm (collect 410 nm to 480 nm), 488 nm (collect 490 to 550 nm), 561 nm (collect 570 to 620 nm), and 633 nm (collect 640 to 660 nm) were used.The pinhole was set to 1 Airy unit.All microscope settings were kept constant.All images shown use the same minimum and maximum settings.

Transportation efficiency study
To study the transportation efficiency of the exosomes across the BBB cell model, we used the Transwell BBB model as described above.At each time point (5 min, 15 min, 30 min, 60 min, 120 min, 180 min, and 240 min), 5 µL of the basolateral side was aliquoted in 195 µL Opti-Mem.The fluorescence signal from the media samples was analyzed with a BioTek plate reader.Opti-Mem was used as the blank, and a negative control of Lipofectamine 3000 + 200 nM Rhodamine Green labeled ATP aptamer sensor added directly to the basolateral side were used.Because the fluorescence intensity is proportional to the functional-DNA sensor, the transportation efficiency of each method was determined as (Ftime -Fmedium)/(Ftotal-Fmedium), where Ftime is the signal at different time points, Fmedium is the Opti-MEM blank, and Ftotal is the negative control of the DNA directly added to the basolateral side.This was performed 3 times.

Delivery of sensors in live mice
Animal care was approved by and followed the guidelines of the University of Texas at Austin's Institutional Animal Care and Committee (IACUC).For in vivo experiments, 150 µg of exosomes (determined by BCA) were loaded with 25 µmol of Cy5-ATP aptamer sensor/kg of mice body weight (~14 g), with unloaded probe purified from the exosomes with 50 kDa Amicon ultrafiltration tubes the day before the experiment and stored at -20°C overnight.Negative controls of PBS, unloaded inactive sensor, and unloaded active sensor were used.Positive controls of PEG-Liposome In Vivo Transfection Reagent (Altogen Biosystems, were prepared according to manufacturer's protocol) and exosomes with exposed Cy5 ATP aptamer (no hybridized quencher) were used.All samples with lipid carriers (exosomes and liposome control) were stained with PHK26 membrane dye (Sigma-Aldrich) as an internal control for delivery efficiency normalization. 3 month old BALB/c mice (Jackson Laboratory) were used as the healthy control, while 3 month old 5xFAD mice (National Institute of Health, Mutant Mouse Resource and Research Centers, purchased through Jackson Laboratory) were used as the Alzheimer's Disease transgenic model mice.Mice were anesthetized using isoflurane (2%) during injections and in vivo imaging.Tail vein injections were performed with a total volume of 200 µL per mouse using a sterile 22 gauge needle.After injections, mice were allowed to recover in their cage, with access to food and water ad libitum during the experiment timecourse.No adverse side effects for the mice were noted after injections.Mice were sacrificed with CO2 as the primary method of euthanasia and cervical dislocation as the secondary method of euthanasia.

In vivo imaging
In vivo imaging was performed with a Xenogen IVIS system.Settings used were 640 nm excitation, 680 nm emission, f number 4, binning factor 8, and 6.6 field of view.To first visualize the delivery system's pharmacokinetics, mice were injected with either PBS or with probe loaded exosomes.The isofluorane-sedated mice were imaged over a 48-hour period at 15 minutes, 1 hour, 2 hours, 4 hours, 6 hours, 12 hours, 24 hours, and 48 hours post-injection.Probe signal in the brain region saturated at 6 hours post injection, with a clear decrease in blood circulation over the next two days.All images shown use the same minimum and maximum settings.

Post-mortem tissue analysis for biodistribution
To study the biodistribution of the exosome systems, the main organs of the mice (brain, heart, liver, spleen, lung, and kidney) were excised at 6 hours post-injection.Immediately after excision, the organs were washed with 3X with PBS to remove blood.The organs were imaged using the Xenogen IVIS Spectrum fluorescent imager, with the mean fluorescence signals analyzed using the instrument software.The same settings were used as the in vivo imaging (640 nm excitation, 680 nm emission, f number 4, binning factor 8, and 6.6 field of view).
Post-mortem tissue analysis of brain sections Six hours post-injection, the mice were sacrificed (primary method: CO2 and secondary method: cervical dislocation) and the brains excised.The brains were fixed overnight in 4% PFA at 4°C and dehydrated with 30% sucrose in PBS at 4°C for 2 days.After embedding in O.C.T. (Scigen Tissue-PLUS OCT Compound, Fisher Scientific) at -20°C, the brains were sectioned to 30 µm with a cryostat.Brain slices were stored in cryopreservation media (30% ethylene glycol, 30% sucrose, 0.1 M PBS, pH 7.35).The dissected brain slices were washed in PBS, stained with Hoechst 33342 for 10 minutes, and mounted on SuperFrost Plus slides (Fisher).The tissue samples were covered with a glass cover slip and sealed with clear nail polish prior to imaging.Lasers of 405 nm (collect 410 nm to 480 nm), 488 nm (collect 490 to 550 nm), and 561 nm (collect 570 to 630 nm) were used.All images shown use the same minimum and maximum settings.

ATP luminescent measurements
One three-month old BALB/c female mouse and one three-month old 5x-FAD female mouse were sacrificed using CO2 as the primary method of euthanasia and cervical dislocation as the secondary method of euthanasia.Their brains were harvested, washed with PBS, and fixed overnight in 4% PFA at 4°C.Brains were then dehydrated with 30% sucrose in PBS at 4°C for 2 days.The whole brain was then suspended in 200 µL PBS in a 1.5 mL Eppendorf tube and flash frozen in liquid nitrogen, allowing the nitrogen to enter the tube.Once frozen, a pestle homogenizer was used to grind the tissue.After complete grinding, the samples were centrifuged at 12,000g for 5 minutes to separate the debris.10 µL of supernatant was added in triplicates to a white opaque 96 well plate (Cayman Chemical) for each sample.Using the Luminescent ATP detection kit (Cayman Chemical), all manufacturer instructions were following for sample preparation and standard curve generation.After sample preparation, samples incubated in the dark at room temperature for 15 minutes before measurement.Luminescence samples were read using a BioTek plate reader, with 1 second integration times.

Supporting Figures
Figure S1: Nanosight to verify exosome size and purity.BEC-Exo and NBC-Exo size, purity, and concentration were characterized by Nanosight NS500.5 µL of the resuspended exosome stock was diluted in 995 µL of PBS.Run times of 1 minute, with the camera level set to 8 at room temperature and screen gain set to 1 were used to collect size and purity data of the freshly purified exosomes.Both exosome sizes peaked at ~110 nm, with minor exosomal aggregation observe through the doublet and multimer formation.Exosomes were loaded with the Rhodamine Green labeled ATP aptamer through electroporation, sonication, freeze-thaw, or incubation (detailed procedures for each loading method are described in the methods section).After each method, 3 cycles of ultrafiltration with 50 kDa MWCO Amicon tubes were performed to purify the aptamer-loaded exosomes from unloaded aptamer.The aptamer-loaded exosomes were incubated with 0.1% Triton X-100 at 37°C for 10 minutes before fluorescent measurements to fully break the exosomal membrane and allow for the Rhodamine Green-labeled ATP aptamer to be released into the solution.To determine the loading efficiency, the fluorescence emission at 520 nm for each sample was fit into the standard curve best fit equation from (D).After solving for the corresponding concentration based on the standard curve, the loading efficacy was calculated as a percentage of the final ATP aptamer concentration over the initial aptamer concentration added to each sample before the loading procedures.(F) After optimizing the loading procedure, NBC-Exos were loaded with the ATP aptamer sensor (hybridized with a quencher strand).The samples were then incubated with either 2 mM ATP or 2 mM ATP + 0.1% Triton X-100 for 10 minutes at 37°C for 10 minutes to determine active sensor response.All measurements were taken in 50 mM Tris, 100 mM NaCl, 5 mM MgCl2 at 37°C.Excitation at 480 nm, emission collected at 520 nm, with 5 mm slit widths.Sensor loaded exosomes that showed slight exosome size distribution changes compared to preloaded exosomes.However, no significant distribution changes occurred, such as major aggregation.(C) Rhodamine Green-labeled aptamer integrity after loading was checked with a 10% urea denaturing PAGE gel.After each tested loading method, 10 nmol of total DNA was heated with 2X gel loading dye stop solution to 95°C for 10 minutes.The samples were loaded onto a 10% urea denaturing PAGE gel and run at 26 W for 1 hour.As each lane showed only one detectable band that corresponded to the control lane (ATP aptamer that was only incubated in buffer), we determined the loading methods did not degrade the aptamer.Detailed loaded methods are outlined in the methods section.To determine how ATP changes in Alzheimer's disease, we carried out the same delivery studies across the BBB in 5xFAD mice.Following the general exosome sample preparation and tail vein injection protocols, mice were sacrificed using CO2 and cervical dislocation.The brains were washed in 3X with PBS in a petri dish.Within 1 hour of excision, the brains were placed on the IVIS instrument and imaged using 640 nm excitation, 680 nm emission, f number 4, binning factor 8, and 6.6 field of view.Two biological replicates were performed for each sensor and experimental sample.Minimal background signal is observed in the unloaded controls due to unwashed sensor remaining in the bloodstream.Exosome sensors display higher signal intensity due to signal sequestered within the brain.Compared to the healthy BALB/c mice, the negative controls displayed higher background fluorescence.This may be due to the weaker BBB structure in neurodegenerative diseases.

Figure S2 :
Figure S2: Quantify exosome loading efficiency and exosome encapsulated sensor response.(A) Hybridized ATP aptamer quencher strand optimization compared to a no quencher control.(B) Zoomed in data on Figure S2A to show quenching efficiency optimization.(C) Hybridized ATP aptamer and hybridized inactive control fluorescence intensities compared to ATP concentrations.The limit of detection was calculated by LOD = 3σ/S, where σ is the standard deviation of blank solutions and S is the slope of calibration curve of the active aptamer.This determined a limit of detection of 0.06 mM ATP. (D) Standard curve of fluorescence intensity vs known concentrations of the Rhodamine Green-labeled ATP aptamer.(E) Loading efficiency of different vesicle loading techniques.Exosomes were loaded with the Rhodamine Green labeled ATP aptamer through electroporation, sonication, freeze-thaw, or incubation (detailed procedures for each loading method are described in the methods section).After each method, 3 cycles of ultrafiltration with 50 kDa MWCO Amicon tubes were performed to purify the aptamer-loaded exosomes from unloaded aptamer.The aptamer-loaded exosomes were incubated with 0.1% Triton X-100 at 37°C for 10 minutes before fluorescent measurements to fully break the exosomal membrane and allow for the Rhodamine Green-labeled ATP aptamer to be released into the solution.To determine the loading efficiency, the fluorescence emission at 520 nm for each sample was fit into the standard curve best fit equation from (D).After solving for the corresponding concentration based on the standard curve, the loading efficacy was calculated as a percentage of the final ATP aptamer concentration over the initial aptamer concentration added to each sample before the loading procedures.(F) After optimizing the loading procedure, NBC-Exos were loaded with the ATP aptamer sensor (hybridized with a quencher strand).The samples were then incubated with either 2 mM ATP or 2 mM ATP + 0.1% Triton X-100 for 10 minutes at 37°C for 10 minutes to determine active sensor response.All measurements were taken in 50 mM Tris, 100 mM NaCl, 5 mM MgCl2 at 37°C.Excitation at 480 nm, emission collected at 520 nm, with 5 mm slit widths.

Figure S3 :
FigureS3: Evaluation of exosome and aptamer integrity pre and post loading.Exosome size, purity, and concentration were characterized by Nanosight NS500.(A) 5 µL of the resuspended exosome stock was diluted in 995 µL PBS buffer.Run times of 1 minute, with the camera level set to 8 at room temperature and screen gain set to 1 were used to collect size and purity data of unloaded exosomes.(B) Rhodamine Green-labeled ATP aptamer sensor loaded exosomes were rerun on the Nanosight using the same settings.Sensor loaded exosomes that showed slight exosome size distribution changes compared to preloaded exosomes.However, no significant distribution changes occurred, such as major aggregation.(C) Rhodamine Green-labeled aptamer integrity after loading was checked with a 10% urea denaturing PAGE gel.After each tested loading method, 10 nmol of total DNA was heated with 2X gel loading dye stop solution to 95°C for 10 minutes.The samples were loaded onto a 10% urea denaturing PAGE gel and run at 26 W for 1 hour.As each lane showed only one detectable band that corresponded to the control lane (ATP aptamer that was only incubated in buffer), we determined the loading methods did not degrade the aptamer.Detailed loaded methods are outlined in the methods section.

AFigure S4 :Figure S5 :Figure S6 :
Figure S4: Monocultured bEnd.3 cell delivery confocal images.Lipofectamine 3000 or the range of native exosomes were used to deliver the Rhodamine Green-labeled ATP aptamer sensor or its scrambled inactive control into monocultured bEnd.3 cells.After a four hour delivery, cells were washed to remove background signal and nuclei were stained with Hoechst 33342.Successful aptamer delivery was visualized through the green fluorescence channel, with all active aptamer groups displaying higher fluorescence signal than their inactive scrambled control.Scale bar is 20 µM.

Figure S7 :
Figure S7: Eight-hour incubation in Transwell BBB model with the ATP aptamer sensor.Lipofectamine 3000 or the range of native exosomes were used to deliver the Rhodamine Greenlabeled ATP aptamer sensor or its scrambled inactive control into the Transwell BBB model.After an eight-hour incubation, cells were washed to remove background signal.Cells were fixed, permeabilized, and stained with anti-Rab11A.(A) Representative images for each sample for the top bEnd.3cell layer.(B) Representative images for each sample for the target SY5Y cell layer.Scale bar is 20 µmeter.Statistical analysis was determined through a one-way ANOVA, with data represented as means of technical replicates and error as standard deviation of the mean.

Figure S8 :
Figure S8: Cell toxicity assay for model cell types.MTT assay to determine exosomal cytotoxicity for brain endothelial cells (A) and brain neuroblastoma cells (B).Positive controls of 30% hydrogen peroxide were used to achieve maximum cell death as a comparison to a negative control of only Opti-MEM and samples of sensor-loaded exosomes.Based on the minimal change in absorbance compared to the negative control, all of the sensor-loaded exosomes displayed minimal cellular toxicity at concentrations used for in vitro cellular studies and in vivo mouse studies.Experiment was performed with three biological replicates.Error show in standard deviation of the mean.

Figure S10 :
Figure S10: Five-hour incubation in Transwell BBB model with the ATP aptamer sensor for recycling endosome modulation.Lipofectamine 3000 or the range of native exosomes were used to deliver the Rhodamine Green-labeled ATP aptamer sensor or its scrambled inactive control into the Transwell BBB model.After a five-hour delivery, cells were washed to remove background signal.Cells were fixed, permeabilized, and stained with anti-Rab11A for simultaneous evaluation with the recycling endosome pathway.(A) Representative images for each sample for the top bEnd.3cell layer.(B) Representative images for each sample for the target SY5Y cell layer.(C) Quantified intracellular fluorescence intensity.Scale bar is 20 µmeter.Statistical analysis was determined through a one-way ANOVA between groups, with data represented as means of technical replicates and error as standard deviation of the mean.

Figure S11 :Figure S12 :
Figure S11: Live mice ATP aptamer sensor-exosome tracking.Following the general protocol for exosome loading and tail vein injections of mice, in vivo imaging was performed to evaluate the circulation of the Cy5-labeled ATP aptamer sensor over 48 hours post tail vein injection.As a negative control to account for bloodstream autofluorescence, we used PBS injections.As a negative control to account for Cy5-DNA remaining in the bloodstream, we injected the unloaded hybridized scrambled control and the unloaded active ATP aptamer sensor.For the positive samples to account for brain delivery, we injected the exosome loaded hybridized scrambled control and the exosome loaded hybridized active ATP aptamer sensor.Image parameters used were 640 nm excitation, 680 nm emission, f number 4, binning factor 8, and 6.6 field of view.Notably, exosome samples contained persistent fluorescent signal over negative controls of unloaded sensor even at the 48 hour time point.Likewise, slight central nervous system signal accumulation was observed with the exosome-mediated samples.

Figure S15 :Figure S16 :
Figure S15: High resolution imaging of the ATP aptamer in the cortex in freshly frozen brain slices after tail vein injections into BALB/c mice.Evaluating the exosome delivery locations, sensor output compared to the inactive control, and exosome and sensor colocalization with the nucleus.Scale bars are 20 µmeter.

Figure S20 :
Figure S20: Positive control of a secondary method to evaluate whole brain ATP levels.(A) Standard curve of ATP concentrations in vitro, as measured by luciferase-luciferin pairs.(B) Luminescence measurements of a healthy mouse brain compared to an Alzheimer disease model 5xFAD mouse brain.When accounting for the sample volume and converting these luminescence measurements to molarity, we found the healthy brain ATP concentration was 2.02 mM and the 5xFAD brain ATP concentration was 1.53 mM.Experiment was performed in technical triplicates, with error bars representing standard deviation.Statistical analysis is an unpaired t-test, with p* < 0.05.