Animal model

A total of 95 C57BL/6JNarl mice, aged 7–9 weeks and weighing 20–30 grams, were used in our study ( Table 1 ). Their health was thoroughly assessed, and they were allowed to adapt to the experimental environment before administration. This research was performed in accordance with the Institutional Animal Care and Use Committee guidelines of the University (approval number 20230261). The experiments ( Table 1 ) were conducted to compare the IN-only group, the IN-and-US (IN+US) group, and the sham group.

Nasal dextran delivery

In the IN delivery, substances were precisely instilled through the nostrils for targeted administration and absorption. Then, 10-kDa dextran (conjugated with Alexa Fluor™ 555, Invitrogen) was prepared at a concentration of 10 mg/mL. IN delivery was conducted on the mice anesthetized by intraperitoneally injecting Zoletil (50 mg/kg) and Xylazine (2.5 mg/kg). The mice were positioned upright using a support apparatus, and 4 µL of the dextran solution was carefully administered to one nostril at a time, alternating between the left and right nostrils ( Fig. 1 a, left). Dextran was IN administered using a 10μL tip (Rainin™; Fig. 1 a, right). The solution was delivered to each nostril for 2 min, with a total delivery of up to 40 μL of dextran. Care was taken to ensure that dextran did not spill out of the nasal cavity throughout the procedure. After dextran was IN administered, the mice were kept in an upright position for 30 min ( Fig. 1 a, b) and fixed in a stereotaxic frame (stereotaxic Instrument 18-7216, BioMed Easy Technologies Co., Ltd) for ultrasound treatment.

Establishment of a nose-to-brain drug delivery model with ultrasound treatment in mice. The mice were positioned upright in a support apparatus (1a, left) and during IN dextran administration with a 10μL tip (1a, right). A flow diagram of the experiments (1b) shows that mice received intranasal drug administration followed by whole-brain ultrasound using a planar ultrasound transducer (1c).

Transcranial ultrasound treatment

The mice were subjected to whole-brain ultrasound treatment twice with 10-min intervals using a self-made 1 MHz planar ultrasound transducer ( Fig. 1 b) [ ]. The head fur was shaved and the scalp was exposed; ultrasound gel was applied between the transducer and the sonicated area at the center of the brain ( Fig. 1 c). The ultrasound transducer was positioned at the midpoint of the line connecting the mouse’s ears and eyes, with the transducer oriented parallel to the exposed skull. This custom-made ultrasound transducer was prepared with a diameter of 15mm based on our previous publications [ , ], and the pressure field of the transducer was measured with a 2.0 mm calibrated needle hydrophone (Precision Acoustics, Dorchester, UK) in the free field within the degassed water ( Supplement Fig. 1A ). The output acoustic pressure (kPa) and intensity of spatial-peak temporal average (Ispta) at the applied input voltage (mVpp) were measured. Before the in-vivo experiment, three ex-vivo conditions were tested using a needle hydrophone to assess the attenuated transcranial acoustic pressure: (1) ultrasound gel alone, (2) the scalp and skull, and (3) the whole mouse head ( supplementary fig. 1B ). The transcranial propagation analysis revealed a 10.4% reduction in acoustic pressure when comparing transmission under condition 2 (scalp and skull) to ultrasound gel alone; additionally, a reduction of 65.9% was observed when comparing condition 3 (whole mouse head) to ultrasound gel alone ( supplement fig. 1C ). Three ex-vivo experiments showed no increase in temperature. Ultrasound was generated using a function generator (Tektronix AFG1022, USA) paired with a power amplifier (E&I 210L, Electronics & Innovation, USA). The parameters were based on previously published settings for the low-energy ultrasound treatment of the whole brain: center frequency = 1 MHz, pulse repetition frequency = 1 kHz, duty factor = 1%, Ispta = 3.68 mW/cm² (equal to pressure level 100 kPa), duration of each stimulation = 5 min [ , ]. Contrast agents were not utilized in our study.

After anesthesia, the mouse’s abdomen and ribcage were carefully opened to expose the heart and liver. A needle was inserted into the left ventricle, and a small incision was made in the right atrium to allow blood to flow out. The circulatory system was initially perfused with physiological saline to flush out blood, as indicated by a pale liver; it was subsequently perfused with a formalin solution (Sigma-Aldrich, HT501128) to fix the tissues. Once perfusion was complete, the brain was harvested for further analysis. The mice were euthanized at 4 and 24 h post-IN administration.

Analysis of the efficiency of nasal drug delivery

Whole brain imaging (Experiment 1)

Before the brain was sectioned, the fluorescence intensity of the mouse brains was imaged from dorsal and ventral perspectives under a fluorescence macroscope (Cat No. MVX10, Tokyo, Olympus) equipped with an ORCA-381 Spark digital CMOS camera (Cat No. C11440-36U, Hamamatsu, Japan). Images of the excised TN were captured under consistent exposure settings throughout the entire imaging session and from all experimental groups ( Fig. 2 a).

Whole brain imaging analysis. The isolated mouse brains were examined from dorsal and ventral views, along with the trigeminal nerves (2a) (OB, olfactory bulb). The macroscopic images of the dorsal and ventral views, and the trigeminal nerves, were captured for the IN-only, IN+US, and sham groups (2b). Quantification of fluorescence intensity for the whole brain (dorsal and ventral views) and trigeminal nerves revealed, through Mann-Whitney U test, tracer delivery to the olfactory bulb in the IN+US group significantly improved compared with that in the IN-only and sham groups from the dorsal view; the delivery to the whole brain and trigeminal nerves also increased in the IN-only and IN+US groups compared with that in the sham group at the 4 h mark (2c).

Immunofluorescence staining (Experiment 4)

Brain sagittal slices (500 or 100 μm) were incubated in 2% PBST (2% Triton X-100 in PBS) at 25°C for 3 days, and washed in PBS thrice at 25°C for 15 min per wash. The slices were blocked (blocking buffer: 10% goat serum, 1% Triton X-100, 2.5% DMSO, 0.05% sodium azide in PBS) at 4°C for 3 days. The slices were incubated with a primary antibody (antibody dilution buffer: 1% goat serum, 0.2% Triton X-100, 2.5% DMSO, 0.05% sodium azide in PBS) at 4°C for 3 days. The sections were washed thrice with the buffer (3% NaCl, 0.2% Triton X-100 in PBS) at 25°C for 1 h, and then incubated at 4°C overnight. Afterward, they were incubated with a secondary antibody at 4°C for 3 days. They were washed thrice with a washing buffer at 25°C for 1 h, incubated at 4°C overnight, treated with RapiClear® 1.49 for 1 h, and mounted on slides. For the endothelial cell marker, CD31 (PECAM1) was labeled using a primary CD31 mouse monoclonal antibody (OriGene, AM32708PU-N) at a 1:100 dilution, and with a secondary goat antimouse antibody conjugated with Alexa Fluor™ 633 (Invitrogen, A-21050) at a 1:200 dilution. Imaging was performed using a laser scanning confocal microscope (Zeiss, Jena, Germany), and the images were analyzed with ImageJ software.

Hematoxylin and eosin (H&E) staining (Experiment 5)

H&E staining was performed on brain sections to identify potential vessel-injury-related erythrocyte extravasation or ischemia-associated cell necrosis and to evaluate the safety of the ultrasound treatment. The following regions were selected for analysis: the olfactory bulb, cerebral cortex, and thalamus. Whole brain tissue blocks were embedded in paraffin, sectioned sagittally at a thickness of 4 μm, and subjected to H&E staining with Tissue-Tek Prisma® Plus Automated Slide Stainer (Sakura™, Nagano, Japan) for histological assessment.

Statistical analysis

Data were statistically analyzed using SPSS 17.0 (IBM SPSS Statistics). The normality of distribution was examined with the Shapiro-Wilk test. Experimental groups were compared using Mann-Whitney U test for the quantification of fluorescence intensity in whole brain imaging, brain slice imaging and tracer concentration of brain tissues and TN. The olfactory bulb was measured as a combined fluorescence quantification of both sides. Data with p < 0.05 were considered statistically significant