Sono-mechanical nanostructures enable sustained and precise ultrasound brain stimulation
DOI: Pending DOI
Synopsis
Transcranial ultrasound is promising for noninvasive deep brain stimulation, but its spatial precision and long-term reliability remain limited. We developed PEGylated hollow silica nanostructures (HSN) that localize and amplify acoustic effects to enable sustained ultrasound neuromodulation without genetic modification. In vitro, HSN enhanced neuronal calcium responses through mechanosensitive ion channel activation. In vivo, HSN enabled spatially confined activation in motor cortex, striatum, ventral tegmental area, and subthalamic nucleus, with functionality maintained for up to 9–11 weeks. In Parkinsonian mice, chronic STN stimulation increased dopaminergic signaling and improved motor performance without evident toxicity.
Background
Ultrasound offers a noninvasive route for deep brain neuromodulation with favorable spatiotemporal resolution, but conventional transcranial ultrasound still lacks the precision required to stimulate small neuronal populations reliably. Nanobubble-assisted strategies can confine and amplify acoustic effects, yet their limited in vivo stability restricts long-term use and repeated administration may increase tissue burden. To overcome these barriers, we engineered hollow silica nanostructures (HSN) with a gas core, iron-doped silica shell, and PEG surface modification. We hypothesized that HSN could provide durable acoustic sensitization, support localized activation of targeted brain circuits, and enable chronic therapeutic neuromodulation without viral transduction or implanted electrodes. This study therefore aimed to establish a stable sono-mechanical platform for precise brain stimulation and to evaluate its efficacy, persistence, mechanism, and biosafety in mice, including its therapeutic potential in Parkinson’s disease models.
Methods
HSN were synthesized as iron-doped, PEGylated hollow silica particles with an approximately 220 nm diameter and characterized by TEM, DLS, elemental mapping, FTIR, ultrasound contrast testing, and neuronal viability assays. In vitro neuromodulation was tested in primary cortical neurons cultured on soft polyacrylamide gels using low-intensity pulsed ultrasound (1.0 MHz; up to 0.32 MPa); calcium imaging, c-Fos staining, and gadolinium blockade were used to assess activation and mechanosensitive channel involvement. For in vivo studies, HSN or saline were stereotactically injected into mouse motor cortex, striatum, ventral tegmental area, or subthalamic nucleus. Neural responses were measured by EMG, c-Fos mapping, fiber photometry of GCaMP6s calcium signals, and dopamine sensing with gDA3m. HSN persistence was assessed by fluorescence imaging, MRI, and ultrasound imaging over weeks. Therapeutic efficacy was evaluated in MPTP and α-synuclein PFF Parkinsonian mouse models using chronic STN ultrasound stimulation with behavioral testing. Biosafety was assessed by body weight, locomotion, memory, cognition, apoptosis, and neuroinflammatory markers.
Results
HSN exhibited strong ultrasound responsiveness, stability, and biocompatibility. In cultured neurons, HSN plus ultrasound produced rapid, reversible, intensity-dependent calcium transients, whereas solid silica nanoparticles or ultrasound alone produced minimal responses. Maximal ΔF/F0 reached 10.5 ± 1.6% at 0.13 MPa, 21.5 ± 2.9% at 0.24 MPa, and 28.3 ± 1.5% at 0.32 MPa. Responses were suppressed by gadolinium to 30.9% of control and recovered to 93.7% after washout, supporting a mechanosensitive ion channel mechanism. More than 83% of neurons responded, and c-Fos expression increased more than threefold. In vivo, HSN enabled reliable, dose-dependent EMG responses following motor cortex stimulation with latencies within 160 ms. In the striatum, ultrasound-induced c-Fos was confined to the HSN-injected region (~200 µm) and not neighboring tissue lacking HSN, demonstrating spatial selectivity governed by nanoparticle localization rather than ultrasound wavelength. HSN remained detectable in brain tissue for over 2 months, with MRI contrast evident up to 10 weeks and echogenicity maintained for at least 11 weeks, greatly exceeding the persistence of conventional synthetic nanobubbles. Chronic functionality was confirmed in the ventral tegmental area, where HSN-enabled calcium responses persisted for up to 9 weeks with stable latencies (~208–212 ms). At week 9, c-Fos-positive cells were significantly increased in the stimulated VTA (20.56 ± 2.51% vs 4.94 ± 1.65% without ultrasound), without increased activation in adjacent midbrain, contralateral VTA, or auditory cortex. In Parkinsonian mice, repeated ultrasound stimulation of the HSN-injected subthalamic nucleus progressively improved motor performance. In the MPTP model, rotarod latency increased from 40.67 ± 6.5 s at baseline to 61.11 ± 10.72 s at week 9, with sustained benefit at week 11 (56.22 ± 10.21 s). Open-field distance also increased from 8.19 ± 2.02 m to 11.78 ± 2.32 m by week 9. Dopamine sensor recording showed rapid striatal dopamine increases after STN stimulation (latency 393.3 ms), and TH preservation in the substantia nigra suggested mitigation of dopaminergic degeneration. In the α-synuclein PFF model, HSN plus ultrasound also improved rotarod performance (58.18 ± 7.38 s vs 40.25 ± 7.05 s with ultrasound alone) and enhanced locomotion and mobile time. Across studies, HSN caused no detectable systemic toxicity, behavioral impairment, apoptosis, or neuroinflammation.
Discussion
These findings establish HSN as a durable sono-mechanical interface that converts low-intensity ultrasound into spatially confined, chronic neuromodulation. By overcoming the short lifetime of nanobubbles, HSN extend functional acoustic sensitization from days to at least 9–11 weeks, enabling repeated stimulation of deep brain circuits without genetic modification, chronic implants, or repeated intracranial dosing. The platform combines precision, persistence, and translational simplicity, and in Parkinsonian mice it reproduces key therapeutic features associated with deep brain stimulation, including circuit activation, dopamine release, and motor rescue. Importantly, the absence of overt toxicity or neuroinflammation supports the feasibility of long-term use. Together, this work positions HSN-enabled ultrasound as a minimally invasive strategy for chronic circuit modulation and a promising foundation for future neurological therapies.
Disclosures
This work was financially supported by the Hong Kong Research Grants Council Collaborative Research Fund (C5053-22GF), General Research Fund (15224323 and 15104520), Hong Kong Innovation and Technology Fund (MHP/014/19), the National Key Research and Development Program of the Ministry of Science and Technology of China (2023YFC2410900), internal funding from The Hong Kong Polytechnic University (G-SACD, 1-W35S, and 1-YWDQ), and the Research Institute of Smart Ageing (1-CDJM). Xuandi Hou, Jianing Jing, Yizhou Jiang, and Lei Sun declare a pending patent related to this manuscript (Application No. US20250114644A1). Zhuohan Shi declares no competing financial interests.