An Ultrasound-Programmable System for Highly Sensitive and Spatiotemporally Controlled Drug Release in Deep Tissues

Abstract

A recent study published in Nature by Wang et al. [1] presents an innovative approach utilizing a focused ultrasound (FUS)-programmable hydrogen-bonded organic frameworks (HOFs) system, which facilitates noninvasive, spatiotemporal drug release deep within tissues, holding significant potential for advancing neuromodulation in deep brain regions. The HOFs self-assemble into highly ordered porous structures through hydrogen bonding and π-π stacking interactions, which undergo controlled dissociation upon FUS stimulation, enabling on-demand drug release (Figure 1A). The weak noncovalent interactions within HOFs permit mechanochemical activation under FUS, allowing for precise spatiotemporal control over therapeutic interventions.

Traditional drug delivery systems (DDS) primarily rely on passive diffusion for targeting diseased areas. However, this passive approach often results in undesirable systemic distribution, limiting the therapeutic efficacy and increasing the off-target risk [2]. Consequently, stimuli-responsive DDS have been developed, including light-triggered and pH-responsive systems, aimed at enhancing drug targeting and controlling release kinetics. Despite these advancements, light-based activation is restricted by the penetration depth of light in biological tissues, which impedes its ability to reach deep-seated lesions. Implantable optical fibers can transmit light signals, but their invasive nature introduces the risk of tissue damage. Additionally, pH-responsive DDS are sensitive to the biochemical microenvironment of diseased tissues, and interpatient variability often leads to inconsistent therapeutic results, preventing clinical translation [3]. Therefore, the development of a highly efficient, noninvasive, and remotely controllable DDS remains a critical goal in precision medicine.

Ultrasound, with its noninvasive nature, deep tissue penetration, and precise spatiotemporal control, has emerged as a promising candidate for next-generation intelligent DDS. FUS, in particular, enables the modulation of deep tissues with millimeter-scale spatial precision, offering significant safety advantages. Ultrasound-induced mechano-responsive cleavage of labile covalent or noncovalent bonds introduces new possibilities for highly controlled drug release [4, 5]. However, the strong valence bond interactions within polymer frameworks often necessitate high power densities for effective ultrasonic dissociation, leading to prolonged response times of several hours. Moreover, the complex topological structures of these frameworks lack established theoretical models to explain the relationships between dissociation efficiency, molecular architecture, and ultrasound power. Therefore, the development of ultrasound-programmable systems with tunable structural stability and ultrasound sensitivity remains a critical challenge in the field.

Porous frameworks such as covalent organic frameworks (COFs) and metal-organic frameworks (MOFs) are known for their excellent drug-loading capacity and well-ordered structures, making them attractive candidates for drug delivery applications. However, the inherent strength of metal–ligand coordination networks impedes their responsiveness to ultrasonic stimuli. In response, Wang et al. [1] engineered a series of four distinct HOFs nanocrystals, incorporating aromatic rings and varying carboxylic acid densities. These HOFs, namely HOF-TATB, HOF-BTB, HOF-101, and HOF-102 (Figure 1B), exhibited dissociation equilibrium under different FUS pressures, following the order: HOF-TATB (91.8%) > HOF-BTB (45.3%) > HOF-101 (11.6%) > HOF-102 (4.7%). The dissociation of these HOFs under ultrasound was a thermodynamic process governed by ultrasound-induced mechanical stress rather than thermal effects. The dissociation efficiency increased with ultrasound peak pressure, with the rate of increase depending on the specific HOF type. This ultrasound-programmable activation of HOFs was termed “UltraHOF.”

A theoretical model was developed to better understand the dissociation mechanism of these ultrasound-programmable HOFs. By calculating the cohesive energy (E_cohesive) required to break down the HOF crystals into their monomeric building units, the study demonstrated a sequential increase in E_cohesive values for HOF-TATB, HOF-BTB, HOF-101, and HOF-102, suggesting a corresponding increase in relative stability. The enhanced ultrasonic stability of these HOFs is attributed to a higher number of hydrogen bonds due to increased carboxyl group content, as well as stronger π-π interactions from extended aromatic fused rings. Since mechanochemical cleavage tends to occur preferentially at weaker bond sites, the design of ultrasound-sensitive HOFs should prioritize organic molecular building units (OMBUs) with fewer hydrogen bonds and simpler aromatic frameworks.

The HOF nanocrystals with good biocompatibility both in vitro and in vivo exhibited exceptional drug-loading capacity with minimal premature release in the absence of ultrasound stimulation. HOFs with lower E_cohesive values demonstrated higher release rates under ultrasound, consistent with theoretical predictions. Among the different HOFs, HOF-TATB proved to be the most ultrasound-sensitive, releasing the drug at relatively low ultrasound pressure (0.51 MPa), accompanied by nanocrystal fragmentation. In addition to loading clozapine N-oxide (CNO) for ultrasound-programmable drug release, the HOFs can serve as a universal platform for activating a variety of drugs.

The HOF-TATB-loaded CNO (TATB@CNO) was utilized to explore deep brain activity under FUS stimulation. At an applied pressure of 1.4 MPa, the acoustic pressure at the ventral tegmental area (VTA) reached approximately 0.9 MPa, sufficient to trigger CNO release and activate designer receptors. Ultrasound irradiation of TATB@CNO within the VTA induced rapid neuronal activation with high temporal resolution, occurring within just 3.5 s. Additionally, each 10-second sono-chemogenetic stimulation resulted in sustained neural activity lasting over 120 s. The sono-chemogenetic modulation of reward-learning behavior in mice increased their preference for environments associated with ultrasound stimulation. Moreover, in the forced swim test, these mice showed a significant increase in swimming distance, indicating enhanced locomotor activity.

Overall, the HOF-based platform proposed by Wang et al. [1] offers exceptional spatial resolution, enabling precise drug release within a millimeter-scale region. Drug release occurs rapidly upon ultrasound activation, and unlike traditional ultrasound-triggered DDS, HOFs feature a lower activation threshold. Their noninvasive, remotely controllable nature eliminates the risks associated with implantable devices, making them a safer, more versatile alternative.

The sensitivity of DDS carriers to ultrasound is a key determinant in the spatiotemporal responsiveness of drug release. Therefore, it is essential to refine theoretical models to systematically investigate the influence of crystal structures on ultrasound-triggered HOF dissociation. By gathering additionally experimental data and computational methods, the optimized models are expected to improve predictive accuracy. Furthermore, the structural and functional enhancement of HOF carriers, such as optimizing pore size and surface properties, should be explored to improve drug-loading capacity and targeted delivery efficiency, better aligning with clinical therapeutic needs. In practical applications, drug release from HOFs should account for both ultrasound sensitivity and the physiological environment factors, such as pH, ion strength, enzymatic degradation, and protein adsorption, to prevent off-target release. Ultrasound propagation in human tissues should be systematically studied, and parameters optimized to enhance penetration depth and energy transfer efficiency. To achieve more precise monitoring of the ultrasound-based therapeutic process, multimodal synergistic treatment strategies should be developed by integrating ultrasound therapy with imaging technologies, such as ultrasound imaging, photoacoustic imaging, and magnetic resonance imaging, enabling real-time monitoring and accurate drug delivery localization. Expanding the range of drugs loaded into HOFs and evaluating the platform efficacy for treating deep-tissue diseases, such as tumors and cardiovascular diseases, will further broaden their biomedical applications. With continued optimization and multifunctional expansion, HOF-based systems hold great promise for advancing biomedical applications, offering more intelligent solutions for precision medicine.

Chenyao Wu: funding acquisition, writing – original draft, visualization. Lili Xia: writing – original draft, visualization. Wei Feng: funding acquisition, writing – review and editing, supervision, formal analysis. All authors have read and approved the final manuscript.

The authors have nothing to report.

The authors declare no conflicts of interest.