The core insight of the review is that focused ultrasound, or FUS, does far more than heat tissue. When an acoustic beam is focused to a small target volume, it deposits energy through three broadly distinct mechanisms: mechanical effects driven by acoustic radiation forces, cavitation-driven effects arising from the dynamics of microscopic gas bubbles, and thermal effects from the absorption of ultrasound energy. Cavitation is particularly dramatic. When pre-existing or newly formed microbubbles oscillate and then implode under acoustic pressure, they generate localized regions of extreme pressure and temperature, producing shockwaves and microjets that can stretch the cell membrane into transient, tiny pores—a phenomenon called sonoporation that allows molecules and ions to pass through without permanently damaging the cell. In parallel, acoustic radiation forces transfer momentum to tissue during sound propagation, displacing and deforming cell membranes at the focal point, while acoustic streaming induces steady shear stresses that further perturb cellular and subcellular structures.

These physical perturbations are not simply destructive; they are informative. The Rice team emphasizes that cells interpret FUS-induced mechanical forces through mechanotransduction—the conversion of mechanical stimuli into biochemical signals. Matrix-anchored cells detect these disturbances through the integrin-adhesion plaque complex, transmitting them along actin stress fibers, while suspended cells experience shear force directly at the plasma membrane. Forces propagating through the cytoskeleton can even reach the nucleus via the linker of nucleoskeleton and cytoskeleton complex, influencing chromatin organization and gene expression. But the most striking mechanistic story involves mechanosensitive ion channels. PIEZO1 has been repeatedly identified as a primary mechano-gated channel responsive to acoustic radiation force-driven membrane tension: low-intensity FUS rapidly activates PIEZO1-dependent calcium influx in osteoblastic precursor cells, promoting ERK signaling and cytoskeletal remodeling, while in prostate cancer models nonthermal ultrasound pulses induce PIEZO1-mediated calcium entry that causes mitochondrial depolarization and caspase-3 activation, sensitizing tumors to TRAIL-mediated apoptosis. The TRPV4 channel, meanwhile, has emerged as a key sonosensor at the blood-brain barrier, where cavitation and radiation force-induced membrane strain gates TRPV4-dependent calcium influx, engaging a Ca²⁺/PKC-δ cascade that drives reversible tight-junction opening. Two-pore domain potassium channels such as TREK-1 and TRAAK add another dimension, converting FUS-induced membrane tension into hyperpolarizing leak currents that dampen neuronal excitability—in one remarkable study, transcranial low-intensity FUS targeting TRAAK-overexpressing brain neurons suppressed sympathetic drive and prevented malignant arrhythmias after myocardial infarction.

The therapeutic implications of this channel-level control are profound. Calcium signaling is a master regulator of cell fate, and FUS can push it in either direction depending on acoustic parameters. Low-intensity pulsed ultrasound enhances tissue regeneration and migration, whereas elevated mechanical forces trigger apoptosis through extensive DNA damage or altered mitochondrial permeability. In hepatocellular carcinoma models, FUS stimulation suppressed tumor proliferation by more than 70 percent in H22-HCC cells and more than 83 percent in Hepa1-6-HCC cells, along with significantly prolonged survival. In immunotherapy contexts, high-intensity ultrasound triggered the calcium-dependent NFAT pathway in T cells, producing stronger immune responses and memory that effectively inhibited tumor recurrence and metastasis. The review also highlights FUS’s capacity to transiently and locally open the blood-brain barrier—a critical translational goal, since passive diffusion across the barrier typically favors only small lipophilic molecules under roughly 400 to 500 Daltons, yet nearly 98 percent of approved small-molecule drugs exceed this threshold. Preclinical and early clinical studies of FUS-mediated BBB opening demonstrate spatially defined, reversible increases in regional permeability that allow therapeutic agents and biologics to access previously restricted brain regions.

The second pillar of the framework concerns nanoparticles as active partners rather than passive cargo holders. Compared with microbubbles—which are 1 to 8 micrometers in diameter, confined largely to vascular compartments, limited in drug-loading capacity, and short-lived in circulation—nanoparticles in the 20 to 200 nanometer range offer tunable size, broad surface functionalization, prolonged systemic circulation, and the ability to access extravascular and interstitial spaces. Crucially, nanoparticles can be engineered as transducers that convert acoustic cues into on-demand structural reconfiguration or bond cleavage. The review catalogs three classes of FUS-triggered chemical bond scission. Surface-anchoring bonds can be severed to shed protective shells: silica core-shell nanoparticles bearing a PEG brush attached via force-labile azo bonds remain stable during circulation until FUS-induced mechanical perturbation triggers PEG detachment, activating free radical generation and cytotoxicity. Prodrug-linker bonds embed sono-labile chemistry at the drug-carrier junction: singlet oxygen generated by therapeutic ultrasound can cleave a urea linkage between carboxyferrocene and methylene blue, switching an inert nanodrug into a Fenton-active ferroptosis inducer at the tumor site. Backbone and crosslink bonds determine whether ultrasound destabilizes the entire carrier framework, as in diselenide-crosslinked microgels that degrade into water-soluble chains under low-frequency ultrasound, or thermosensitive hydrogels that disintegrate under mild FUS hyperthermia to release ultrasmall 1-to-5-nanometer doxorubicin-loaded secondary nanoparticles deep into tumor microvasculature.

Not all FUS-nanoparticle interactions require covalent bond rupture, however. The review details reversible physical mechanisms in which ultrasound controls membrane properties, aggregation states, or spatial distribution without permanent chemical modification. Thermosensitive liposomal bilayers tuned with DPPC/DSPC/MSPC compositions remain stable at 37 degrees Celsius but generate transient membrane defects under mild FUS hyperthermia, accelerating release of encapsulated carboplatin and membrane-associated SN-38. In a triple-negative breast cancer model, FUS-triggered doxorubicin liposomes increased vascular permeability, promoted immunogenic cell death, and reprogrammed a suppressive tumor microenvironment into an immune-responsive one that enhanced checkpoint blockade efficacy. Piezoelectric barium titanate nanoparticles activated by FUS generate reactive oxygen species or trigger nitric oxide release, altering stromal components such as collagen and fibronectin—demonstrating that nanoparticles can actively reshape the tumor microenvironment in concert with acoustic stimulation.

The third and perhaps most forward-looking pillar of the review is its argument for advanced in vitro testing platforms. The authors note that the National Institutes of Health has recently shifted research priorities toward human-based technologies, establishing the Office of Research Innovation, Validation, and Application to reduce reliance on animal models, which frequently fail to translate—many candidate therapies fail in phase I and II clinical trials despite promising rodent results, owing to fundamental interspecies differences in metabolism, molecular interactions, and disease progression. Conventional two-dimensional cell culture fares no better: flat, rigid substrates cannot capture the three-dimensional multicellular architecture of tumors, the mechanical cues of extracellular matrix stiffness, or the cell-cell interactions—including bidirectional mitochondrial transfer between cancer and immune cells—that regulate therapeutic response. Intermediate systems such as Transwell chambers, 3D hydrogel cultures, and tumor organoids each address parts of this gap, but they remain limited in their ability to support controlled perfusion and spatiotemporal regulation.

Cancer-on-a-chip platforms close this remaining gap. These microfluidic systems integrate self-assembled vascular networks, defined extracellular matrix structures, and regulated flow within optically accessible formats, enabling real-time, quantitative analysis of nanoparticle penetration, distribution, and release under physiologically controlled conditions. The review describes how vascularized chip models—including glioblastoma-on-a-chip systems—allow assessment of nanodrug formulations designed to preserve vascular integrity during FUS exposure, while stiffness-tunable hydrogel microfluidic systems reveal how matrix mechanics regulate cancer cell migration and invasion. When FUS is incorporated directly into these chips, researchers can resolve in real time how acoustic stimulation, nanoparticle activation, and tumor-vascular-immune interactions couple together—effects that static culture systems average away and that xenograft models obscure. One cited study integrated FUS with microbubble oscillation in an organ-on-chip model to disrupt the extracellular matrix and enhance interstitial drug transport, while other work showed FUS activating microglia, hinting at immune modulation possibilities in brain tumors.

The authors are candid about the challenges that remain. The effective and safe ultrasound dose range for combined FUS-nanoparticle therapy, as well as repeated dosing strategies, is still unclear, and complex multicomponent formulations need standardization for large-scale production, quality control, and regulatory approval. Yet the trajectory is clear: next-generation cancer-on-a-chip platforms that reconstruct vascular perfusion, matrix mechanics, immune infiltration, and a tunable field for FUS stimulation—ideally built from heterogeneous patient samples—could serve as a translational bridge from nanomedicine design to clinical implementation, ultimately enabling personalized assessment of FUS-responsive therapies. If that bridge is crossed, the humble sound wave, working in concert with engineered nanoparticles, could become one of the most versatile tools in oncology: a knife-less surgeon, a courier for drugs, and a reprogrammer of the tumor microenvironment, all in one focused beam.

The appeal of mEVs begins with their membrane. Cholesterol, sphingomyelin and ceramides help create a relatively robust lipid bilayer that shields internal cargo from environmental damage. Studies summarized in the review indicate that milk vesicles can remain intact under simulated gastrointestinal conditions and protect RNA from digestive enzymes such as ribonucleases. After oral administration, vesicles may interact with intestinal epithelial cells through endocytosis, membrane fusion or receptor-mediated uptake, allowing their contents to enter recipient cells. Some experimental work also suggests that milk vesicles or their cargo can reach tissues beyond the gut, including the brain. This possibility is particularly important because the blood-brain barrier restricts many therapeutic molecules. However, crossing that barrier in animal or cellular models does not establish effective delivery in people. Biodistribution depends on vesicle size, surface proteins, cargo, dose, species of origin, processing conditions and administration route. The review therefore treats gastrointestinal stability and barrier transport as promising properties, not guarantees of therapeutic performance.

Milk vesicles are not a single uniform material. The broader extracellular-vesicle population includes exosomes, microvesicles and apoptotic bodies, which differ in size and how they form. Exosomes develop inside multivesicular bodies when endosomal membranes bud inward to create intraluminal vesicles; multivesicular bodies can then fuse with the plasma membrane and release them. Microvesicles form by outward budding of the cell surface, involving calcium-dependent cytoskeletal changes and redistribution of membrane lipids. In milk, the vesicle population is shaped by the animal species, lactation stage, maternal physiology, diet and health status. Bovine, human, goat, camel, porcine and equine milk can therefore contain different mixtures of proteins and regulatory RNAs. Commonly measured surface or intracellular markers include CD9, CD63, CD81, TSG101 and Alix, but marker detection alone does not define purity or biological function. The review calls for multi-method characterization that combines particle sizing, microscopy, protein analysis, RNA profiling and functional testing.

Obtaining clean vesicles from milk is technically difficult because milk is a complex mixture containing casein micelles, soluble proteins, fat globules and other particles with overlapping physical properties. Differential ultracentrifugation remains widely used, separating material according to size and density, but it can be slow and may promote aggregation or structural damage. Ultrafiltration and polyethylene-glycol precipitation are easier to scale, yet they can recover non-vesicular contaminants. Size-exclusion chromatography separates particles by hydrodynamic size and is often combined with other methods to improve purity. Immunoaffinity capture can select vesicles carrying particular surface markers, although it may reduce recovery and exclude biologically relevant subpopulations. Emerging microfluidic and immunomagnetic systems could enable automated processing of small volumes, while tangential-flow filtration combined with chromatography offers a possible route toward larger-scale production. Across all approaches, researchers need consistent measurements of particle number, size distribution, morphology, membrane integrity, cargo and contaminating milk proteins.

As delivery systems, mEVs could carry molecules that otherwise degrade quickly or dissolve poorly. The review describes experimental loading with polyphenols such as curcumin, resveratrol, quercetin and epigallocatechin gallate, as well as chemotherapeutic compounds including paclitaxel and doxorubicin. Encapsulation may improve aqueous dispersal, protect cargo during digestion and increase contact with intestinal tissues. Vesicles have also been investigated for transporting small interfering RNA, microRNA, messenger RNA, peptides and proteins. In principle, this creates a dual-purpose platform: the vesicle’s native cargo may influence recipient cells, while an added therapeutic molecule supplies a designed activity. Surface engineering could further attach targeting ligands or alter tissue distribution. Yet loading is not straightforward. Passive incubation, membrane permeabilization and other approaches can produce different encapsulation efficiencies and may damage the vesicle. A useful product would require reproducible cargo content, predictable release kinetics and evidence that the loaded molecule reaches the intended tissue at a clinically meaningful dose.

The biological effects reported across models are broad but uneven. In intestinal systems, milk vesicles have been associated with stronger tight junctions, including proteins such as ZO-1, occludin and claudin-1, and with reduced inflammatory signaling. MicroRNAs including miR-148a, miR-21, miR-30a and miR-146b may influence pathways involving NF-κB, Toll-like receptors, DNA methylation and the NLRP3 inflammasome. In cell and animal studies, these mechanisms have been linked to lower inflammatory cytokines, improved barrier function and protection from oxidative stress. Other experiments report effects on macrophage polarization, with some mEV preparations encouraging an anti-inflammatory state. Researchers have also examined bone, liver, heart, lung and pancreatic applications. Milk vesicles have been tested in models of colitis, metabolic dysfunction, fibrosis, osteoporosis, vascular injury and pulmonary inflammation. These findings suggest multiple possible mechanisms, including direct cargo transfer, modulation of gut microbiota and communication along the gut-liver or gut-heart axes, but they do not demonstrate that drinking milk or consuming an unstandardized vesicle preparation treats disease.

Several findings illustrate why careful interpretation is essential. In mice, orally administered milk vesicles have been reported to cross the blood-brain barrier, increase hippocampal dendritic complexity and improve selected cognitive or motor outcomes. Other studies have found changes in microglial DNA-methylation machinery or neuronal survival in cellular models. At the same time, neurological results vary with dose, species, metabolic context and experimental design. A separate line of research has raised hypotheses about interactions between milk exosomes and excessive galactose exposure, while other work found that aging had stronger effects on rat brain lipid profiles and cognition than an extracellular-vesicle-rich supplement. Lung studies likewise include contrasting observations: some mEV preparations protect epithelial barriers or deliver anti-fibrotic compounds, whereas bovine vesicles increased inflammatory macrophage polarization in mice exposed to agricultural dust. In cancer research, vesicles have delivered drugs and gene regulators to tumour models, but one study reported that oral bovine milk vesicles slowed primary-tumour growth while accelerating metastasis. Such results make clear that mEVs are biologically active, context-dependent materials rather than universally beneficial particles.

Translation will depend on proving safety and manufacturing consistency as much as on demonstrating biological activity. Milk origin does not automatically eliminate risk. Preparations can retain caseins, beta-lactoglobulin and other proteins that may trigger reactions in people with milk allergy. Repeated exposure also requires assessment of immune activation, liver and kidney function, oxidative stress, tissue distribution and delayed toxicity. Thermal processing can reduce vesicle yield or disrupt structure, complicating the relationship between fresh milk, pasteurized products and purified formulations. Regulators will also need to determine whether a given product is a food, supplement, biologic, nanomedicine or combination product, since each category carries different requirements. The review points toward standardized isolation protocols, validated vesicle markers, sensitive contaminant testing, single-vesicle and multi-omic analysis, stable storage methods and Good Manufacturing Practice production. Human pharmacokinetic and clinical studies will be decisive. For now, milk-derived extracellular vesicles represent a promising natural nanocarrier platform whose future rests on converting intriguing laboratory observations into reproducible, well-controlled and demonstrably safe interventions.

A further advantage of mEVs is that their value may extend beyond their role as passive containers. Their membranes carry naturally occurring adhesion molecules, tetraspanins and transport-related components that can influence how vesicles are recognized, internalized and distributed. This biological interface distinguishes them from liposomes and polymeric nanoparticles, whose composition can be tuned with considerable precision but generally requires deliberate surface engineering to achieve comparable interactions with cells. The contrast is not absolute: synthetic systems offer stronger control over particle uniformity, drug loading and release kinetics, while mEVs offer a more physiologically integrated membrane and a potentially favorable safety profile. Hybrid designs that combine EV membranes with synthetic cores therefore represent one strategy for balancing biological compatibility with manufacturing control.

The native cargo also complicates how mEV products should be designed and evaluated. A preparation intended to deliver an added drug or RNA may simultaneously contain endogenous proteins, lipids, microRNAs and metabolites capable of altering immune or metabolic responses. Those constituents could contribute to efficacy, but they could also vary with animal species, lactation conditions, feed, health status and processing history. Consequently, measuring total particle concentration is insufficient for comparing products. Functional potency assays will need to establish whether a defined preparation produces a reproducible cellular response, while molecular profiling can help identify which cargo components are retained, enriched or lost during isolation and loading. This is especially important when the desired activity depends on cooperation between the vesicle membrane and its internal contents rather than on a single therapeutic molecule.

Manufacturing scale is promising but should not be confused with readiness for routine clinical use. Milk provides a comparatively accessible starting material, and the review describes ultracentrifugation, size-exclusion chromatography and precipitation methods as established approaches, with tangential-flow and related technologies offering routes toward process intensification. At larger scale, however, purification must preserve membrane integrity while removing abundant non-vesicular milk constituents and maintaining consistent particle and cargo characteristics. The field has reached early translational milestones, including a reported first clinical trial involving mEV-based formulations for RNA therapeutics and anticancer agents, but such studies are only an initial test of feasibility. Results from carefully controlled human investigations will need to define dose, absorption, biodistribution, immune effects and clinically meaningful benefit before the farm-to-pharmacy concept can support approved interventions.

Subject of Research: Milk-derived extracellular vesicles as nutritional and therapeutic nanocarriers

Article Title: Milk-derived extracellular vesicles: nutritional significance, nano-delivery potential, and emerging therapeutic applications – an updated review

Article References: Wang, S., Shaukat, A., Al-Rasheed, M., Tareen, A. M., Arain, M. A., & Luo, C. (2026). Milk-derived extracellular vesicles: nutritional significance, nano-delivery potential, and emerging therapeutic applications – an updated review. Food Science of Animal Resources, 46(1), Article 96. https://doi.org/10.1007/s44463-026-00104-6

Image Credits: AI Generated

DOI: 10.1007/s44463-026-00104-6

Keywords: extracellular vesicles, milk nanocarriers, nutraceutical delivery, drug delivery, microRNAs, precision medicine, gut health, nanomedicine, Milk-derived, extracellular, vesicles, nutritional

Cell membranes are extraordinarily effective barriers. Their lipid bilayers protect the contents of a cell while allowing only selected molecules to enter or leave through channels, receptors, and transport proteins. For researchers trying to deliver proteins, nucleic acids, drugs, or molecular sensors into cells, this barrier is both essential and frustrating. Many delivery systems rely on passive diffusion, membrane-disrupting chemicals, viral vectors, or endocytosis, in which a cell engulfs an external object inside a vesicle. These methods can be inefficient, difficult to control, or potentially damaging. A device that could actively guide a defined cargo across the membrane, rather than simply attach to the cell or become trapped inside an endosome, would address one of the central problems in molecular delivery.

The nanosyringe described by Ding, Fan, Hao and colleagues is based on DNA origami, a method that turns DNA from a genetic information carrier into a programmable construction material. In DNA origami, a long scaffold strand is folded into a predetermined three-dimensional shape by hundreds of shorter “staple” strands. Each staple binds to specific sections of the scaffold, allowing researchers to design structures with nanometre-scale dimensions and precisely positioned functional elements. The resulting objects are not biological needles in the conventional sense. Instead, they are molecular machines assembled through predictable base pairing, with their shape, flexibility, binding sites, and mechanical movements encoded in the DNA sequence.

The syringe concept adds a functional architecture to that molecular framework. A conventional syringe has a barrel that holds cargo, a needle that reaches a target, and a plunger that applies force. At the nanoscale, those components must be recreated through molecular geometry and interactions rather than metal, plastic, or mechanical seals. A DNA origami structure can provide a hollow compartment or channel, while programmable strands and structural elements can act as gates, hinges, anchors, or moving parts. By engineering these features into one object, the researchers aim to transform membrane crossing from an uncontrolled physical event into a sequence of coordinated molecular steps.

The central challenge is not merely touching a membrane, but crossing it in a directed and useful way. A DNA nanostructure approaching a cell encounters a chemically complex surface covered with proteins, carbohydrates, and charged molecules. The membrane itself is a dynamic two-dimensional fluid, and its interior is hydrophobic, making it energetically difficult for a water-loving DNA object to pass through. A successful nanosyringe therefore needs mechanisms that help it bind to the correct surface, orient itself, interact with the lipid bilayer, and create or exploit a transient pathway. The device’s programmability is important because these stages can, in principle, be adjusted independently through changes to its DNA sequence and attached molecular components.

This strategy differs from many nanocarriers that deliver material by entering cells through vesicles. Endocytic uptake can bring a particle into the cell, but the cargo may remain enclosed in an endosome and later be degraded or recycled. Direct membrane translocation seeks to bypass that route by establishing a more immediate connection between the external environment and the cell interior. Such a pathway would be valuable for cargoes that must reach the cytoplasm, where many therapeutic and synthetic-biology functions take place. It could also allow researchers to study membrane transport under controlled conditions, separating the effects of targeting, membrane penetration, and cargo release instead of treating delivery as a single, poorly defined step.

The programmable nature of the system is what makes the work especially significant. DNA nanotechnology allows binding sequences to be selected for particular molecular targets, potentially enabling a nanosyringe to distinguish between different membrane environments. It may also permit control over when the structure opens, closes, docks, or releases its cargo. In principle, such instructions could be triggered by complementary DNA or RNA strands, changes in molecular concentration, environmental conditions, or interactions with a chosen receptor. This does not mean that a universal cell-penetrating machine has already been created. Rather, the work points toward a modular platform in which targeting and translocation behaviours can be redesigned without rebuilding the entire device from scratch.

The technology could eventually have applications in precision medicine, although substantial barriers remain before any clinical use. A practical delivery system would need to function in the complex fluids of the body, avoid rapid degradation, reach the intended tissue, distinguish diseased cells from healthy ones, and release its cargo at the correct location and dose. DNA nanostructures can be vulnerable to nucleases, immune recognition, aggregation, and clearance. Their performance may also change dramatically in the presence of serum proteins and crowded biological surroundings. Scaling up production with consistent quality would present another challenge. These issues are familiar across the field of DNA nanotechnology, and the nanosyringe concept does not eliminate them; it offers a new architecture in which they can be investigated systematically.

Beyond medicine, the device could become a tool for building more sophisticated artificial cells and molecular robots. Researchers have long sought ways to make synthetic compartments exchange materials with their surroundings in a controlled fashion. A programmable nanosyringe could contribute to artificial systems that sense their environment, import selected molecules, and respond through predefined chemical circuits. It could also help scientists explore how membrane permeability, mechanical force, and molecular recognition operate together. Because DNA origami structures can be imaged, modified, and produced with sequence-level precision, they provide an experimental platform for testing design principles that are difficult to isolate in natural membrane proteins.

The study’s broader message is that DNA nanotechnology is moving beyond static shapes and passive carriers toward active devices with defined tasks. A DNA origami nanosyringe does not simply provide a container for cargo; it is conceived as a coordinated molecular instrument that can engage a membrane and direct translocation. Whether that vision leads to reliable therapeutic delivery will depend on future experiments addressing efficiency, selectivity, safety, and operation in living organisms. For now, the work demonstrates how programmable molecular construction can be paired with one of biology’s most demanding engineering problems. By turning a familiar macroscopic tool into a nanoscale DNA machine, the researchers have offered a striking example of how synthetic structures may one day perform targeted operations inside living systems.

Subject of Research: Programmable DNA origami nanosyringe for directed membrane translocation and molecular cargo delivery

Article Title: A programmable DNA origami nanosyringe for directed membrane translocation

Article References: Ding, L., Fan, S., Hao, X. et al. A programmable DNA origami nanosyringe for directed membrane translocation. Nat. Nanotechnol. (2026). https://doi.org/10.1038/s41565-026-02249-3

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41565-026-02249-3

Keywords: DNA origami, nanotechnology, nanosyringe, membrane translocation, molecular machines, targeted delivery, synthetic biology, nanomedicine