The central obstacle the review addresses is the hyaluronic acid (HA)-rich extracellular matrix (ECM) that surrounds cells in virtually all soft tissues, including the nasal submucosa, the perineural and perivascular compartments, and the brain parenchyma itself. HA is a linear, non-sulfated glycosaminoglycan built from repeating disaccharide units of D-glucuronic acid and N-acetyl-D-glucosamine, synthesized at the plasma membrane by HA synthases and capable of reaching molecular weights in the megadalton range. Because HA can bind up to 1,000 times its own weight in water, it creates a hydrated, gel-like interstitial environment that resists the movement of large particles. For therapeutics the size of AAV capsids, which measure roughly 20 to 25 nanometers in diameter, or for cells and vesicles that are orders of magnitude larger, this matrix imposes viscosity, steric hindrance, and hydrodynamic resistance that limit penetration from any injection site.

Hyaluronidases attack this barrier with biochemical precision. These endoglycosidases cleave the β-1,4-glycosidic bonds between N-acetylglucosamine and glucuronic acid residues, depolymerizing long HA chains into short tetrasaccharides and hexasaccharides. The recombinant human PH20 isoform, known commercially as HYLENEX and approved by the FDA as a subcutaneous delivery adjunct, acts as an endo-β-N-acetylhexosaminidase targeting those same bonds. The enzymatic cut produces several interrelated changes: reduced viscosity and elastic modulus, increased effective pore size within the interstitium, and decreased interstitial fluid pressure in HA-rich tissues. HA fragments generated by the enzyme may also act as danger-associated molecular patterns that activate innate immune signaling through Toll-like receptors and cell-surface receptors such as CD44 and RHAMM, a property the review suggests could potentially synergize with therapeutic mechanisms.

What makes hyaluronidase particularly attractive as a delivery adjunct is the reversibility of its effect. Endogenous HA synthesis by the HAS1, HAS2, and HAS3 enzymes progressively restores the matrix after enzyme exposure. Measurable reductions in HA polymer length and tissue viscosity appear within minutes to hours of local administration, peak permeability effects typically occur within the first one to four hours, and HA content begins returning toward baseline within approximately 15 hours, with near-complete restoration by 24 to 48 hours. This transient permeability window creates a logical dosing strategy: administer the enzyme 30 to 60 minutes before the therapeutic payload so that vectors or cells arrive in the tissue during peak permeability, then allow the matrix to reconstitute itself without lasting structural disruption.

The clinical precedent for this approach is well established in a different context. rHuPH20 is already an approved enabling excipient in several high-volume subcutaneous biologic products, including formulations of rituximab, trastuzumab, and daratumumab, where it transiently modifies the subcutaneous ECM to increase absorption rate, improve bioavailability, and permit larger injection volumes. Decades of clinical use support a generally favorable safety profile at approved doses, with common adverse events limited to mild, self-limited injection-site reactions and rare hypersensitivity events, particularly with older animal-derived preparations. The shift to the recombinant human enzyme has substantially reduced immunogenicity concerns; a comprehensive nonclinical toxicology review found no evidence of organ toxicity, developmental toxicity, or cross-reactive antibody effects, and treatment-emergent anti-rHuPH20 antibodies in clinical products have not been clearly associated with loss of efficacy or increased adverse events.

The foundational preclinical evidence for CNS applications comes from a 2009 study by Danielyan and colleagues, in which fluorescently labeled rat mesenchymal stem cells were administered intranasally to adult mice with or without 100 USP units of hyaluronidase applied 30 minutes beforehand. Without pretreatment, only 584 plus or minus 152 of the 300,000 applied cells arrived in the olfactory bulb within one hour. Hyaluronidase pretreatment significantly increased delivery to the olfactory bulb (p = 0.02) and to the total brain area (p = 0.04), with labeled cells subsequently identified in the thalamus, hippocampus, cortex, and subarachnoid space. Importantly, the delivered cells were predominantly not associated with CD45-positive immunocompetent cells, suggesting they traversed the cribriform plate largely intact rather than being carried as phagocytic cargo. Two migration routes were identified: a parenchymal route via the rostral migratory stream and olfactory bulb, and a cerebrospinal fluid-mediated route along the cortical surface, potentially driven by the perivascular pumping mechanism of arterial pulsation.

More recent work has added a biological targeting layer on top of this physical enhancement. Multilineage-differentiating stress-enduring (MUSE) cells, a pluripotent stem cell subpopulation identified by SSEA-3 expression in bone marrow mesenchymal cells and human umbilical cord MSCs, express high levels of sphingosine-1-phosphate receptor 2 (S1PR2). This receptor mediates active homing toward sites of tissue injury where sphingosine-1-phosphate is released by damaged and inflamed cells. In a Parkinson’s disease mouse model, intranasally administered MUSE cells achieved selective CNS targeting while non-MUSE MSCs showed significant pulmonary accumulation with no detectable head signal. Brain section analysis confirmed MUSE cell presence in the cerebral cortex and substantia nigra, and functionally, treated mice showed higher tyrosine hydroxylase-positive and NeuN-positive cell density in the substantia nigra pars compacta, reduced S1P and TNF-α brain expression with increased BDNF and GDNF, improved motor performance on pole and wire hang tests, and near-normalization of substantia nigra histomorphology. The review argues that combining hyaluronidase with MUSE cells addresses two distinct barriers in a complementary manner: the enzyme removes the physical ECM obstacle while the cells’ S1P-S1PR2 axis provides directional migration, though the authors note this mechanistic synergy remains an expert assessment pending confirmatory human-scale studies.

Small extracellular vesicles (sEVs) represent a third cargo class for this strategy. These lipid bilayer-delimited particles, classified under the MISEV2023 framework by measurable physical and biochemical attributes rather than presumed biogenesis, carry nucleic acids and proteins including miRNAs, siRNAs, and mRNAs encoding therapeutic proteins or gene-editing tools. Quantitative studies demonstrate saturable, transcytosis-mediated transport of EVs across the intact blood-brain barrier in both directions, with flux enhanced by inflammation, and multiple groups have shown that intranasally administered sEVs reach widespread brain regions via perivascular and perineural pathways with therapeutic effects in stroke and neurodegeneration models. Hyaluronidase pretreatment of the nasal mucosa has been reported to further enhance CNS sEV distribution and increase perivascular space accumulation in rodent stroke models, suggesting a natural pairing between the enzyme and vesicle-based gene modulation platforms.

The review is explicit about the translational caveats. Rodent nasal anatomy differs substantially from human anatomy, with a proportionally larger olfactory epithelium that may exaggerate nose-to-brain access relative to humans, meaning quantitative distribution improvements should be read as proof-of-principle rather than direct predictors of human outcomes. Human safety and efficacy data for intranasal AAV with hyaluronidase are entirely absent, and all supporting evidence for the CNS applications remains preclinical. The authors also flag concerns around perineuronal nets, specialized HA-based ECM structures surrounding subsets of neurons that regulate memory consolidation, circuit stability, and protection against excitotoxicity; intentional disruption of these structures carries poorly characterized risks of cognitive or seizure-threshold effects. On the regulatory side, combining hyaluronidase with gene therapy products raises combination-product questions, requiring demonstration that the enzyme does not adversely affect vector potency, purity, or stability, and early engagement with the FDA or EMA is advised.

To structure the path forward, the review assigns applications to three evidence tiers: Tier 1 covers established FDA-approved subcutaneous uses; Tier 2 covers off-label procedural applications supported by case series and extended clinical experience; and Tier 3 covers investigational intranasal AAV, EV, MSC, and MUSE cell CNS delivery, supported only by preclinical rodent data and requiring formal first-in-human evaluation under investigational new drug applications, ethics board approval, and explicit informed consent. The authors propose a detailed investigational intranasal protocol, including 50 to 75 USP units of hyaluronidase per naris injected submucosally 30 minutes before therapeutic administration, with strict volume constraints of 1.0 to 1.5 milliliters per naris to avoid pocket rupture, and discharge instructions for neurological symptom monitoring. Priority research areas include controlled trials comparing intranasal AAV, MUSE cell, and EV-based gene therapies with and without hyaluronidase, human-scale characterization of the S1P gradient, longitudinal immunogenicity tracking, and development of regulatory frameworks specific to hyaluronidase-vector combinations. Used judiciously as a permeability adjunct that complements rather than replaces therapeutic biology, the authors conclude, hyaluronidase offers meaningful practical utility across the spectrum of modern gene and cell therapy delivery challenges, provided its investigational status is respected at every step.

Subject of Research: Hyaluronidase-enhanced intranasal delivery of gene therapy vectors and regenerative cells to the central nervous system

Article Title: Hyaluronidase-enhanced delivery in gene therapy and regenerative medicine for improved vector, cell, and cargo access

Article References: Sewell, P. E., & Jensen, C. (2026). Hyaluronidase-enhanced delivery in gene therapy and regenerative medicine for improved vector, cell, and cargo access. Frontiers in Drug Delivery, 6, Article 1906950. https://doi.org/10.3389/fddev.2026.1906950

Image Credits: AI Generated

DOI: 10.3389/fddev.2026.1906950

Keywords: hyaluronidase, gene therapy, AAV vectors, intranasal delivery, nose-to-brain, mesenchymal stem cells, MUSE cells, extracellular vesicles, hyaluronic acid, extracellular matrix, CNS drug delivery, rHuPH20