A new study in Nature Neuroscience reports that a clinical expression cassette designed to restore production of survival motor neuron, or SMN, protein does not produce detectable sensorimotor toxicity when delivered using self-complementary adeno-associated virus serotype 9, commonly known as scAAV9. The findings address a central safety question in gene therapy for spinal muscular atrophy: whether raising SMN levels throughout the nervous system could itself disrupt movement or sensory function. Balch, Blatnik, Kray and colleagues examined the consequences of using the same broad molecular strategy that has helped make SMN replacement a clinical reality, focusing specifically on the possibility that treatment-related overexpression might impair motor performance or other neurological functions.

Spinal muscular atrophy is caused by damaging variants in the SMN1 gene, which normally provides instructions for the SMN protein. This protein is required for the assembly of spliceosomal complexes that process messenger RNA, a basic cellular operation essential to the survival and maintenance of motor neurons. When SMN levels fall, the largest motor neurons in the spinal cord become especially vulnerable, leading to progressive muscle weakness, loss of movement and, in severe cases, respiratory failure. Gene-replacement therapies attempt to correct the underlying deficiency by delivering a functional SMN sequence to cells using a viral vector. Because mature motor neurons do not readily divide, a vector capable of entering long-lived neurons and sustaining expression is particularly valuable.

AAV9 has become one of the most important delivery vehicles in neurological gene therapy because it can cross biological barriers and reach tissues in the central and peripheral nervous systems. The vector is engineered so that it cannot replicate as a wild virus. Instead, it carries a therapeutic DNA cassette containing the gene of interest and regulatory elements that control its expression. In a self-complementary AAV, the vector genome is configured to form double-stranded DNA more rapidly after entering a cell. This can shorten the delay before therapeutic expression begins, although the design also reduces the amount of genetic material the vector can carry. The compact architecture makes scAAV9 an attractive platform for small therapeutic cassettes such as those used to express SMN.

The safety concern explored in the study arises from the same feature that makes SMN replacement effective: the treatment is intended to raise protein production in many cells, rather than only in a narrowly defined population of motor neurons. SMN is a widely required protein, and its expression must be carefully controlled. Too little protein causes disease, but the consequences of sustained, widespread or excessive expression have remained an important question for therapeutic development. In particular, abnormalities in motor coordination, muscle control, reflexes or sensation could be difficult to distinguish from disease progression unless they are evaluated directly. The researchers therefore centered their analysis on sensorimotor toxicity, an umbrella term covering adverse effects on movement, coordination and sensory-motor integration.

According to the report, animals receiving the clinical SMN expression cassette in scAAV9 did not show evidence of the sensorimotor deficits that would be expected if the treatment were damaging neural circuits or interfering with neuromuscular function. The conclusion is significant because it concerns a clinically relevant cassette rather than an abstract laboratory construct. Expression cassettes contain more than the coding sequence alone: promoters and other regulatory components determine where, when and how strongly the therapeutic gene is produced. A favorable safety result with the clinical configuration therefore provides information about the actual therapeutic design, including the regulatory elements that are intended to support durable SMN expression after vector delivery.

Sensorimotor toxicity is a particularly important endpoint for SMN gene therapy because the disease itself affects the systems being tested. A treated animal may move differently because of pre-existing weakness, developmental delay or incomplete rescue, while a toxic effect could produce additional impairment. Meaningful assessment requires investigators to examine several dimensions of function, including spontaneous activity, gait, coordination, strength and responses that depend on intact sensory processing. It also requires comparison with appropriate control groups and attention to the timing of any abnormalities. The study’s central message is that, under the conditions tested, the clinical scAAV9-SMN design did not generate a detectable adverse sensorimotor phenotype.

The work also illustrates why vector design cannot be separated from biological safety. AAV9 particles can distribute beyond the original injection site, and systemic administration can expose tissues throughout the body. The amount of vector delivered, the route of administration, the efficiency of cell entry and the activity of the promoter can all influence the final level and distribution of SMN protein. Self-complementary genomes may accelerate expression, but they also impose a strict packaging limit and can alter the kinetics of the therapeutic response. Evaluating the complete product in relevant experimental settings is therefore more informative than studying the SMN protein or the viral capsid in isolation.

The new findings do not mean that every possible use of scAAV9-SMN will have identical safety characteristics. Gene-therapy risk depends on species, age, dose, route of administration, immune status and the distribution of vector genomes among organs. A result showing no sensorimotor toxicity also does not eliminate other categories of risk, such as liver injury, immune reactions, manufacturing impurities or effects that might emerge only after prolonged observation. Nor does it establish that increased SMN expression is harmless in every tissue or at every level. Instead, the study narrows a specific uncertainty: the clinically configured cassette, delivered through scAAV9, did not produce measurable toxicity in the sensorimotor domain examined by the investigators.

That distinction matters for the continued development of treatments for spinal muscular atrophy and potentially for other disorders involving motor neurons. Therapeutic programs must balance biological activity against the risks of broad and persistent gene expression. If a vector restores a deficient protein but causes neurological dysfunction, the treatment window may be too narrow for practical use. Evidence that a clinical SMN cassette can achieve its intended molecular purpose without causing detectable movement or sensory impairment supports the rationale for further investigation, including longer-term monitoring, dose optimization and studies in additional models. It also strengthens confidence in the principle that correcting a severe protein deficiency does not necessarily require compromising neural function through toxic overexpression.

The study arrives as the field moves from demonstrating that gene replacement can work to defining how such therapies should be designed, dosed and monitored over a lifetime. For patients with SMA, durable SMN restoration remains the therapeutic goal, but durability must be accompanied by a detailed understanding of safety. The report by Balch and colleagues contributes to that effort by testing a clinically relevant vector-and-cassette combination against a focused neurological safety question. Its conclusion—that scAAV9 carrying the clinical SMN expression cassette lacks detectable sensorimotor toxicity in the studied setting—offers a reassuring result while leaving the broader task of long-term, multi-organ surveillance firmly in view.

Subject of Research: Safety of a clinical SMN expression cassette delivered by self-complementary AAV9, with emphasis on sensorimotor toxicity and gene therapy for spinal muscular atrophy.

Article Title: Lack of sensorimotor toxicity when using the clinical SMN expression cassette in scAAV9

Article References:

Balch, M.H.H., Blatnik, A.J., Kray, K.M. et al. Lack of sensorimotor toxicity when using the clinical SMN expression cassette in scAAV9. Nat Neurosci (2026). https://doi.org/10.1038/s41593-026-02411-z

Image Credits: AI Generated

DOI: https://doi.org/10.1038/s41593-026-02411-z

Keywords: spinal muscular atrophy, SMN, gene therapy, scAAV9, AAV9, sensorimotor toxicity, motor neurons, viral vectors, neurological safety

Tags: adeno-associated virus in neurological treatmentclinical outcomes of SMN cassettegene therapy safety profilemotor neuron preservationmuscle weakness prevention strategiesneuronal function and gene overexpressionneurotoxicity in gene therapyscAAV9 vector safetysensorimotor toxicity assessmentSMN protein gene deliveryspinal muscular atrophy gene therapyspliceosomal complex and RNA processing