Rabies virus remains one of the most lethal human pathogens, and once its neurological symptoms appear, no effective treatment exists. A new study now reports that a gene-silencing approach delivered by a viral vector can sharply suppress rabies virus replication in laboratory cells and substantially improve survival in infected mice, both when administered before infection and when given after the virus has already entered the body.
The research, published as a short communication in the journal Animal Diseases, focused on recombinant adeno-associated virus serotype 2, or rAAV2, as a delivery vehicle for short interfering RNAs, commonly called siRNAs. These small RNA molecules can silence specific genes by binding to their messenger RNA transcripts and triggering their degradation. The team designed siRNAs targeting two rabies virus genes: the G gene, which encodes the glycoprotein responsible for neuronal entry and transport, and the L gene, which encodes the RNA polymerase catalytic subunit essential for viral transcription and replication.
The researchers first screened six candidate siRNAs, three against each target gene, in mouse neuroblastoma cells and baby hamster kidney cells infected with the rabies virus CVS-11 strain. Direct immunofluorescence assays showed that four of the candidates, si-G193, si-G672, si-G997, and si-L5055, significantly inhibited viral replication, with a marked reduction in fluorescent spots per field of view. Viral titer measurements confirmed that si-G997 and si-L5055 produced significantly lower titers than the negative control. Quantitative reverse transcription PCR revealed that all siRNAs targeting the G gene effectively reduced G messenger RNA levels in both cell types, while si-L5055 significantly decreased L messenger RNA levels.
To verify that these effects reflected true gene knockdown rather than indirect antiviral activity, the team cotransfected human embryonic kidney cells with either si-G997 or si-L5055 together with plasmids expressing tagged rabies virus G or L proteins. Both siRNAs significantly reduced the expression of their respective target proteins compared with a scrambled negative control. Under infection conditions, the protein levels of the viral N and G proteins were reduced by both siRNAs, and the G protein was suppressed more markedly than the N protein. The authors attributed the poor performance of the other four siRNAs to differences in the thermodynamic properties of the target sequences and target site accessibility.
Chemically synthesized siRNAs, although rapid in silencing target genes, are prone to degradation in the body and act only briefly. Viral vectors such as adenovirus, retrovirus, and lentivirus can deliver nucleic acids but often provoke strong host immune responses that generate neutralizing antibodies and weaken therapeutic effects. The researchers therefore turned to adeno-associated virus, which they describe as a relatively safe delivery system that supports long-term episomal expression in most tissues. Serotype 2 was chosen because most AAV vectors show predominant tropism for liver and muscle, whereas serotype 2 specifically targets the central nervous system, a property that fits the requirements of rabies virus research.
The team encapsulated siRNA expression cassettes, each containing a CMV enhancer, a U6 promoter, and a shRNA coding sequence, into the rAAV2 vector, producing the constructs rAAV-G997 and rAAV-L5055. In HEK293T cells, these vectors significantly reduced the expression of the G protein and L protein, respectively. A CCK-8 assay showed that neither vector affected the viability of BHK-21 cells. When BHK-21 cells were transduced with the vectors 24 hours before rabies virus infection, both N and G protein levels decreased compared with a negative control vector, with the reduction in G expression more pronounced. In neuroblastoma cells transduced with rAAV-G997, G messenger RNA levels and viral titers remained consistently lower than in the control group at all time points examined, and rAAV-L5055 similarly reduced L messenger RNA levels and viral titers at every time point.
The in vivo experiments used six-week-old Kunming mice divided into groups of ten. Three groups received intracerebral injections of rAAV2 at a dose of 2 × 10¹¹ vector genomes per kilogram, followed 48 hours later by an intramuscular challenge with 1 × 10⁵ focus-forming units of the CVS-11 strain. Among mice injected with rAAV-G997, rAAV-L5055, or a negative control vector before infection, the rAAV-G997 group maintained relatively stable body weight throughout a 21-day observation period, while survival rates reached 90 percent for rAAV-G997, 90 percent for rAAV-L5055, and 50 percent for the control.
In a postexposure setting, mice received the same vectors 24 hours after viral challenge. The rAAV-G997 and rAAV-L5055 groups again showed relatively stable body weight changes compared with the control group, and survival rates were 88.9 percent for both treatment groups versus 55.6 percent for the control. A separate confirmatory experiment using 18 additional mice, six per group, measured viral loads in brain tissue on day 7 after infection. Both pretreatment groups exhibited significantly decreased viral loads in brain tissue compared with the negative control. The authors note that the lack of significant difference in efficacy between the two vectors suggests the strategy has a broad therapeutic window.
The protective effect is attributed to two factors. Rabies virus primarily infects the central nervous system, and AAV2 exhibits natural tropism for neurons, enabling efficient transduction of the primary sites of viral replication when delivered directly to the brain. Intracerebral injection also bypasses the blood-brain barrier, delivering the therapeutic vector straight to the brain parenchyma. Consequently, even 24 hours after viral challenge, the vector-delivered siRNAs effectively suppressed viral replication within the central nervous system, an intervention that directly targets the intracellular viral life cycle and offers a prolonged therapeutic window that the authors say surpasses that of conventional biologics.
The translational path faces considerable hurdles. The evaluations indicated that a relatively high viral dose is required for efficacy, which increases production costs. Owing to time constraints, the study was unable to assess the purity of the packaged rAAV2, and product-related impurities such as empty capsids may cause batch-to-batch variability and potentially exacerbate host immune responses. The authors conclude that rAAV2-mediated delivery of siRNAs targeting the rabies virus G and L genes can inhibit viral replication in vitro and provide significant protection in vivo, highlighting the potential of this platform as a therapeutic approach for rabies while acknowledging that addressing manufacturing and purity issues remains critical for any future clinical application.
Subject of Research: RNA interference delivered by recombinant adeno-associated virus serotype 2 as an experimental antiviral strategy against rabies virus replication in cells and mice.
Article Title: Recombinant adeno-associated virus serotype 2-mediated RNA interference inhibits rabies virus replication
Article References: Wang, Y., Yu, Y., Huang, R., Zhao, S., Peng, Y., Tan, P., He, H., Zhang, Y., Zhang, M., Wang, Y., Luo, Y., Guo, X., & Luo, J. (2026). Recombinant adeno-associated virus serotype 2-mediated RNA interference inhibits rabies virus replication. Animal Diseases, 6(1), Article 42. https://doi.org/10.1186/s44149-026-00253-7
Image Credits: AI Generated
DOI: 10.1186/s44149-026-00253-7
Keywords: rabies virus, RNA interference, siRNA, recombinant adeno-associated virus, rAAV2, gene silencing, postexposure prophylaxis, neurotropic virus, antiviral therapy, mouse model, CVS-11 strain, viral vector delivery
News Source: Juliet Wilcox. (October 30, 2026). Gene-Silencing Vector Shows Strong Protection Against Rabies in Mice. Scienmag.
