Optimization of adeno-associated viral vector-mediated transduction of the corticospinal tract: comparison of four promoters

The corticospinal tract (CST) is the principal descending motor pathway controlling limb and trunk movements, and damage to this pathway causes paralysis. Because adult corticospinal neurons have poor intrinsic regenerative capacity, gene therapy is a promising approach to enhance axon regeneration and functional recovery. Adeno-associated viral (AAV) vectors are widely used for CNS gene delivery, with serotype and promoter choice critically determining transduction efficiency, cell-type specificity, and expression strength. Prior work identified AAV1 as highly efficient for transducing layer V corticospinal neurons after cortical injection, but the optimal promoter for strong and selective expression in these neurons had not been defined through a side-by-side test. This study aimed to identify a compact promoter that drives robust, neuron-preferential expression in the CST, directly comparing sCAG, hCMV, mPGK, and hSYN promoters packaged in AAV1 and delivered to the rodent sensorimotor cortex.

Previous comparative serotype studies showed AAV1 (and AAV5 to a lesser extent) efficiently transduces layer V cortical neurons following direct cortical injection in rodents. AAV transduction efficiency depends on capsid–receptor interactions, endosomal escape, nuclear trafficking, and transcriptional control. Promoters fundamentally shape transgene expression magnitude and cellular specificity, and promoter size matters due to AAV’s ~4.7 kb packaging limit. While promoters such as CMV, CAG, PGK, and synapsin are commonly used in the brain, a direct, side-by-side promoter comparison specifically targeting corticospinal neurons had not been performed, motivating the present study.

Study design: Side-by-side comparison of four compact promoters (sCAG, hCMV, mPGK, hSYN) driving eGFP in AAV1 to optimize transduction of corticospinal neurons in rats and mice. Transgene expression level and cellular specificity were assessed 6 weeks post-injection. Plasmid constructs: AAV-sCAG-eGFP (868 bp sCAG), AAV-hCMV-eGFP (752 bp hCMV upstream of ~400 bp β-globin intron), AAV-mPGK-eGFP (551 bp mPGK upstream of ~400 bp β-globin intron), AAV-hSYN-eGFP (499 bp hSYN; WPRE removed from pTRUF20B-SEW). All constructs contained ITRs and eGFP reporter; polyA not shown in schematic. In vitro transfection: Primary cortical neurons from E18 Sprague–Dawley rats were cultured on poly-D-lysine-coated glass dishes in neural Q basal medium with GS21 and GlutaMAX at 37 °C, 7% CO2. At 10 DIV, plasmids (7 µg/dish) were transfected using oscillating nanomagnetic transfection (NeuroMag) over a 2 Hz, 0.2 mm amplitude magnetic array for 30 min. Cells were fixed at 14 DIV with 4% PFA and mounted for imaging. Quantification with SynD (MATLAB) measured neurite and soma eGFP fluorescence, soma area, and Sholl analysis of dendritic branching. AAV production and titration: Recombinant AAV1 vectors were produced at small scale per established protocols. Genomic titers were quantified by qPCR and matched to 1.42 × 10^12 gc/ml for in vivo injections. Animals and injections: Adult female Lister Hooded rats (n=20) and C57BL/6 mice (n=20) were group-housed with ad libitum food/water. Stereotaxic injections of titer-matched AAV1 vectors were made into the sensorimotor cortex to transduce CST neurons. Analyses were performed 6 weeks post-injection. Histology and imaging: Brain and spinal cord sections underwent immunostaining for neuronal (NeuN), astrocytic, microglial, oligodendrocyte lineage markers, and WFA for perineuronal nets (PNNs). Fluorescence imaging used Leica DM6000 B microscope for quantification; representative images acquired with Leica SP5/TCS SPE confocal systems. NeuN-stained slides were scanned on Zeiss AxioScan Z1. Quantification: Transduction area per section was manually delineated in ImageJ. Transduced cortical neurons (eGFP+ NeuN+) and mean per-neuron eGFP intensity were quantified with a custom ImageJ macro, using background thresholds from non-transduced cells and intra-region measurements. Counts were scaled based on section series to estimate total transduced neurons across ~480 µm AP span. Validation against manual counts showed no significant difference. dCST axons at cervical spinal cord were counted using an eyepiece reticle in three grids (80×80 µm each) and scaled (×10 rats; ×4 mice) to estimate total transduced axons; axonal eGFP intensity measured from ten axons minus background. PNN-bearing interneurons co-expressing eGFP were manually counted.

- AAV1 with hCMV and sCAG promoters yielded broad transgene expression across neurons, astrocytes, and oligodendrocytes in cortex. - mPGK and hSYN promoters produced the strongest overall eGFP expression levels. - mPGK drove expression in cortical neurons and oligodendrocytes (not neuron-specific). - hSYN drove neuron-specific expression, including perineuronal net-expressing interneurons and layer V corticospinal neurons. - The results provide promoter-specific trade-offs between expression strength and cell-type specificity for CST-targeted gene delivery in rodents.

The study addressed the critical question of which compact promoter best balances strength and neuronal specificity for transducing corticospinal neurons using AAV1. Findings show that while hCMV and sCAG enable broad, multi-lineage expression (useful for glial targeting), mPGK and hSYN achieve higher expression, with hSYN conferring neuron-specificity that includes key CST projection neurons and PNN-bearing interneurons. Thus, for CST-focused interventions where neuronal specificity is essential (e.g., manipulating intrinsic growth programs without glial confounds), hSYN is advantageous. Conversely, mPGK may be preferred when strong expression in both neurons and oligodendrocytes is desired. These insights refine vector design for CNS applications, particularly spinal cord injury research where effective and selective modulation of corticospinal neurons is a central objective.

This side-by-side comparison identifies hSYN as a compact promoter that delivers strong, neuron-specific expression in the corticospinal system with AAV1, while mPGK provides strong but broader neuronal/oligodendroglial expression, and hCMV/sCAG support multi-lineage transduction. The work guides promoter selection for CST-targeted gene therapies and basic studies. Future directions include evaluating additional promoters/enhancers, testing other AAV capsids and delivery routes, assessing long-term stability and potential silencing, quantifying functional outcomes after SCI, and expanding cell-type resolution with single-cell transcriptomic and spatial analyses.