The plasmids pAAV-Hb9-LMO3 (sbGLuc-VChR1-eYFP) and pAAV-hSyn-mCherry were synthesized by Addgene, Watertown, MA, USA (Addgene plasmid: 114103, catalog #114472) according to a previously reported specification [17]. A schematic representation of the developed constructs is shown in Fig. 1A. Luminopsin (LMO3) is a fusion protein of Gaussia luciferase variant (sbGLuc), Volvox canalorhodopsin-1 (VChR1) and enhanced yellow fluorescent protein (eYFP) under the regulation of the Homeobox 9 (Hb9) promoter. The plasmid pAAV-CMV-eGFP was provided by Marlin Biotech, Moscow, Russia.

Fig. 1.

Experimental design and targeting strategy for AAV-mediated gene delivery to the spinal cord. (A) Schematic representation of the main components of the gene cassette packaged into the recombinant AAV vector. (B) Intrathecal injection of gene constructs (6 $\times$ 10¹³ vg/mL, 20 µL) into rats: AAV9-CMV-eGFP, n = 6; AAV9-hSyn-mCherry, n = 5; AAV9-Hb9-LMO3-eYFP, n = 5. (C) Visualization of spinal cord segments used for spatial analysis of neurons expressing reporter proteins. (D) Manual mapping of reporter-positive neurons was performed using a mask that includes anatomical regions of the spinal cord, shown here at the L1–L2 segmental level. Scale bar = 100 µm. AAV, adeno-associated virus; CMV, cytomegalovirus; hSyn, human synapsin; Hb9, homeobox 9; GFP, green fluorescent protein.

To produce plasmid DNA of pAAV-Hb9-LMO3 (sbGLuc-VChR1-eYFP), pAAV-hSyn-mCherry, pAAV-CMV-eGFP in preparative quantities, the Escherichia coli strain NEB® Stable (Thermo Fisher Scientific Inc., Waltham, MA, USA) was transformed. For this purpose, bacterial cells were cultured in Luria-Bertani (LS-LB) medium without addition of antibiotics at 37 °C and vigorous shaking (200–300 rpm) until OD600 = 0.30–0.35. In this work, low-salt Luria-Bertani LS-LB medium was used for cultivation of E. coli NEB® Stable cells (per 1 L of deionized water): salt-free yeast extract (Sigma-Aldrich, St. Louis, MI, USA USA)—5 g, tryptone (Sigma-Aldrich, USA)—10 g, 5 M NaCl (Sigma-Aldrich, USA)—5 g, pH 7.5. Agarized LS-LB medium was also used (per 1 L of deionized water): salt-free yeast extract—5 g, tryptone—10 g, 5 M NaCl—5 g, agar (Sigma-Aldrich, USA)—15 g. Sterilization of nutrient media was carried out in autoclave in liquid or agar mode. Competent cells were prepared using CaCl₂ method. Plasmid DNA from bacterial cells was isolated by alkaline lysis using a commercial GeneJET Plasmid MiniprepKit (Thermo Fisher Scientific Inc., USA) according to the methodology recommended by the manufacturer. For this purpose, a single bacterial colony of recombinant Escherichia coli NEB® Stable strain containing the required plasmid DNA was seeded in a flask with 2–5 mL of LS-LB medium in the appropriate selective antibiotic. The culture was incubated in a shaker (300 rpm) at 37 °C for 8 hours. Next, the initial bacterial culture was diluted from 1:500 in 50 mL of LB medium. After seeding, the culture was incubated in a shaker (300 rpm) at 37 °C for 12–16 hours. The culture should reach a cell density of approximately 3–4 $\times$ 10⁹ cells per 1 mL. Bacterial cells were harvested by centrifugation in 50 mL tubes at 6000 g for 15 min at 4 °C. Plasmid DNA was isolated from the obtained bacterial biomass.

SH-SY5Y cells were cultured in DMEM/F-12 medium (PanEco, Moscow, Russia) supplemented with 10% fetal bovine serum (Bioserum, South America), 2 mM L-glutamine (PanEco, Russia) and 1$\times$ penicillin and streptomycin antibiotic mixture (PanEco, Russia). Cells were cultured in an incubator at 37 °C in a humidified atmosphere containing 5% CO₂. The transfection agent linear polyethylenimine (LPEI) (Thermo Fisher Scientific Inc., USA) was used to transfect SH-SY5Y cells with plasmid DNA. For transfection, SH-SY5Y cells were cultured on 6-well culture plates. A cell culture with a monolayer density of 60–70% was used. To 400 µL of serum-free culture medium, 2 µL of LPEI transfection agent was added, and the mixture was thoroughly mixed on a vortex. 2 µg of plasmid DNA—pAAV-Hb9-LMO3 or pAAV-hSyn-mCherry or pAAV-eGFP were added to the mixture, the mixture was thoroughly mixed, precipitated and incubated at room temperature for 15–20 min. Then 400 µL of the transfection mixture was added dropwise to the wells of the culture plate. The plate was gently shaken to achieve uniform distribution of the complexes immediately after the addition of the transfection reagent. The plates were incubated in an incubator at 37 °C in a humidified atmosphere containing 5% CO₂. Transgene expression was analyzed after 24–48 hours. For immunofluorescence analysis, the cells were fixed in 4% buffered formalin, washed with 0.1% Triton X-100 in PBS and counterstained with 4^′,6-Diamidino-2-phenylindole (DAPI) (10 mg/mL in PBS, Sigma, St. Louis, MI, USA) to visualize the nuclei. The results were analyzed using an LSM 780 Confocal Microscope (Carl Zeiss, Jena, Germany).

Human embryonic kidney 293 cells (HEK293T; ATCC CRL-11268, American Type Culture Collection, Manassas, VA, USA) were used in this study. The cells were cultured in high-glucose DMEM (Gibco, Waltham, MA, USA) supplemented with 10% (v/v) fetal bovine serum (FBS; Biosera Europe, France) and Penicillin-Streptomycin (Pen-Strep). All cell lines were maintained at 37 °C in a humidified atmosphere of 5% CO₂ and 95% relative humidity, and were confirmed to be free of mycoplasma contamination by PCR testing. Cell line authentication was performed by the American Type Culture Collection using short tandem repeat (STR) profiling.

Primary motor neurons were isolated from neonatal Wistar rats at postnatal day 3 (P3). Animals were euthanized by decapitation (a detailed description of the procedure is provided in Section 2.9), after which the spinal cord was extracted and transferred into sterile physiological saline supplemented with antibiotics. Tissue was then incubated in 0.25% trypsin solution (PanEco, Russia) containing antibiotics at 37 °C for 20 minutes with gentle trituration every 10 minutes. Following enzymatic digestion, the suspension was centrifuged at 2000 rpm for 10 minutes. The supernatant was discarded, and the pellet was resuspended in calcium- and magnesium-free Dulbecco’s phosphate-buffered saline (DPBS, 450 mL), followed by a second centrifugation under the same conditions. The DPBS wash was repeated twice. After the final centrifugation, cells were resuspended in DMEM/F12 medium without glutamine (PanEco) supplemented with B27 (ncB27 Supplement, serum-free, Shownin, Hefei, China) and EGF (1:4464, Cloud-Clone, catalog #APA560Hu02, USA). Cells were seeded into 6-well plates pre-coated 24 hours in advance with human fibronectin (IMTEK, Moscow, Russia). An initial small volume of the cell suspension was added to allow attachment; after 10–15 minutes of incubation at 37 °C, the final volume of culture medium was added. These preparations were established as de novo primary motor neuron cultures. Cultures were maintained under standard conditions (37 °C, 5% CO₂, 95% relative humidity) and monitored daily. Medium replacement was performed as needed but not earlier than day 3; only 50% of the medium volume was replaced during each change. Culture was confirmed to be free of mycoplasma contamination by PCR testing. On day 7 of culture, cells were transduced with recombinant AAV9 vectors (AAV9-Hb9-LMO3 or AAV9-hSyn-mCherry) at a multiplicity of infection (MOI) of 50,000. Transgene expression was evaluated 48 hours post-transduction. For immunofluorescence analysis, cells were fixed with 4% paraformaldehyde in PBS, permeabilized with 0.1% Triton X-100 in PBS, and stained with DAPI (10 mg/mL, Sigma-Aldrich) to visualize nuclei. Confocal imaging was performed using an LSM 780 microscope (Carl Zeiss, Germany). To verify the cellular identity within the cultures, immunofluorescent staining was performed using antibodies against neurons (NeuN, catalog #DF614, Affinity, Shanghai, China)/$\beta$III-tubulin (Tuj1, catalog #AF700, Affinity, China), and astrocytes (GFAP, catalog #PAA068Ra01, Cloud-Clone Corp.), confirming the neuronal phenotype of the majority of cells; representative images are shown in Supplementary Fig. 1.

Recombinant viral vectors AAV9-Hb9-LMO3, AAV9-hSyn-mCherry, AAV9-eGFP were produced in HEK293T cells by the triple transient transfection method on a total area of 7500 cm². Viral preparations were purified from cellular debris, impurity proteins, and empty viral capsids according to the methodology previously developed in the laboratory [22, 23, 24]. Viral vector preparations were sterilised by filtration using a syringe filter with a pore diameter of 0.22 µm, then frozen in viral preparation storage buffer: 1$\times$PBS/213 mM NaCl/0.001%, Pluronic F-68 in Eppendorf tubes. Aliquots of viral preparations were taken before freezing for quality control. The viral titer was determined through quantitative PCR using primers (forward: 5^′-3^′-GGAACCCCCCTAGTGATGGAGTT and reverse: 5^′-3^′-CGGCCTCAGTGAGCGA) and a probe targeting ITRs (5^′-3^′(FAM) CACTCCCTCTCTCTGCGCGCGCTCG (BBQ)). The titer of viral particles in the final preparation was 3.52 $\times$ 10¹³ GC/mL for AAV9-Hb9-LMO3; 3.36 $\times$ 10¹³ GC/mL for AAV9-hSyn-mCherry; and 3.85 $\times$ 10¹³ GC/mL for AAV9-eGFP.

The study was conducted on 16 female Wistar rats (weight: 200–250 g, 3–4 months; source - KrolInfo LLC, Moscow region, Orekhovo-Zuyevsky city district, Russia). The animals were kept in standard conditions (23 °C $\pm$ 2 room temperature, 12 h:12 h light-dark cycle starting at 8 AM, 60–65% humidity, and free access to water and food ad libitum). The animals were randomly assigned into three groups: AAV9-CMV-eGFP (n = 6), AAV9-hSyn-mCherry (n = 5), AAV9-Hb9-LMO3-eYFP (n = 5). All experiments involving animals were performed in accordance with the guidelines set forth by the European Communities Council Directive 86/609/EEC. The experimental protocols were approved by the Animal Care and Use Committee of the Kazan Federal University (protocol No. 50, dated 26.09.2024).

Prior to the intrathecal injection procedure, the animals were anesthetized with a combination of Zoletil (Virbac Laboratories, Inc., 100 mg/mL, Karros, France) and Xylanite (Nita-Farm, Saratov, Russia). The anesthetic mixture was administered via intramuscular injection at a dose of 40 mg/kg for Zoletil and 10 mg/kg for Xylanite. All rats had the abdomen and lumbosacral area shaved at the start of the injection. The shaved areas were cleaned with a diluted solution of the alcohol-based antiseptic «Chisteya Plus» (LLC «VITA-PUL», Moscow, Russia). Rats were positioned in sternal recumbency with their pelvic limbs brought under the abdomen as cranially as possible in order to arch the lumbosacral area. The intrathecal injection was performed using a Hamilton syringe (75N 20 µL SYR 25G). The intrathecal injection of gene constructs was administered in a 20 µL, 6 $\times$ 10¹³ copies/mL. The injection site was between the last lumbar (L6) and the 1st sacral (S1) vertebra Fig. 1B. The injection was considered successful if one of the 2 following signs were noted: presence of CSF in the needle hub prior to injection and/or twitch of the tail during the injection. If none of these signs were noted, or if blood was visible in the needle hub prior to injection, the needle was withdrawn and another sterile needle was used to repeat the procedure. All intrathecal injections were successful at the 1st or 2nd attempt.

Euthanasia was performed by intramuscular administration of Zoletil (40 mg/kg) and Xylanite (10 mg/kg), followed by transcardial perfusion with cold 0.01 M phosphate-buffered saline (PBS; pH 7.4), followed by 4% paraformaldehyde (PFA) in PBS. For neonatal rat pups (P3), anesthesia was induced with the same drug combination at equivalent doses prior to decapitation.

For histological analysis, spinal cords were extracted and post-fixed in 4% paraformaldehyde in PBS. Tissues were cryoprotected by sequential immersion in 15% and 30% sucrose solutions. After embedding in Tissue-Tek O.C.T. Compound, 20 µm transverse sections were cut using a RWD Minux® FS800A. For each animal, five slides from each analyzed spinal cord segment (cervical C3–4, thoracic Th8–9, lumbar L1–2, and sacral S1), each containing five sections, were selected for analysis. Segment choice was based on their functional relevance: C3–4 for forelimb control, Th8–9 for trunk coordination, L1–2 for the locomotor rhythmogenic kernel, and S1 for autonomic and urinary control [25, 26, 27, 28]. Sections were rinsed in PBS, counterstained with 4^′,6-diamidino-2-phenylindole (DAPI; Sigma-Aldrich), and mounted with medium for confocal microscopy.

Spatial analysis of neurons expressing reporter proteins was conducted in the gray matter of the spinal cord at the level of the studied segments Fig. 1C. Detailed imaging was carried out using an LSM 780 confocal microscope (Carl Zeiss, Germany). Visualization was performed using confocal microscopy with a 20$\times$ magnification. Inverted single-channel images were used for quantitative analysis. The study was limited to the portion of the gray matter located between laminae VII and X Fig. 1D. For manual mapping of neurons, a mask encompassing the anatomical regions of the spinal cord at the L2 segment level was applied. The following structures were used as coordinate landmarks: the lateral dorsal commissural nucleus, intermediolateral nucleus, dorsal nucleus, intercalated nucleus, and the boundary between the gray and white matter.

Spatial analysis of neurons expressing reporter proteins was conducted in the gray matter of the spinal cord at the level of the studied segments. Visualization was performed using confocal microscopy with a 20$\times$ magnification. The study was limited to the portion of the gray matter located between laminae VII and X. For anatomical orientation, reference images from published atlases of the rat spinal cord [29] and comparative neuroanatomical studies [30]. Representative examples of these reference-based orientations, including 10$\times$ and 20$\times$ images of the central canal region. For manual mapping of neurons, a mask encompassing the anatomical regions of the spinal cord at the L2 segment level was applied. The following structures were used as landmarks: the lateral dorsal commissural nucleus (LDCom), intermediolateral nucleus (IML), dorsal nucleus (D), intercalated nucleus (ICL), and the boundary between the gray and white matter. Quantification in laminae VII, VIII, and IX was performed bilaterally, while lamina X, recognized as an asymmetrical region, was analyzed as a single area surrounding the central canal.

Statistical analyses were performed using Origin 10.0 SR0 (OriginLab Corporation, Northampton, MA, USA). Differences among groups were first assessed using the Kruskal–Wallis test, and significant results were further examined by pairwise Mann–Whitney U tests with Bonferroni correction. Significance levels were set at *p 0.05, **p 0.01, and ***p 0.001.

To confirm the integrity and correct structure of the developed plasmid constructs, restriction digestion was performed. Analysis of pAAV-CMV-eGFP, pAAV-hSyn-mCherry, and pAAV-Hb9-LMO3-eYFP revealed restriction fragments of expected sizes, consistent with the theoretical plasmid maps (Fig. 2A). These results confirm the successful cloning and structural accuracy of the designed vectors.

Fig. 2.

Analysis of plasmid DNA integrity and transgene expression in primary motor neuron cultures. (A) Restriction analysis of plasmid DNA and electrophoresis in 0.8% agarose gel. Lanes 1, 3, 5—circular plasmid DNA; lanes 2, 4, 6—plasmid DNA after restriction digestion with SmaI. (B) Transgene expression analysis 48 hours after transduction of primary motor neuron cultures with AAV9-CMV-eGFP, AAV9-hSyn-mCherry or AAV-Hb9-LMO3-eYFP. Scale bar: 100 µm.

To evaluate the functionality of the constructs, primary motor neuron cultures were transduced with AAV9-CMV-eGFP, AAV9-hSyn-mCherry or AAV9-Hb9-LMO3-eYFP vectors. Confocal imaging revealed expression of eGFP, mCherry or eYFP in transduced cells, confirming that both constructs drive effective transgene expression under control of the respective promoters (Fig. 2B). Notably, fluorescence was localized to neuronal pericarion, suggesting successful transduction of motor neuron populations in vitro.

On day 7 after intrathecal administration of recombinant AAV9 vectors, the number of transduced neurons was analyzed in lamina VII at four spinal cord levels: cervical (C3–4), thoracic (Th8–9), lumbar (L1–2), and sacral (S1). No expression of reporter proteins eGFP or eYFP was detected at lamina VII in the AAV9-CMV-eGFP and AAV9-Hb9-LMO3-eYFP groups at any of the examined levels, indicating a lack of tropism of these constructs for neurons in this lamina. In contrast, the AAV9-hSyn-mCherry group exhibited consistent mCherry expression in neurons at all analyzed levels, with notable segmental variability (Fig. 3A,3A’).

Fig. 3.

Selective transduction patterns of AAV vectors in distinct spinal cord lamina. (A) Confocal images of spinal cord sections at the L1–L2 level for experimental groups. Scale bar: 60 µm. (A’) Transduction of neurons in lamina VII was observed following intrathecal injection of AAV-hSyn-mCherry, whereas AAV-CMV-eGFP and AAV-Hb9-LMO3-eYFP showed no tropism for these neurons. (B) Number of transduced neurons in lamina VIII (C) and X at C3–4, Th8–9, L1–L2, and S1 spinal cord levels. Statistical analysis was performed using the Mann–Whitney U test with Bonferroni correction. (B’) Confocal imaging of spinal cord lamina VIII (C’) and X at the L1–L2 level in experimental groups. White arrowheads indicate transduced neurons. Insets with dashed borders show enlarged views of selected regions. Scale bar: 60 µm; inset scale bar: 15 µm.

The highest mean number of transduced neurons was observed at C3–4 (15.88 $\pm$ 4.08). At the S1 level, the number of labeled neurons was 10.20 $\pm$ 1.70, while at Th8–9 and L1–2 the values were lower, amounting to 7.62 $\pm$ 2.74 and 6.24 $\pm$ 6.12, respectively. Despite the presence of transduced neurons at all examined levels, mCherry expression in lamina VII showed pronounced segmental heterogeneity: the average number of labeled cells was lower at the thoracic and lumbar levels compared to the cervical and sacral levels. Statistical analysis revealed a significant reduction in the number of transduced neurons at Th8–9 (p = 0.006) and L1–2 (p = 0.001) compared to C3–4. No statistically significant differences were observed between Th8–9, L1–2, and S1.

Transduction of neurons in lamina VIII was analyzed in the C3–4, Th8–9, L1–2, and S1 spinal cord segments on day 7 after intrathecal administration of recombinant AAV9 vectors (Fig. 3B,3B’). In the AAV9-Hb9-eYFP group, eYFP expression in neurons was minimal (on average, fewer than one cell per section), consistent with the motoneuron specificity of the Hb9 promoter and the absence of motoneurons in this region.

In the groups using a strong ubiquitous promoter (CMV) and a neuron-specific promoter (hSyn), a moderate number of transduced cells were observed. At the L1–2 level, the mean number of eGFP-positive neurons in the AAV9-CMV-eGFP group was 16.46 $\pm$ 4.61, which was approximately 2.8 folds higher than in the AAV9-hSyn-mCherry group (5.83 $\pm$ 2.54; p = 0.00034) and substantially exceeded the value in the AAV9-Hb9-eYFP group (0.71 $\pm$ 1.44; p = 0.00032). The difference between the AAV9-hSyn-mCherry and AAV9-Hb9-eYFP groups was also statistically significant (p = 0.005). A similar pattern was observed at the S1 level: the mean number of transduced neurons was 11.83 $\pm$ 3.62 in the AAV9-CMV-eGFP group, 3.61 $\pm$ 1.61 in the AAV9-hSyn-mCherry group, and 1.35 $\pm$ 1.57 in the AAV9-Hb9-eYFP group. Transduction efficiency with the CMV promoter was significantly higher than with hSyn (p = 0.000072) and Hb9 (p = 0.0093).

At the C3–4 level, a significant difference was found only between the AAV9-hSyn-mCherry and AAV9-Hb9-eYFP groups (p = 0.0000025), with markedly higher expression observed using the hSyn promoter. No statistically significant differences were detected between groups at the Th8–9 level.

Neuronal transduction in lamina X was analyzed in spinal cord segments C3–4, Th8–9, L1–2, and S1 on day 7 after intrathecal administration of recombinant AAV9 vectors (Fig. 3C,3C’). In the AAV9-CMV-eGFP group, eGFP expression in lamina X neurons was observed only at the L1–2 and S1 levels, where the mean number of transduced cells was 12.73 $\pm$ 4.92 and 10.15 $\pm$ 2.27, respectively. In the AAV9-hSyn-mCherry group, moderate transduction was detected in all examined segments, including C3–4 (4.58 $\pm$ 1.68), Th8–9 (4.92 $\pm$ 3.06), L1–2 (7.13 $\pm$ 2.83), and S1 (5.31 $\pm$ 4.35), without pronounced segmental differences. At the L1–2 and S1 levels, the number of transduced neurons in the AAV9-CMV-eGFP group was significantly higher than in the AAV9-hSyn-mCherry group (p = 0.0149 and p = 0.0144, respectively).

In the AAV9-Hb9-eYFP group, no eYFP expression was detected in lamina X neurons at any spinal level, which is consistent with the motoneuron-specific activity of the Hb9 promoter and the absence of such cells in this region.

Transduction of neurons in lamina IX was evaluated at the spinal cord levels C3–4, Th8–9, L1–2, and S1 on day 7 after intrathecal injection of recombinant AAV9 vectors (Fig. 4A,4A’). All three groups demonstrated consistent reporter expression in accordance with the expected localization of motoneurons in this lamina. The average number of transduced neurons ranged from 3.58 $\pm$ 1.44 to 5.92 $\pm$ 1.79, with no statistically significant differences between groups at most segmental levels, except for the lumbar region.

Fig. 4.

Analysis of transduction specificity and distribution of AAV vectors in lamina IX of spinal cord. (A) Confocal images of spinal cord sections at the L1–L2 level showing lamina IX in experimental groups. (A’) Transduced neurons are indicated by arrowheads. Higher-magnification images show motoneurons, highlighting differences in pericarion and process transduction among constructs with different promoters. (B) Number of transduced neurons in lamina IX of the L1–L2 and (C) S1 spinal cord segments on day 7 after intrathecal injection of AAV constructs. Statistical analysis was performed using the Mann–Whitney U test with Bonferroni correction. (D) In the AAV-CMV-eGFP group, reporter expression was observed in both glial cells (asterisk) and neurons within lamina IX of the S1 segment. (A,A’,D) Scale bars: 50 µm. In Fig. 4A, the pink arrows indicate the neurons, which are represented on an enlarged scale in Fig. 4A’.

At the L1–2 level, eGFP expression in the AAV9-CMV-eGFP group (5.92 $\pm$ 1.79) appeared higher than in the AAV9-hSyn-mCherry (4.92 $\pm$ 1.53) and AAV9-Hb9-eYFP (4.50 $\pm$ 2.03) groups; however, no statistically significant differences were found between the groups (Fig. 4B,C). Notably, eGFP expression was detected not only in neurons but also in glial cells, particularly in segments close to the intrathecal injection site (Fig. 4D).

The pattern of eYFP expression under the control of the Hb9 promoter displayed distinct characteristics compared to the other vectors: the fluorescent signal was predominantly confined to motoneuron somata, with minimal labeling of neuronal processes, in contrast to the more diffuse labeling observed in the groups with CMV and hSyn promoters.