Tunable dual-AAV sparse labeling of PV+ retinal ganglion cells enables single-neuron projection by fMOST

Introduction: Sparse and bright labeling of retinal ganglion cell (RGC) is essential for correlating single-cell morphology with brain-wide visual circuitry. This study aimed to develop a cell-type-specific, sparse labeling strategy for parvalbumin-expressing RGCs (PV⁺ RGCs) in the transgenic mouse retina using recombinant adeno-associated virus (rAAV) and to map the whole-brain projection patterns of single PV⁺ RGCs via fluorescence micro-optical sectioning tomography (fMOST).

Methods: A cell-type-specific dual AAV system was employed, co-packaging a Cre-dependent Flpo plasmid and an Flpo-dependent enhanced yellow fluorescent protein (EYFP) plasmid. Key parameters-including the mixing ratio of core plasmids (ranging from 1/100 to 1/1000), gene copy number of Flpo and EYFP (single versus double), and AAV serotype (AAV2.2 versus engineered AAV2.NN)-were systematically optimized. Transduction efficiency and labeling sparsity under each condition were compared. Whole-retina-to-brain imaging was performed using fMOST on samples injected with the optimal condition (AAV2.2-double-1/1000), enabling the reconstruction of complete axonal trajectories of individual PV⁺ RGCs from the retina to the brain.

Results: The sparsity and signal intensity of labeled RGCs varied significantly with the core plasmid ratio, AAV serotype, and gene copy number. The engineered AAV2.NN serotype increased transduction efficiency and labeling density under equivalent conditions, which facilitated the morphological subclassification of PV⁺ RGCs into ON, ON-OFF, and OFF types based on their stratification relative to ChAT bands. Axonal projections of single PV⁺ RGCs were successfully traced to the superior colliculus (SC), dorsal and ventral lateral geniculate nuclei (dLGN/vLGN).

Discussion: This viral labeling platform effectively resolves the classical trade-off between sparsity and signal intensity, providing a robust methodology for whole-brain mapping of individual RGC projections. The approach establishes a practical foundation for future mechanistic and therapeutic studies investigating subtype-selective vulnerability in RGCs.

Introduction

The retina is a part of the central nervous system (CNS), and its intricate neural circuit structure underlies visual information processing (Corso-Díaz et al., 2018). Understanding the morphological characteristics of various types of neurons in the retina, such as the dendritic branching, axon projections, and synaptic connections they form, is critical for uncovering the fundamental principles of visual perception (Wernet et al., 2014; Ibrahim et al., 2019; Helmstaedter et al., 2013). Such insights are also essential for the diagnosis and treatment of retina-related diseases, such as glaucoma and diabetic retinopathy (Mead and Tomarev, 2016; Wang et al., 2020; Goyal et al., 2023).

To overcome the limitations of bulk labeling, sparse labeling techniques have been developed (Ku et al., 2016). These methods utilize genetic or viral tools to stochastically label only a small subset of neurons within a tissue, significantly reducing the complexity of neuronal imaging (Jefferis and Livet, 2012; Jamal et al., 2020; Veldman et al., 2020). Sparse labeling has thus become a powerful strategy for studying neuronal morphology and connectomics, allowing researchers to obtain clear and precise data on individual neuronal structures (Li et al., 2010; Kuramoto, 2019; Han et al., 2018). However, current sparse labeling methods, including recombinant adeno-associated virus (rAAV) delivery and transgenic mice models, face significant challenges (Qiu et al., 2022). A primary technical hurdle is the balance between labeling density and signal intensity (Lu, 2021; Matsumoto et al., 2024). Diluting viral vectors to achieve sparsity often results in faint labeling, compromising the visibility and traceability of labeled neurons. Conversely, increasing viral titers to enhance brightness typically results in excessive labeling density, obscuring individual neuron details. This trade-off complicates efforts to achieve both optimal sparsity and sufficient signal strength. Despite the advantages of AAV vectors, such as low immunogenicity and sustained transgene expression (Wu et al., 2025), efficiently achieving sparse labeling of parvalbumin-positive (PV⁺) RGCs remains a significant challenge.

To address these challenges, we developed an optimized AAV-based sparse labeling strategy integrated with fluorescence micro-optical sectioning tomography (fMOST). This innovative approach enabled the effective sparse labeling of individual PV⁺ RGCs and facilitated comprehensive mapping of their whole-brain projection patterns. Additionally, by screening a range of AAV serotypes, we were able to classify distinct subtypes of PV⁺ RGCs based on their labeling profiles. This research offers a novel methodology for analyzing retinal neural circuits at a higher resolution. Furthermore, the insights gained and experimental tools developed in this study provide a foundation for advancing gene therapies targeting retinal degenerative diseases, paving the way for new therapeutic strategies.

Materials and methods

Construction of virus vector

HEK293T cells were seeded at a density of 7.0 × 10⁶ cells per disk 24 h prior to transfection. The transfection involved employing a combination of an adenoviral helper plasmid, a Rep/Cap plasmid, and a plasmid mixture comprising DIO-Flpo and FDIO-EYFP or DIO-Flpo-Flpo and FDIO-EYFP-EYFP in specific ratios, delivered using PEI Pro (Cwbio, CW9309M). After a 72-h incubation period, the cells were harvested and collected by centrifugation. Viral particles were liberated through the application of a high-salt lysis buffer, followed by five cycles of freezing and thawing. To eliminate genomic DNA and residual plasmids, Benzonase (Sigma, 9025-65-4) was added prior to the precipitation of proteins, including AAVs, with polyethylene glycol (PEG, Aladdin, P103734) for 1 h. The precipitated PEG pellet was resuspended overnight, and AAVs were subsequently purified using an iodixanol gradient to eliminate contaminating proteins and empty capsids. The virus-laden iodixanol fraction was concentrated and underwent buffer exchanged through Amicon Ultra-15 filtration tubes (Merck, UFC910024). The concentrated AAV solution was sterile, filtrated, aliquoted, and stored at −80 °C for later use. The viral titer is determined using real-time quantitative PCR (qPCR) and adjusted to 5.0 × 10¹² vg/mL.

The nomenclature for dual AAV vectors adheres to the format: AAV2.X (where X represents the serotype) + Gene copy (single/double indicating the gene copy number) + ratio (the mixing ratio of Cre-dependent Flpo plasmid to Flpo-dependent enhanced yellow fluorescent protein). For instance, AAV2.2-single-1/100 indicates that this dual AAV vector was produced by co-packaging DIO-Flpo and FDIO-EYFP at a ratio of 1/100 with the AAV2.2 serotype.

Animal model and intravitreal injection

Experiments were conducted using 6–8 weeks old PV-Cre mice (weighing ∼20 g). All animal procedures were performed in accordance with Animal Ethics Committee of Wuhan University (MRI2024-LAC010). Mice were anesthetized (1.25% avertin, 20 μL/g i.p.), and then intravitreal injection was performed using a microsyringe (Hamilton, #65). The injection volume was 1.5 μL per eye, and the injection site was located 0.5 mm posterior to the limbus, mouse was injected into the right eye only for fMOST. After the injection, mice were kept in a warm environment until fully awake.

Retinal flat-mount and immunofluorescence staining

Five weeks after injection, mice were deeply anesthetized and then underwent cardiac perfusion with ice-cold 0.1 M phosphate-buffered saline (PBS), followed by fixation with 4% paraformaldehyde (PFA). Ocular globes were carefully extracted and further fixed in 4% paraformaldehyde for 45 min before dissecting the intact retina for retinal flat-mount preparation. The optic nerve was transferred into a sucrose solution for gradient dehydration (at 4 °C). Subsequently, frozen sections were prepared at a thickness of 30 μm. Both retinal flat mounts and sections were washed with PBS and blocked overnight in a solution containing 4% BSAT at 4 °C. Immunofluorescence staining was performed using anti-GFP antibody (Abcam, ab13970) and anti-ChAT antibody (Millipore, AB144P), with all steps conducted at 4 °C. Images were acquired using laser scanning confocal microscopes (Zeiss LSM 880 and Leica Stellaris 5 WLL) and a high-resolution rapid fluorescence microscope (Leica Thunder).

Tissue preparation for fMOST

The mice were deeply anesthetized and subjected to cardiac perfusion using ice-cold 0.1 M PBS followed by 4% PFA. Post-extraction from the cranial cavity, the brain and eyes were fixed in 4% PFA at 4 °C for 24 h, followed by a 24-h rinse in PBS. The brain and eyes were then prepared for resin embedding.

The samples underwent a graded dehydration process through a sequential series of ethanol solutions at concentrations of 50%, 75%, 95%, and 100%, with each step lasting 2 h. Following dehydration, the specimens were infiltrated with LR-White resin in increasing concentrations (50%, 70%, 85%, and 100%), each for 2 h. They were then immersed overnight in 100% LR-White resin overnight for 36 h at 4 °C, with the resin solution refreshed after 12 h. The final polymerization of the brains was conducted in a vacuum oven set at 38 °C for 24 h. The 100% LR-White resin was prepared by combining 100 g of LR-White with 0.24 g of ABVN as a polymerization initiator.

fMOST imaging and data processing

The LR-White resin-embedded brains were imaged using the fMOST system (Wuhan OE-Bio Co., Ltd.). Each brain was mounted with the fMOST automated data acquisition system, which comprises a 473 nm laser, a 40× water immersion objective (Olympus, N2667700), and a time delay and integration charge-coupled device (TDI-CCD) for signal detection. Semi-thin coronal sections of 2 μm thickness were sequentially imaged from anterior to posterior, with the data acquisition phase extending over 4–5 days per sample. This meticulous process generated approximately 6,500 coronal sections, facilitating the creation of a comprehensive brain dataset. The raw images, with a resolution of 0.35 μm × 0.35 μm × 2 μm, were processed using specialized software for stripe stitching and brightness normalization. The processed images were then used for 3D reconstruction using Imaris software. The resulting brain dataset was downsampled and aligned to the Allen Reference Brain Atlas, using the Common Coordinate Framework version 3 (CCFv3) for standardized spatial orientation.

Data analysis and statistical processing

The data were processed using ImageJ and Aivia software. The images and fluorescence intensities were processed or analyzed by ImageJ. To ensure more reasonable normalization of fluorescence signal intensity, we first standardized the imaging parameters for capturing fluorescent images to avoid signal intensity deviations caused by differences in exposure time, gain, etc. All images were obtained under the same microscope settings. Then we standardized the method for fluorescence signal quantification:

1. Open ImageJ software and load the prepared image files into the software.

2. Click on Image→Adjust→Brightness/Contrast and set the Max parameter to 255.

3. Click on Analyze→Tools→ROI Manager.

4. Use selection tool to choose the EYFP-positive somata.

5. Click the Add button in the ROI Manager to add the selected regions.

6. Once all regions have been added, click the Measure button to perform calculations. The results, including parameters such as Area and Mean gray value for each region, will be automatically generated for subsequent statistical analysis.

7. To ensure data accuracy, perform the same experiment three times or more and record the corresponding values, the specific n-values are indicated in the legends of the corresponding images.

Finally, apply the normalization formula: subtract the average gray value of the background (background mean gray value) from the average gray value of the target region to complete the fluorescence intensity correction.

The number of fluorescent cells in retinal flat mounts refers to the total count of all EYFP-positive cells in each whole retina, for each group, we analyzed the retinas from five different mice. For each retina, we selected multiple EYFP-positive RGC somata as regions of interest and calculated the mean fluorescence intensity per retina; these per-retina values were then used for statistical analysis. Statistical analysis was performed using GraphPad Prism 9.0. One-way analysis of variance (ANOVA) was used for comparisons between groups, followed by Tukey’s post-hoc test for multiple comparisons. Data are presented as mean ± standard deviation (SD), and differences with p < 0.05 were considered statistically significant (ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Results

Screening of dual plasmid mixing ratios for AAV sparse labeling

We employed PV-Cre mice and a dual AAV system for specifically and sparsely targeting of PV⁺ RGCs. The dual AAV system is consisted of two vectors: a Cre-dependent Flpo plasmid and Flpo-dependent enhanced yellow fluorescent protein (EYFP) plasmid. We obtained four dual AAV vectors products (AAV2.2-single-1/100, AAV2.2-single-1/300, AAV2.2-single-1/600, AAV2.2-single-1/1000) by mixing the Cre-dependent Flpo plasmid and Flpo-dependent EYFP plasmid at four different ratios (1/100, 1/300, 1/600 and 1/1000) during the AAV production. We intravitreally injected these four products into the retina of Cre mice and collected the retina and optic nerve 5 weeks postinjection (Figure 1A). In the WT control mice, EYFP expression in neither retinal cell bodies nor the optic nerve was observed. Meanwhile, the results showed that as the mixing ratio increases from 1/100 to 1/1000, the number of labeled PV⁺ cells progressively decreased (Figure 1B). The number of fluorescent cells in retinal flat mounts is the total count of all EYFP-positive cells in each whole retina. Specifically, the product of 1/100 ratio labeled approximately 542.0 ± 39.2 cells and the product of 1/300 labeled approximately 144.0 ± 47.6 cells, but it is still insufficient to observe the complete morphology of individual cells. The product of 1/600 ratios labeled 35.4 ± 5.8 cells, while the product of 1/1000 ratio resulted in up to 4 labeled cells. Surprisingly, in one retina of 1:1000 ratio group, only a single PV⁺ RGC was labeled (Figure 1C). Accordingly, the number of EYFP-positive optic nerves showed a similar trend to that of EYFP-positive somata, ∼7–8 axons were labeled in 1/100 group, while only 1 axon was labeled in 1/1000 group (Figure 1D). There was no significant difference in the fluorescence intensity of individual retina soma in PV-Cre mice across the different ratio groups (Figure 1E). Thus, we established a novel strategy for sparse and high-brightness labeling of specific types of retinal cells by combining transgenic mice with a dual-plasmid AAV viral system.

FIGURE 1

Figure 1. Sparse labeling of RGCs using a dual-plasmid strategy and the resulting labeling patterns at different plasmid mixing ratios. (A) Schematic diagram of virus packaging and intravitreal injection in mice. (B) The representative image of virus expression in the retina, enlarged view (zoom-in) of dashed square part of left image of retina and optic nerve, 1/100, 1/300, 1/600, 1/1000 represent mixing ratio of the core plasmid. From the left to right, scale bar = 200, 50, 100 μm. (C) Statistical chart of GFP positive cell numbers in different groups, n = 5 retinas in each group. (D) Statistical chart of GFP positive RGC axons in different groups, n = 5 optic nerves in each group. (E) Statistical chart of signal intensity of cell bodies in different groups, n = 5 retinas in each group. Data are represented as mean ± SD, with statistical significance indicated by asterisks (ns = not significant, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Optimization of sparse labeling conditions by increasing the gene copy number

Next, we systematically optimized the copy number parameters of the target gene in the core plasmid rAAV-EF1a-DIO-Flpo-Flpo-WPRE-hGH-pA and rAAV-nEF1a-FDIO-EYFP-EYFP-WPRE-hGH-pA, within the payload capacity of AAV2.2. By increasing the copy number of Flpo and EYFP within these plasmids, we successfully elevated the expression levels of the EYFP protein, as reflected in the enhanced fluorescence intensity seen in labeled cells (Figure 2A). Interestingly, at a plasmid ratio of 1/600, there was no significant difference observed in the number of EYFP-positive cells or the labeled axons when comparing double-copy with single-copy groups (Figures 2B, D). Nonetheless, the average fluorescence intensity per cell exhibited a significant difference between these groups. Specifically, the mean signal intensity in the single-copy group was 70.2 ± 10.9, whereas the double-copy groups’ intensity was significantly higher 94.1 ± 11.4 (p < 0.01) (Figure 2C). This optimization paves the way for more detailed studies on the morphology of RGCs and their projections to downstream brain regions, enhancing our understanding of neural pathways.

FIGURE 2

Figure 2. Labeling efficiency of AAV2.2 vectors harboring different transgene copy numbers. (A) The representative image of virus expression in the mouse retina, partial enlarged view and optic nerve, single copy and double copies indicate the copy number parameters of the target gene of the core plasmid. From the left to right, scale bar = 200, 50, 100 μm. (B) Statistical chart of GFP positive cell numbers in different groups, n = 5 retinas in each group. (C) Signal intensity of single cell in different groups, n = 5 retinas in each group. (D) Statistical chart of GFP positive RGC axons in different groups, n = 5 optic nerves in each group. Data are represented as mean ± SD, with statistical significance indicated by asterisks (ns = not significant, **p < 0.01).

Mapping of individual RGC projections to brain based on fMOST imaging

To elucidate the precise projection patterns of RGC to visual centers at single-cell resolution, we employed fMOST to acquire high-resolution three-dimensional data from the whole brains of transgenic mice subjected to anterograde labeling. Through image registration, neuron tracing, and three-dimensional reconstruction, we successfully reconstructed the complete and morphologically intact axon of a single RGC, along with its distinct projection pathways and terminal branches in all target areas, including the superior colliculus and lateral geniculate nucleus. We performed whole-brain imaging and data reconstruction of sparsely labeled PV⁺ RGC 5 weeks after virus injection to the right eye of PV-Cre mice (Figure 3A). The complete projection patterns of individual neurons were reconstructed (Figure 3B). Computational neuroanatomical analysis revealed that the neuron demonstrated origin locations and distinct projection patterns.

FIGURE 3

Figure 3. Visualization of RGC projections to brain targets using fMOST imaging. (A) Schematic diagram of intravitreal injection in the mouse (right eye), tissue processing and fMOST imaging. (B) fMOST imaging of projection from single retinal neurons to brain regions, scale bar = 1000 μm. (C) Labeling and enlarged view of projecting nerve fibers in the dLGN and vLGN brain region, bregma = –2.4 mm, scale bar = 1000 μm (left) and 400 μm (right). (D) Labeling and enlarged view of projecting nerve fibers in the SC brain region, bregma = –3.84 mm, scale bar = 1000 μm (left) and 400 μm (right).

We found that PV⁺ RGCs primarily project to regions such as the superficial layers of the dorsal lateral geniculate nucleus (dLGN), the ventral lateral geniculate nucleus (vLGN) (Figure 3C) and the superior colliculus (SC) (Figure 3D). Supplementary Video 1 demonstrates the complete projection continuously traced from single-cell labeling in the retina to brain regions. We observed the overall trajectory of individual RGC axon upon exiting the optic chiasm. As the axon approach the midbrain and forebrain regions, it exhibits distinct target-specific branching behaviors, directing their projections to SC and LGN, respectively. Notably, we did not observe any RGC axon branches projecting to non-visual nuclei, consistent with their projection specificity. Within the LGN, axon terminals form highly intricate, densely branched arborizations (Figure 3C). Compared to the terminals in the SC, those in the LGN generally display greater complexity, with more branching points, forming a smaller but more compact structural volume (Figure 3D). In summary, our fMOST imaging data reveal a distinct pattern: individual RGC transmit information to both the SC and LGN in a parallel manner via target-specific axonal branching. However, the presynaptic structures formed in these primary visual centers differ significantly in spatial scale, morphological complexity, and subnuclear localization, likely reflecting their distinct functional requirements in processing motion perception (SC) versus pattern and detail vision (LGN-cortical pathway).

Optimization of AAV serotypes and PV⁺ RGC classification

AAV2.2 vectors have been sufficient to meet the needs of studies involving sparse labeling and central projection tracing (Chen et al., 2010). To better delineate neuronal morphology, we selected AAV2.NN as a candidate, which exhibits stronger penetration of the inner limiting membrane in the retina and a larger payload capacity compared to AAV2.2 (Pavlou et al., 2021). We found that the RGC projections labeled by AAV2.NN were more distinct and denser under the same condition compared to that of AAV2.2 (Figure 4A). Whether the number or the fluorescence intensity of labeled cells, AAV2.NN is greater than AAV2.2 (Figures 4B, C). The number of axons also exhibits a corresponding trend (Figure 4D). Therefore, we performed morphological classification of PV⁺ RGCs labeled with AAV2.NN.

FIGURE 4

Figure 4. Comparison of labeling efficiency across different AAV serotypes. (A) The representative image of virus expression in the mouse retina, partial enlarged view and optic nerve. From the left to right, scale bar = 200, 50, 100 μm. (B) Statistical chart of GFP positive cell numbers in different groups, n = 5 retinas in each group. (C) Signal intensity of single cell in different groups, n = 5 retinas in each group. (D) Statistical chart of GFP positive RGC axons in different groups, n = 5 in optic nerves each group. Data are represented as mean ± SD, with statistical significance indicated by asterisks (*p < 0.05, ****p < 0.0001).

The classification of RGC subtypes is instrumental in understanding parallel processing of visual information at the retinal level. In optic nerve degenerative diseases like glaucoma, different RGC subtypes demonstrate varying susceptibility to damage (Denniss et al., 2025; Liu et al., 2023). Recognizing this selective vulnerability is essential for developing targeted neuroprotective therapies. Furthermore, ensuring that regenerating axons in optic nerve regeneration correctly reach their target brain regions relies heavily on their subtype identity.

Recent authoritative studies have classified mouse RGCs into more than 40–50 transcriptomic subtypes (Goetz et al., 2022; Sanes and Masland, 2015), highlighting the complexity and diversity of these cells. In the retina, there is a specific stratified correspondence between the dendrites of ON and OFF RGCs and the dendrites of starburst amacrine cells (SACs) within the inner plexiform layer (IPL) (Sun et al., 2015; Chen et al., 2021; Guadagni et al., 2016). The dendrites of ON RGCs are primarily distributed in the inner sublayer (sublamina b) of the IPL, located at approximately 40% depth (measured from the ganglion cell layer, GCL), overlapping with the dendritic band of ON-type SACs (Supplementary Video 2; Baden et al., 2016; Sanes and Masland, 2015). Conversely, the dendrites of OFF RGCs are concentrated in the outer sublayer (sublamina a) of the IPL, at a depth of about 77% (measured from the GCL), corresponding to the dendritic band of OFF-type SACs (Supplementary Video 3). The dendrites of ON-OFF cells extend into both the ON and OFF sublayers (Supplementary Video 4; Baden et al., 2016).

As shown in Figure 5, PV⁺ RGCs can be subdivided based on their dendritic positioning within SAC synapses into three types: ON, ON-OFF, and OFF (Yi et al., 2012). Morphologically, ON cells can be further classified into distinct subtypes: PV_on1, characterized by a large soma and a large dendritic field (large parasol-like), probably with fast conduction speed, primarily responsible for motion detection and luminance information (Goetz et al., 2022; Sanes and Masland, 2015); PV_on2, featuring a small soma and a small dendritic field (small parasol-like), which might have slower conduction speeds and is mainly involved in high spatial resolution and fine vision (Goetz et al., 2022); PV_on3, having a large soma with a small dendritic field (sunflower-like); and PV_on4, distinguished by dendrites exhibiting bilateral symmetry (butterfly-like). This detailed classification and understanding of RGC subtypes provide a foundation for exploring their functional roles and developing interventions that address specific vulnerabilities and regeneration strategies.

FIGURE 5

Figure 5. Classification of PV⁺ RGC subtypes based on sparse labeling and morphological analysis. Z-stack layered scanning of the retina with x-z overlay and x-y overlay images. The white dashed lines represent the ChAT (choline acetyltransferase, marker of SACs) labeling schematic, with the upper part indicating ON and the lower part indicating OFF, PV_on1–4 represents four different types of ON PV⁺ RGCs, scale bar = 20 μm.

Discussion

We present a practical framework for celltype-specific sparse labeling in the adult retina that reconciles sparsity with brightness and enables brain-wide single-neuron reconstruction. By independently tuning plasmid ratio and gene copy number, we obtain single PV⁺ RGC labels suitable for fMOST without losing visibility. The observed target-specific collateralization to SC/LGN and distinct terminal morphologies highlight pathway specialized presynaptic architectures at single-neuron resolution. Relative to prior sparse-labeling approaches (Lin et al., 2018; Ibrahim et al., 2019), our dual-AAV design requires only a single Cre driver line and standard intravitreal delivery, facilitating adoption. AAV2.NN further broadens use cases by boosting efficiency for subtype surveys and mesoscale mapping.

Sparse labeling is essential for resolving detailed neuronal morphology at the single-cell level (Costa et al., 2016). By optimizing the ratio of the viral core plasmids, we maintained the labeling density within an ideal range, effectively minimizing signal overlap that often arises from excessive labeling. This methodological framework offers a valuable reference for investigating other sparsely distributed neuronal populations. Furthermore, the implementation of a double-copy strategy–by increasing the copy number of core viral components–enhanced labeling clarity, thereby improving the accuracy of neuronal projection tracing. Such advances are critical for elucidating the structural and functional architecture of neural circuits. The application of fMOST technology further permitted the analysis of brain-wide connectivity with single-cell resolution. Our observation of topologically organized projections aligns with findings reported by Jiao et al. (2025) in hypothalamic neurons, suggesting that such organizational principles may be a general feature of brain network architecture. The selection of an appropriate AAV serotype is critical for achieving high transduction efficiency in the retina. Our results indicate that the engineered serotype AAV2.NN transduces RGCs with significantly greater efficiency than the conventional AAV2.2, corroborating recent reports (Zhang et al., 2025). This enhancement is likely attributable to specific mutations in the viral capsid proteins, which improve affinity for cell surface receptors and facilitate viral internalization and intracellular trafficking (Weinmann et al., 2022; Zhang et al., 2025). These improvements in AAV-mediated gene delivery not only support precise neuronal labeling in neurobiology research but also hold therapeutic potential for vision disorders involving RGCs (Zhang et al., 2022; Tribble et al., 2014; Nimkar et al., 2025). Continued refinement of capsid engineering may further extend the utility of this approach to other neuronal subtypes and tissues.

Nonetheless, our study has several limitations. First, all experiments were performed in adult mice, which precluded the examination of development dynamics in PV⁺ RGCs projection patterns (Solomon et al., 2018; Beck et al., 2005). The whole eye–brain level single-neuron tracking system is used to provide a representative example of the long-range projection pattern of a sparsely labeled PV⁺ RGC. We processed the fMOST data by overlaying 25 consecutive 2-μm sections to generate a refined 3D reconstruction. This processing improved the visualization of RGC axonal arborizations and yielded clearer videos showing coronal (Supplementary Video 5) and sagittal (Supplementary Video 6) views of the soma and dendrites in the retina, and the axonal projections in the SC, vLGN and dLGN (Supplementary Videos 7–9). This reconstruction illustrates the feasibility and resolution of our approach, but it does not imply that all PV⁺ ON, OFF, and ON-OFF subtypes share the same projection pattern (Attallah et al., 2021).

Future research directions may include: (1) developing more specific promoters to enable selective labeling of RGC subtypes, a systematic comparison among ON, OFF, and ON-OFF subtypes will require additional, subtype-specific reconstructions; (2) integrating optogenetics with in vivo calcium imaging to investigate functional properties of distinct projection subtypes; and (3) establishing disease models to assess alterations in PV⁺ RGC projections under pathological conditions. From a technical standpoint, optimizing algorithms for fMOST data processing will be a key future focus. Isotropic resolution recovery techniques based on deep learning, such as Self-Net, have already demonstrated significant potential (Bastug et al., 2024; Fotaki et al., 2022; Menze et al., 2024). Finally, our findings open new avenues for treating retinal degenerative diseases. For example, AAV vectors could be engineered to deliver neurotrophic factors specifically to PV⁺ RGCs, promoting neuronal survival and axonal regeneration (Rodger et al., 2012; Cen et al., 2017; Yungher et al., 2017). Sparse labeling and neuronal reconstruction–encompassing the precise three-dimensional mapping of somata, axons, and dendrites–remain foundational techniques for realizing these goals (Abdellah et al., 2018).

Data availability statement

The original contributions presented in this study are included in this article/Supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement

The animal study was approved by Animal Ethics Committee of Wuhan University (MRI2024-LAC010). The study was conducted in accordance with the local legislation and institutional requirements.

Author contributions

LZ: Validation, Writing – review & editing, Formal analysis, Writing – original draft, Data curation, Visualization. GT: Investigation, Software, Writing – review & editing, Validation. YL: Data curation, Writing – review & editing. MY: Data curation, Writing – review & editing. SJ: Methodology, Writing – review & editing. WZ: Software, Writing – review & editing. QW: Writing – original draft, Data curation. YS: Funding acquisition, Conceptualization, Writing – review & editing, Resources, Project administration.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by grants from the National Natural Science Foundation of China (NSFC No. 82471086) and the Hubei Provincial Health and Technology Project.

Conflict of interest

SJ and QW were employed by Zhongmou Therapeutics Co., Ltd.

The remaining authors declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fncir.2025.1740624/full#supplementary-material

References

Abdellah, M., Hernando, J., Eilemann, S., Lapere, S., Antille, N., Markram, H., et al. (2018). NeuroMorphoVis: A collaborative framework for analysis and visualization of neuronal morphology skeletons reconstructed from microscopy stacks. Bioinformatics 34, i574–i582. doi: 10.1093/bioinformatics/bty231

PubMed Abstract | Crossref Full Text | Google Scholar

Attallah, N., Horsburgh, J., Beckwith, A., and Tracy, R. (2021). Residential water meters as edge computing nodes: Disaggregating end uses and creating actionable information at the edge. Sensors 21:5310. doi: 10.3390/s21165310

PubMed Abstract | Crossref Full Text | Google Scholar

Baden, T., Berens, P., Franke, K., Román Rosón, M., Bethge, M., and Euler, T. (2016). The functional diversity of retinal ganglion cells in the mouse. Nature 529, 345–350. doi: 10.1038/nature16468

PubMed Abstract | Crossref Full Text | Google Scholar

Bastug, B. T., Guneri, G., Yildirim, M. S., Corbaci, K., and Dandil, E. (2024). Fully Automated detection of the appendix using U-net deep learning architecture in CT scans. J. Clin. Med. 13:5893. doi: 10.3390/jcm13195893

PubMed Abstract | Crossref Full Text | Google Scholar

Beck, R., King, M., Ha, G., Cushman, J., Huang, Z., and Petitto, J. M. (2005). IL-2 deficiency results in altered septal and hippocampal cytoarchitecture: Relation to development and neurotrophins. J. Neuroimmunol. 160, 146–153. doi: 10.1016/j.jneuroim.2004.11.006

PubMed Abstract | Crossref Full Text | Google Scholar

Cen, L., Liang, J., Chen, J., Harvey, A., Ng, T., Zhang, M., et al. (2017). AAV-mediated transfer of RhoA shRNA and CNTF promotes retinal ganglion cell survival and axon regeneration. Neuroscience 343, 472–482. doi: 10.1016/j.neuroscience.2016.12.027

PubMed Abstract | Crossref Full Text | Google Scholar

Chen, H., Xu, H. P., Wang, P., and Tian, N. (2021). Visual deprivation retards the maturation of dendritic fields and receptive fields of mouse retinal ganglion cells. Front. Cell. Neurosci. 15:640421. doi: 10.3389/fncel.2021.640421

PubMed Abstract | Crossref Full Text | Google Scholar

Chen, S., Ma, H., Han, J., Lu, R., Tao, P., Law, P., et al. (2010). Antinociceptive effects of morphine and naloxone in mu-opioid receptor knockout mice transfected with the MORS196A gene. J. Biomed. Sci. 17:28. doi: 10.1186/1423-0127-17-28

PubMed Abstract | Crossref Full Text | Google Scholar

Corso-Díaz, X., Jaeger, C., Chaitankar, V., and Swaroop, A. (2018). Epigenetic control of gene regulation during development and disease: A view from the retina. Prog. Retin. Eye Res. 65, 1–27. doi: 10.1016/j.preteyeres.2018.03.002

PubMed Abstract | Crossref Full Text | Google Scholar

Costa, M., Manton, J., Ostrovsky, A., Prohaska, S., and Jefferis, G. S. (2016). NBLAST: Rapid, sensitive comparison of neuronal structure and construction of neuron family databases. Neuron 91, 293–311. doi: 10.1016/j.neuron.2016.06.012

PubMed Abstract | Crossref Full Text | Google Scholar

Denniss, J., Cheloni, R., Martin, J., and Spitschan, M. (2025). Pupil responses to melanopsin-isolating stimuli as a potential diagnostic biomarker for glaucoma. PLoS One 20:e0324373. doi: 10.1371/journal.pone.0324373

PubMed Abstract | Crossref Full Text | Google Scholar

Fotaki, A., Fuin, N., Nordio, G., Velasco Jimeno, C., Qi, H., Emmanuel, Y., et al. (2022). Accelerating 3D MTC-BOOST in patients with congenital heart disease using a joint multi-scale variational neural network reconstruction. Magn. Reson. Imaging 92, 120–132. doi: 10.1016/j.mri.2022.06.012

PubMed Abstract | Crossref Full Text | Google Scholar

Goetz, J., Jessen, Z., Jacobi, A., Mani, A., Cooler, S., Greer, D., et al. (2022). Unified classification of mouse retinal ganglion cells using function, morphology, and gene expression. Cell. Rep. 40:111040. doi: 10.1016/j.celrep.2022.111040

PubMed Abstract | Crossref Full Text | Google Scholar

Goyal, V., Read, A., Ritch, M., Hannon, B., Rodriguez, G., Brown, D., et al. (2023). AxoNet 2.0: A deep learning-based tool for morphometric analysis of retinal ganglion cell axons. Transl. Vis. Sci. Technol. 12:9. doi: 10.1167/tvst.12.3.9

PubMed Abstract | Crossref Full Text | Google Scholar

Guadagni, V., Cerri, C., Piano, I., Novelli, E., Gargini, C., Fiorentini, C., et al. (2016). The bacterial toxin CNF1 as a tool to induce retinal degeneration reminiscent of retinitis pigmentosa. Sci Rep. 6:35919. doi: 10.1038/srep35919

PubMed Abstract | Crossref Full Text | Google Scholar

Han, Y., Kebschull, J., Campbell, R., Cowan, D., Imhof, F., Zador, A., et al. (2018). The logic of single-cell projections from visual cortex. Nature 556, 51–56. doi: 10.1038/nature26159

PubMed Abstract | Crossref Full Text | Google Scholar

Helmstaedter, M., Briggman, K., Turaga, S., Jain, V., Seung, H., and Denk, W. (2013). Connectomic reconstruction of the inner plexiform layer in the mouse retina. Nature 500, 168–174. doi: 10.1038/nature12346

PubMed Abstract | Crossref Full Text | Google Scholar

Ibrahim, L. A., Huang, J. J., Wang, S. Z., Kim, Y. J., Zhang, L. I., and Tao, H. W. (2019). Sparse labeling and neural tracing in brain circuits by STARS strategy: Revealing morphological development of type II spiral ganglion neurons. Cereb. Cortex 29:1700. doi: 10.1093/cercor/bhy154

PubMed Abstract | Crossref Full Text | Google Scholar

Jamal, L., Kiyama, T., and Mao, C. (2020). Genetically directed sparse labeling system for anatomical studies of retinal ganglion cells. Methods Mol. Biol. 2092, 187–194. doi: 10.1007/978-1-0716-0175-4_13

PubMed Abstract | Crossref Full Text | Google Scholar

Jefferis, G. S., and Livet, J. (2012). Sparse and combinatorial neuron labelling. Curr. Opin. Neurobiol. 22, 101–110. doi: 10.1016/j.conb.2011.09.010

PubMed Abstract | Crossref Full Text | Google Scholar

Jiao, Z., Gao, T., Wang, X., Wang, A., Ma, Y., Feng, L., et al. (2025). Projectome-based characterization of hypothalamic peptidergic neurons in male mice. Nat. Neurosci. 28, 1073–1088. doi: 10.1038/s41593-025-01919-0

PubMed Abstract | Crossref Full Text | Google Scholar

Ku, T., Swaney, J., Park, J., Albanese, A., Murray, E., Cho, J., et al. (2016). Multiplexed and scalable super-resolution imaging of three-dimensional protein localization in size-adjustable tissues. Nat. Biotechnol. 34, 973–981. doi: 10.1038/nbt.3641

PubMed Abstract | Crossref Full Text | Google Scholar

Kuramoto, E. (2019). Method for labeling and reconstruction of single neurons using Sindbis virus vectors. J. Chem. Neuroanat. 100:101648. doi: 10.1016/j.jchemneu.2019.05.002

PubMed Abstract | Crossref Full Text | Google Scholar

Li, L., Tasic, B., Micheva, K., Ivanov, V., Spletter, M., Smith, S., et al. (2010). Visualizing the distribution of synapses from individual neurons in the mouse brain. PLoS One 5:e11503. doi: 10.1371/journal.pone.0011503

PubMed Abstract | Crossref Full Text | Google Scholar

Lin, R., Wang, R., Yuan, J., Feng, Q., Zhou, Y., Zeng, S., et al. (2018). Cell-type-specific and projection-specific brain-wide reconstruction of single neurons. Nat. Methods 15, 1033–1036. doi: 10.1038/s41592-018-0184-y

PubMed Abstract | Crossref Full Text | Google Scholar

Liu, P., Chen, W., Jiang, H., Huang, H., Liu, L., Fang, F., et al. (2023). Differential effects of SARM1 inhibition in traumatic glaucoma and EAE optic neuropathies. Mol. Ther. Nucleic Acids 32, 13–27. doi: 10.1016/j.omtn.2023.02.029

PubMed Abstract | Crossref Full Text | Google Scholar

Lu, M. (2021). Single-molecule FRET imaging of virus spike-host interactions. Viruses 13:332. doi: 10.3390/v13020332

PubMed Abstract | Crossref Full Text | Google Scholar

Matsumoto, N., Barson, D., Liang, L., and Crair, M. (2024). Hebbian instruction of axonal connectivity by endogenous correlated spontaneous activity. Science 385:eadh7814. doi: 10.1126/science.adh7814

PubMed Abstract | Crossref Full Text | Google Scholar

Mead, B., and Tomarev, S. (2016). Evaluating retinal ganglion cell loss and dysfunction. Exp. Eye Res. 151, 96–106. doi: 10.1016/j.exer.2016.08.006

PubMed Abstract | Crossref Full Text | Google Scholar

Menze, R., Hesse, B., Kusmierczuk, M., Chen, D., Weitkamp, T., Bettink, S., et al. (2024). Synchrotron microtomography reveals insights into the degradation kinetics of bio-degradable coronary magnesium scaffolds. Bioact. Mater. 32, 1–11. doi: 10.1016/j.bioactmat.2023.09.008

PubMed Abstract | Crossref Full Text | Google Scholar

Nimkar, K., Tsai, N. Y., Zhao, M., Yi, Y., Lum, M. R., Garrett, T. R., et al. (2025). Molecular and spatial analysis of ganglion cells on retinal flatmounts identifies perivascular neurons resilient to glaucoma. Neuron 113, 3390–407.e8. doi: 10.1016/j.neuron.2025.07.025

PubMed Abstract | Crossref Full Text | Google Scholar

Pavlou, M., Schön, C., Occelli, L., Rossi, A., Meumann, N., Boyd, R., et al. (2021). Novel AAV capsids for intravitreal gene therapy of photoreceptor disorders. EMBO Mol. Med. 13:e13392. doi: 10.15252/emmm.202013392

PubMed Abstract | Crossref Full Text | Google Scholar

Qiu, L., Zhang, B., and Gao, Z. (2022). Lighting up neural circuits by viral tracing. Neurosci. Bull. 38, 1383–1396. doi: 10.1007/s12264-022-00860-7

PubMed Abstract | Crossref Full Text | Google Scholar

Rodger, J., Drummond, E., Hellström, M., Robertson, D., and Harvey, A. (2012). Long-term gene therapy causes transgene-specific changes in the morphology of regenerating retinal ganglion cells. PLoS One 7:e31061. doi: 10.1371/journal.pone.0031061

PubMed Abstract | Crossref Full Text | Google Scholar

Sanes, J. R., and Masland, R. H. (2015). The types of retinal ganglion cells: Current status and implications for neuronal classification. Annu. Rev. Neurosci. 38, 221–246. doi: 10.1146/annurev-neuro-071714-034120

PubMed Abstract | Crossref Full Text | Google Scholar

Solomon, A., Westbrook, T., Field, G., and McGee, A. (2018). Nogo receptor 1 is expressed by nearly all retinal ganglion cells. PLoS One 13:e0196565. doi: 10.1371/journal.pone.0196565

PubMed Abstract | Crossref Full Text | Google Scholar

Sun, L., Brady, C., Cahill, H., Al-Khindi, T., Sakuta, H., Dhande, O., et al. (2015). Functional assembly of accessory optic system circuitry critical for compensatory eye movements. Neuron 86, 971–984. doi: 10.1016/j.neuron.2015.03.064

PubMed Abstract | Crossref Full Text | Google Scholar

Tribble, J., Cross, S., Samsel, P., Sengpiel, F., and Morgan, J. E. (2014). A novel system for the classification of diseased retinal ganglion cells. Vis. Neurosci. 31, 373–380. doi: 10.1017/S0952523814000248

PubMed Abstract | Crossref Full Text | Google Scholar

Veldman, M., Park, C., Eyermann, C., Zhang, J., Zuniga-Sanchez, E., Hirano, A., et al. (2020). Brainwide genetic sparse cell labeling to illuminate the morphology of neurons and glia with Cre-dependent MORF mice. Neuron 108, 111–27.e6. doi: 10.1016/j.neuron.2020.07.019

PubMed Abstract | Crossref Full Text | Google Scholar

Wang, A., Lee, P., Bui, B., Jobling, A., Greferath, U., Brandli, A., et al. (2020). Potential mechanisms of retinal ganglion cell type-specific vulnerability in glaucoma. Clin. Exp. Optom. 103, 562–571. doi: 10.1111/cxo.13031

PubMed Abstract | Crossref Full Text | Google Scholar

Weinmann, J., Söllner, J., Abele, S., Zimmermann, G., Zuckschwerdt, K., Mayer, C., et al. (2022). Identification of broadly applicable adeno-associated virus vectors by systematic comparison of commonly used capsid variants in vitro. Hum. Gene Ther. 33, 1197–1212. doi: 10.1089/hum.2022.109

PubMed Abstract | Crossref Full Text | Google Scholar

Wernet, M., Huberman, A., and Desplan, C. (2014). So many pieces, one puzzle: Cell type specification and visual circuitry in flies and mice. Genes Dev. 28, 2565–2584. doi: 10.1101/gad.248245.114

PubMed Abstract | Crossref Full Text | Google Scholar

Wu, W., McRae, J., Brown, A., and Agbaga, M. (2025). Tropism and retinal transduction efficiency of adeno-associated virus serotypes in mice. Invest. Ophthalmol. Vis. Sci. 66:18. doi: 10.1167/iovs.66.12.18

PubMed Abstract | Crossref Full Text | Google Scholar

Yi, C., Yu, S., Lee, E., Lee, J., and Jeon, C. (2012). Types of parvalbumin-containing retinotectal ganglion cells in mouse. Acta Histochem. Cytochem. 45, 201–210. doi: 10.1267/ahc.11061

PubMed Abstract | Crossref Full Text | Google Scholar

Yungher, B. J., Ribeiro, M., and Park, K. K. (2017). Regenerative responses and axon pathfinding of retinal ganglion cells in chronically injured mice. Invest. Ophthalmol. Vis. Sci. 58, 1743–1750. doi: 10.1167/iovs.16-19873

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, N., He, X., Xing, Y., and Yang, N. (2022). Differential susceptibility of retinal ganglion cell subtypes against neurodegenerative diseases. Graefes Arch. Clin. Exp. Ophthalmol. 260, 1807–1821. doi: 10.1007/s00417-022-05556-2

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, T., Wang, F., Wu, Y., Cao, J., and Shen, Y. (2025). Retinal transduction profiling of diverse AAV serotypes via intravitreal injection. J. Virol. 99:e0063725. doi: 10.1128/jvi.00637-25

PubMed Abstract | Crossref Full Text | Google Scholar