Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 29.
Published in final edited form as: Science. 2025 Oct 16;390(6770):eadr9109. doi: 10.1126/science.adr9109

Mesoscale volumetric fluorescence imaging at nanoscale resolution by photochemical sectioning

Wei Wang 1,†, Xiongtao Ruan 2,†, Gaoxiang Liu 2,†, Daniel E Milkie 3, Wenping Li 1, Eric Betzig 2,3,4, Srigokul Upadhyayula 2,5,6,*, Ruixuan Gao 1,7,*
PMCID: PMC13617513  NIHMSID: NIHMS2209202  PMID: 41100598

Abstract

Optical nanoscopy of intact biological specimens has been transformed by recent advancements in hydrogel-based tissue clearing and expansion, enabling the imaging of cellular and subcellular structures with molecular contrast. However, existing high-resolution fluorescence microscopes are physically limited by objective-to-specimen distance, which prevents the study of whole-mount specimens without physical sectioning. To address this challenge, we developed a photochemical strategy for spatially precise sectioning of specimens. By combining serial photochemical sectioning with lattice light-sheet imaging and petabyte-scale computation, we imaged and reconstructed axons and myelin sheaths across entire mouse olfactory bulbs at nanoscale resolution. An olfactory-bulb-wide analysis of myelinated and unmyelinated axons revealed distinctive patterns of axon degeneration and de-/dysmyelination in the neurodegenerative brain, highlighting the potential for peta- to exabyte-scale super-resolution studies using this approach.

Attempts to understand human cognition, behavior, and consciousness have catalyzed new technologies to uncover the anatomical complexity and molecular workings of the brain (1, 2). Among these, tissue clearing and expansion microscopy (ExM) have emerged as effective tools for optical investigation of intact brain tissue (3–6). In particular, hydrogel-based tissue clearing and expansion has enabled volumetric mapping of biomolecules down to nanoscale resolution via conventional fluorescence microscopy, while simultaneously minimizing optical scattering and aberration (5–8). Tissue expansion plus light-sheet fluorescence microscopy has created a scalable platform to study synaptic ultrastructure, connectivity, and molecular compositions of invertebrate and rodent brains with high speed, low photobleaching, and 3D nanoscale resolution (9–12).

In ExM, both the expansion factor and the microscope parameters contribute to the overall resolution and scalability. Optimizing both is crucial for achieving isotropic nanoscale resolution, high sensitivity, low photobleaching, and fast imaging across large expanded samples. In expansion lattice light-sheet microscopy (ExLLSM), for example, while imaging speed exceeds 100 million voxels per second with minimal photobleaching and sub-100 nm resolution at a modest ~4-fold expansion (9), the imaging depth is physically limited by the short working distance of the high numerical aperture (NA) objectives used (13). More recent studies have adopted ultra-long-working-distance objectives for large-scale imaging (14), but often at the expense of NA, and hence resolution, for a given expansion factor. Nevertheless, balancing NA and working distance continues to pose an obstacle in scaling up ExM-based methods to whole-mount specimens without sample sectioning.

Electron microscopy (EM) has long faced similar challenges of limited penetration in volumetric imaging. In response, techniques such as serial-section electron microscopy (ssEM) and focused-ion-beam scanning electron microscopy (FIB-SEM) employ mechanical sectioning or surface-ablation followed by on-section or block-face imaging (Fig. 1A) (15–18). While analogous serial sectioning has been applied to volumetric fluorescence imaging of fixed tissue (19–25), mechanical sectioning of hydrogel-expanded specimens often leads to sample distortion, tearing, and loss (fig. S1, fig. S2, Movie S1, Movie S2). Hydrogel stiffening (e.g., increasing monomer concentration, re-embedding, or modifying monomer chemistry) can reduce these problems (12, 26), but the resulting changes in polymer density and composition often lead to reduced imaging resolution (fig. S3) and compromised compatibility with established expansion protocols that require specific hydrogel densities and compositions. Furthermore, in applications such as brain connectomics, sectioning must be near-perfect, as even a single tear or lost section can be catastrophic for tracing and reconstruction of thin processes (27, 28). Yet standard vibratomes and (ultra)microtomes struggle to accommodate large mammalian tissues, especially post-expansion (28).

Fig. 1. Volumetric imaging via photochemical sectioning (VIPS).

Fig. 1.

Open in a new tab

(A) Volumetric imaging via serial section or focused-ion-beam (FIB) as practiced in electron microscopy (EM, left), and volumetric optical imaging via photochemical sectioning (PS) in VIPS (right). The sectioning angle (θ) of photochemical sectioning illumination [hv (PS)] can either be tangent or normal to the block. (B) Schematic of degradation of acrylamide (AA)/sodium acrylate (SA) polymer chains via photocleavage of the crosslinker (photocleavable crosslinker, PC). Inset, detail showing photocleavage of PC under UV illumination [hv (UV)]. (C) Synthesis of PC (compound 3). The second step employs a Passerini reaction for facile synthesis. (D) Pre- and post-expansion images of PC-crosslinked polyacrylamide/sodium polyacrylate hydrogel (PC-gel). Grid size: 5 mm. Inset, PC-gel (labeled with green dye) in water before (i), during (ii), and after (iii) 405 nm laser light-sheet illumination from the side. (E) Pre- (top) and post-405 nm-laser illumination (bottom) images of HEK cells, fluorescently labeled for microtubules (β-tubulin, AF488) and nuclei (SiR-DNA) and expanded with the PC-gel. The illuminated region-of-interest (ROI) at lower left was completely removed. Scale bars: 10 μm (48 μm). Hereafter, all scale bars are at pre-expansion scale, with post-expansion sizes indicated in brackets, unless otherwise noted. (F) Similar experiment for expanded HEK cells with β-tubulin labeling showing removal of the illuminated ROI when expanded with the PC-gel (left column) versus the non-photodegradable, bis-acrylamide-crosslinked polyacrylamide/sodium polyacrylate gel (Bis-gel, control) (right column). Scale bars: 10 μm (48 μm for PC-gel and 45 μm for Bis-gel). (G) Photodegradation time of PC-gel-expanded HEK cells under 405 nm illumination versus PC concentration [n = 3 samples at each concentration, SEM = standard error of the mean, rM = molar ratio of PC to bis-acrylamide used for Bis-gel].

To overcome these challenges, we developed Volumetric Imaging via Photochemical Sectioning (VIPS), a method that extends nanoscale fluorescence imaging to centimeter-scale volumes and beyond. In analogy to FIB-SEM, VIPS employs a spatially precise photochemical degradation process to perform sequential sample degradation and on-block volumetric optical imaging of hydrogel-embedded biological specimens (Fig. 1A, Movie 1). VIPS uses a superabsorbent hydrogel that incorporates a photocleavable crosslinker to rapidly and completely disintegrate under either UV illumination or multiphoton excitation. By combining sequential on-block volumetric lattice light-sheet microscopy (LLSM) imaging, light-sheet photochemical sectioning, and petabyte-scale computation, we imaged and reconstructed axonal projections and myelination sheaths across two complete mouse olfactory bulbs at nanoscale resolution. Finally, we performed a comparative analysis of the wild-type and neurodegenerative mouse olfactory bulb, which revealed unique spatial patterns of axon degeneration and de-/dysmyelination in the neurodegenerative sample at single-axon resolution.

Movie 1.

Download video file (465.3MB, mp4)
Open in a new tab

Animation showing the general concept and workflow of Volumetric Imaging via Photochemical Sectioning (VIPS) in both macroscale and nanoscale views.

Design and synthesis of photodegradable superabsorbent hydrogel

Photoresponsive hydrogels have been used to modulate mechanical and chemical environments around encapsulated cells (29–32). After evaluating various photochemistries for their optical properties and compatibilities with tissue expansion and clearing, we selected polyacrylamide/sodium polyacrylate hydrogel for VIPS. We hypothesized that this polymer, widely used for tissue clearing and expansion, could undergo swift and complete photodegradation by incorporating a photocleavable crosslinker (PC) into its polymer network (Fig. 1B). To enhance the usability and accessibility of VIPS, we prioritized the following design principles for PC: (a) photocleavage at wavelengths distinct from popular fluorophores; (b) bio-orthogonal and water-soluble; (c) low molecular weight for enhanced sample permeability; (d) compatible with established hydrogel recipes by simply replacing the standard crosslinker; (e) synthesizable in a minimal number of steps with common wet-lab equipment.

Following these principles, we designed the molecular structure and synthetic pathway of PC (Fig 1C). This PC molecule, equipped with a ~2 kDa polyethylene glycol (PEG) backbone and two photocleavable o-nitrobenzyl ester moieties, can be synthesized in the ambient environment with standard glassware and wet-lab equipment (Materials and Methods, fig. S4). Using PC as the crosslinker, we then synthesized a photocleavable version of polyacrylamide/sodium polyacrylate hydrogel via free-radical chain-growth polymerization (Fig. 1D). The resulting optically transparent and mechanically elastic hydrogel (“PC-gel”) underwent the same osmotic swelling as its non-photodegradable counterpart (“Bis-gel”, polyacrylamide/sodium polyacrylate hydrogel crosslinked with bis-acrylamide) (5). Consistent with the Bis-gel, the expansion factor of the PC-gel increased as the PC concentration decreased during polymerization (fig. S5) (5). In practice, ~4–5-fold expanded samples were the most convenient to handle because the PC-gel maintained a high mechanical integrity around these expansion factors. Unless otherwise noted, we used close to this optimal expansion factor for all biological sample preparations. The global expansion errors of the PC-gel were within ~0.3–1.3% over 20 μm, comparable to the Bis-gel (fig. S6) (5).

Spatially precise photodegradation of PC-gel

Under continuous UV (365 nm) illumination, bulk PC-gel underwent complete photodegradation, transforming into a disintegrated liquid state (fig. S7). Although PC photocleavage occurs most efficiently around 270–350 nm (29), the more commonly used 405 nm light was also effective, requiring ~80 min at ~0.01 W/cm2 to cleave the gel in water using a ~200 μm thick light sheet [Fig. 1D (inset), Movie 2]. We could also photodegrade PC-gel-expanded cell samples across a defined region-of-interest (ROI) with a confocal microscope (Fig. 1E, Table S1). The Bis-gel controls confirmed that the signal loss within the photodegraded ROI resulted from the diffusion of freed, originally gel-anchored fluorophores, rather than photobleaching (Fig. 1F, fig. S8). Finally, gelled samples with higher crosslinker (PC) density required longer illumination times at the same laser power to complete the photodegradation (Fig. 1G).

Movie 2.

Download video file (12.3MB, mp4)
Open in a new tab

Accelerated video showing spatially controlled photochemical sectioning of PC-gel with a thin sheet of 405 nm laser from the side.

Volumetric fluorescence imaging via two-photon photochemical sectioning

Single-photon widefield or confocal illumination is typically not suitable for photochemical sectioning in a single-objective microscope setup because out-of-focus light passes through the PC-gel and causes undesirable photodegradation outside the intended plane. One option is to exploit the nonlinear optical properties of two-photon (2P) absorption, which photodegrades only those regions through which a raster-scanned 2P-focused spot travels. A large volume can thus be imaged with 2P VIPS by repeated rounds of: (a) imaging the sample to a depth less than the focal depth of the objective; (b) photodegrading or photoslicing via 2P illumination to remove the sample to a depth slightly shallower than that of the imaging (to leave an overlap region for image stitching); and (c) iterating this process until the desired sample volume is imaged (Fig. 2A).

Fig. 2. Volumetric fluorescence imaging of mouse brain tissue at nanoscale resolution with two-photon (2P) VIPS.

Fig. 2.

Open in a new tab

(A) Schematic of the optical setup for 2P VIPS. Imaging beyond the microscope working distance (WD) is achieved by sequential volumetric fluorescence imaging and 2P photochemical sectioning (PS). (B) XY (top) and XZ (bottom) views of a 2P-photodegraded ROI in mouse brain tissue fluorescently labeled for synaptic proteins (Homer1) and expanded with the PC-gel. Scale bars: 50 μm (post-expansion). (C) 3D rendered view of the 2P-photodegraded ROI in B. (D) Profiles of fluorescence intensity along the orange dotted lines in B. (E) Schematic of a mouse brain tissue strip vertically embedded and expanded with the PC-gel for imaging by 2P VIPS. (F) Vertical orthoslice through the tissue strip, imaged to a ~3.8 mm depth (beyond the ~2.6 mm objective WD) via three rounds of sequential on-block volumetric confocal microscopy imaging and 2P photochemical sectioning. Myelin sheaths, labeled with fluorescent myelin basic protein (MBP), are shown. Lines 1–3 indicate the maximum imaging depth for each round. (G) Enlarged views of regions i-iv as denoted in F. Scale bar: 50 μm (post-expansion).

To demonstrate this, we first used a 740 nm 2P laser to photodegrade a 3D ROI within a densely labeled PC-gel-expanded mouse tissue sample and then used single-photon confocal microscopy to image the encompassing volumes post-photodegradation (Fig. 2B) (29). As a result, the lateral and axial photodegradation boundaries matched well with the defined 3D ROI (Fig. 2C–2D). The PC-gel remained unaffected at 920 nm excitation (Fig. S9), demonstrating that traditional 2P imaging of visible-range fluorophores is compatible with 2P VIPS. Next, we applied 2P VIPS to a PC-gel-expanded mouse brain strip (Fig. 2E). We performed three rounds of on-block volumetric confocal microscopy imaging covering depths of ~1.5 mm, ~1.5 mm, and ~1.8 mm, respectively, with each imaging round separated by ~1.0 mm of 2P photodegradation. The three stitched subvolumes spanned a ~3.8 mm thick tissue volume, well beyond the objective’s working distance of ~2.6 mm (Fig. 2F–2G).

Volumetric fluorescence imaging via light-sheet photochemical sectioning

As point scanning methods, volumetric confocal microscopy imaging and 2P photochemical sectioning are too slow and confocal’s propensity for photobleaching is too great for either of them to be easily applied to centimeter-scale volumetric imaging. Light-sheet microscopy negates these issues via confinement of fluorescence excitation to the focal plane and parallel readout of all pixels within this plane. ExLLSM in particular has the requisite speed and low photobleaching for mesoscale imaging at nanoscale resolution (9, 33), albeit limited by the effective working distances of its two objectives. It therefore makes an ideal platform for VIPS (Fig. 3A, fig. S10, Movie 1).

Fig. 3. Volumetric fluorescence imaging of mouse olfactory bulb (OB) at nanoscale resolution with lattice light-sheet microscopy (LLSM) VIPS.

Fig. 3.

Open in a new tab

(A) Schematic of the optical setup. Imaging beyond the WD of the two angled LLSM objectives is achieved by light-sheet photochemical sectioning (PS) with a third objective tangent to the block. (B) Strategy for whole OB imaging showing multiple rounds of overlapped LLSM volumetric imaging and light-sheet PS to ensure lossless reconstruction of the entire imaged volume. (C) (Top) 3D rendered view of a 7-week wild-type (WT) mouse OB fluorescently labeled for axons (NF-200) and myelin sheaths (MBP) (~3.3 by 4.9 by 2.2 mm, pre-expansion, ~6.4 by 9.4 by 4.3 mm, post-expansion, wedge removed to show interior). (Bottom) The approximate location of OB in the mouse brain. Scale bar: 1 mm (1.93 mm). A: anterior; P: posterior; D: dorsal; V: ventral (D) 3D rendered slab view (~80 μm thick, pre-expansion) of the WT mouse OB, downsampled 5-fold in each direction for visualization. Scale bar: 1 mm (1.93 mm). Top inset, 3D view indicating the location in the OB of the imaged slab. Scale bar: 1 mm (1.93 mm). (E-I) Full-resolution maximum intensity projection (MIP) views (~80 μm thick, pre-expansion) of the boxed regions in D near: GL (E); EPL/MCL (F); IPL (G); GCL (H); and LOT (I). Region I is from a different slab view from the one shown in D. Scale bars: 50 μm (96.5 μm). GL: glomerular layer; EPL: external plexiform layer; MCL: mitral cell layer; IPL: internal plexiform layer; GCL: granule cell layer; LOT: lateral olfactory tract.

To achieve spatially precise photochemical sectioning parallel to the sample surface, we integrated a separate Gaussian light-sheet illumination path (fig. S11). Initial benchmark using a ~80 mW 405 nm laser (spread over a uniform ~70–100 μm thick light sheet across a ~20 mm by 20 mm plane, ~0.02 W/cm2) showed that photoslicing of a single ~6 mm by 10 mm PC-gel-embedded sample block-face could be completed in ~20 minutes for ~4–5-fold-expanded samples in water, and in ~2 hours for ~twofold-expanded samples in 1x PBS as a result of the increased crosslinker density (Materials and Methods). Further characterization of sample features in the photosliced plane showed uniform surface flatness, minor distortion on par with the baseline expansion errors, and negligible photobleaching from the photoslicing process itself (fig. S12, fig. S13, fig. S14).

We then validated the capabilities of LLSM VIPS on a ~4.4-fold expanded human hippocampus tissue sample labeled for axons and myelin sheaths. We performed eight rounds of sequential volumetric LLSM imaging followed by photoslicing across a ~0.25 mm by 0.23 mm by 1.6 mm volume—achieving five times the depth attainable without VIPS (fig. S15). Imaging and photoslicing times totaled ~15 minutes and ~2 hours, respectively, with each photosliced top gel block floated off and removed from the imaging bath immediately. The imaged subvolumes were then deconvolved, registered, and stitched using PetaKit5D with an effective Nyquist-sampled voxel size of ~22 nm by 22 nm by 59 nm (34). In the stitched volume, we were able to trace and reconstruct individual axons with diameters as small as ~150 nm, confirming the preservation of connectivity and traceability of anatomical features across multiple rounds of imaging and photochemical sectioning, and providing the motivation to tackle even larger volumes.

Mapping axons and myeline sheaths in whole mouse olfactory bulb

The mammalian olfactory bulb (OB) is a complex neural structure essential for processing and relaying olfactory information. To demonstrate the scalability of LLSM VIPS, we performed two-color imaging of an entire 7-week wild-type (WT) mouse OB at ~twofold expansion (~56 nm by 56 nm by 135 nm effective voxel size). Although modest, this expansion factor yielded resolution sufficient to trace and reconstruct most axons in the mouse brain, as estimated from volume EM (35), and avoids the tens of petabytes of data for the raw, intermediate, and fully processed volumes that would be necessary at higher expansion factors.

We imaged the entire expanded OB (~twofold expanded to a volume of ~6.4 mm by 9.4 mm by 4.3 mm) at peak acquisition rates of ~2 teravoxels per hour over 10 days, utilizing nine rounds of interleaved on-block volumetric LLSM imaging and photoslicing (Fig. 3B). Imaging and photoslicing times totaled ~142 hours and ~20 hours, respectively, and yielded 17,280 volumetric tiles, constituting ~0.5 petabytes of raw data. We reconstructed the dataset in two stages. First, parallelized PetaKit5D processing handled deskewing, rotation, deconvolution, and stitching to rebuild the full OB (Fig. 3C, Movie 3) (34). Second, PetaVIPS, an extension of PetaKit5D, enabled scalable and parallel downstream workflows, from axon and myelin segmentation to tractography analysis (Materials and Methods) (36).

Movie 3.

Download video file (853.4MB, mp4)
Open in a new tab

Volumetric rendering showing a 7-week wild-type (WT) mouse olfactory bulb (OB) imaged with lattice light-sheet microscopy (LLSM) VIPS.

The reconstructed WT mouse OB dataset revealed distinct axonal projections and myelination patterns across all its major anatomical layers (37) (Fig. 3D). For example, detailed 3D visualization highlighted the unique axonal and myelination patterns within individual glomeruli of the glomerular layer (GL), where the axonal shafts formed a dense, peripheral basket encompassing the central neuropil (Fig. 3E). The vertical axonal projections of tufted cells traversed through the external plexiform layer (EPL) and mitral cell layer (MCL) (Fig. 3F), while the dense lateral axonal projections spanned the internal plexiform layer (IPL) (Fig. 3G). The axonal bundles from the mitral and tufted cells were observed passing through the granule cell layer (GCL) (Fig. 3H), exiting the olfactory bulb, and projecting through the lateral olfactory tract to the olfactory cortex (Fig. 3I). Overall, axons travelling through the GL and EPL/MCL exhibited high tortuosity, while those in the deeper layers displayed strong directionality. Furthermore, the nanoscale resolution of the dataset enabled segmentation, skeletonization, and reconstruction of individual myelinated and unmyelinated axon fibers, allowing for precise measurements such as calculating the total lengths of segmentable axons throughout the entirety of mouse OB (Fig. 4A–4B, fig. S16, fig. S17).

Fig. 4. Olfactory-bulb (OB)-wide analysis of axons and myelin sheaths at single-axon resolution.

Fig. 4.

Open in a new tab

(A) Total length of segmented axons, myelinated axons, and unmyelinated axons across a WT mouse OB (excluding the ONL). (B) Total length of segmented axons in distinct anatomical layers. (C) Full-resolution 3D rendered views of representative ROIs in GL, EPL/MCL, and IPL of 7-week WT (top) and NPC1 (Npc1−/−) (bottom) mouse OBs. NF-200, MBP, and composite views of the same region are shown in separate columns. Scale bars: 50 μm (96.5 μm). A: anterior; P: posterior; D: dorsal; V: ventral. (D) Comparison of the volume density of segmented axons across the entirety of and in distinct anatomical layers of the WT and NPC1 mouse OBs. (E) Similar comparison of segmented myelin sheaths and myelinated axons across the two OBs. (F) Percentage of myelinated axons to all segmented axons across distinct anatomical layers of the two OBs. (G) Box plots of the percentage of myelinated axons across ~50 ROIs (~66 by 66 by 87 μm, pre-expansion) evenly sampled across distinct anatomical layers of the WT and NPC1 mouse OBs. Whisker ends denote maximum and minimum values, dots denote individual values, and the upper, middle, box lines denote 75th, 50th (median), and 25th percentile, respectively. ONL: olfactory nerve layer; GL: glomerular layer; EPL: external plexiform layer; MCL: mitral cell layer; IPL: internal plexiform layer; GCL: granule cell layer.

Axon degeneration and de-/dysmyelination in a neurodegenerative mouse olfactory bulb

Axonal degeneration and de-/dysmyelination, which often precede clinical symptoms, are hallmarks of neurodegeneration and aging. In the fatal neurodegenerative lysosomal storage disorder Niemann-Pick type C1 (NPC1) disease, mutations in the NPC1 gene cause early-stage dysmyelination (38). Pathophysiological changes in the NPC1 animal’s olfactory system represent an early indicator of disease progression (39, 40), which is also characterized by the reduced myelin basic protein (MBP) expression in the OB at the neonate and adolescent stage (41). However, the precise mechanism behind de-/dysmyelination remains unclear, particularly regarding whether the reduced expression of MBP in the OB persists into adulthood and results in anatomical loss at the single axon/myelin sheath level. Furthermore, the progression of axon and myelin sheath loss across the distinct layers of OB remains unexplored.

To study axon degeneration and de-/dysmyelination in the neurodegenerative brain, we imaged a 7-week NPC1 (Npc1−/−) mouse OB (~twofold expanded to a volume of ~6.0 mm by 11.2 mm by 3.6 mm; ~0.6 petabytes of raw data) (Movie 4). The expanded and reconstructed WT and NPC1 mouse OB volumes enabled direct comparison of axon and myelin sheath morphologies across the entire OB (Movie 3, Movie 4, Fig. 4C), which revealed distinct anatomical changes in the NPC1 OB. These included apparent reduction in axon and myelin sheath densities in the GL, EPL/MCL, and IPL, indicative of potential axon degeneration and de-/dysmyelination in these OB layers (Fig. 4C).

Movie 4.

Download video file (1,004.8MB, mp4)
Open in a new tab

Volumetric rendering showing a 7-week NPC1 (Npc1−/−) mouse OB imaged with LLSM VIPS.

An OB-wide quantitative analysis of axon fibers revealed a reduced volume density of segmented axons in the NPC1 OB, consistent with previous studies showing early-onset axonal degeneration across all major parts of the NPC1 brain (Fig. 4D) (42–44). The severity of axon degeneration varied across different OB layers, with the EPL/MCL exhibiting the greatest reduction (-56.6%) and the IPL showing the least (-29.6%). An exception was in the GCL, which exhibited an increase in axon density (+18.6%) (Fig. 4D). We hypothesize this may reflect compensatory axonal sprouting in the GCL following widespread neuronal loss (45). Given that axonal damage often precedes loss of cell bodies (43, 46), which can be located in distant anatomical layers and brain regions, these results highlight VIPS’s capability to capture the local variability of axonal integrity and survival both within and across the OB layers.

We also compared myelination profiles of the WT and NPC1 mouse OBs. The density of total segmented myelin sheaths and myelinated axons was reduced throughout the NPC1 OB (Fig. 4E), affirming that early-stage dysmyelination persists as the animal develops (41). Difference in the percentage of myelinated axons to all axons also varied across the layers. Only marginal or no decreases from the WT to the NPC1 OB were observed in the GL and the EPL/MCL (Fig. 4F), indicating a similar rate of loss for both myelinated and unmyelinated axons in these layers. In contrast, the IPL and GCL showed decreases of 36.2% and 46.9%, respectively, suggesting that myelinated axons in these layers may have experienced greater loss than unmyelinated ones (47). Such spatial heterogeneity was also pronounced within each layer and myelination characteristics varied widely across the spatially sampled ROIs (Fig. 4G, fig. S18).

Olfactory-bulb-wide topography and tractography

Leveraging the nanoscale resolution and whole-volume continuity provided by VIPS, we performed single-axon topography and tractography to evaluate axonal connectivity across the WT and NPC1 OB volumes. In the WT OB, the number of segmented axons traversing between adjacent layers remained constant throughout the bulb (Fig. 5A). In the NPC1 OB, however, progressively fewer segmented axons traveled from the IPL-GCL to the EPL/MCL-IPL and then to the GL-EPL/MCL interfaces, indicating that degeneration begins in the outer layers and advances inward.

Fig. 5. Olfactory-bulb (OB)-wide topography and tractography at single-axon resolution.

Fig. 5.

Open in a new tab

(A) Total number of segmented axons crossing distinct anatomical layers in 7-week WT (left) and NPC1 (Npc1−/−) (right) mouse OBs. Axon numbers are normalized to those crossing between the IPL and GCL in each OB. (B, C) 3D rendered slab views (~80 μm thick, pre-expansion) of azimuthal (ϕ) angles of segmented axons in the two OBs, downsampled 10-fold in each direction for visualization. Scale bars: 200 μm (386 μm). (D, E) Frequency histograms of azimuthal (ϕ) angles of all axons segmented across the entirety of the two OBs (excluding the ONL). (F, G) Frequency histograms of azimuthal (ϕ) angles of all axons segmented across distinct anatomical layers of the two OBs. (H, I) Frequency histograms and full-resolution 3D rendered views of azimuthal (ϕ) angles of segmented axons in representative ROIs in GL, EPL/MCL, and IPL of the two OBs. Scale bars: 50 μm (96.5 μm) and 10 μm (19.3 μm) for the enlarged boxed regions. A: anterior; P: posterior; D: dorsal; V: ventral. ONL: olfactory nerve layer; GL: glomerular layer; EPL: external plexiform layer; MCL: mitral cell layer; IPL: internal plexiform layer; GCL: granule cell layer.

Single-axon tractography enabled visualization (Fig. 5B–5C, fig. S19, fig. S20, Movie 5) and quantitative analysis (Fig. 5D–5E, fig. S21) of the 3D organization of the segmented axons across the entire OB, as well as within specific OB layers (Fig. 5F–5G, fig. S21) and localized regions (Fig. 5H–5I, fig. S22). Within each layer, axon orientations shifted from a ventral-dorsal axis in the outer layer (e.g., GL) to an anterior-posterior axis in the deeper layer (e.g., GCL), reflecting the shifting paths of information flow (Fig. 5F–5I, fig. S21). Compared to the WT OB, the NPC1 OB showed most prominent reduction in axons projecting along the ventral-dorsal axis within the IPL, suggesting that the degenerative process can be selective for axons with specific orientations. Within this overall architecture, however, there existed substantial local heterogeneity, with axons projecting in many directions, even if the average orientations in each region followed an overall trend (Fig. 5F–5I).

Movie 5.

Download video file (346.4MB, mp4)
Open in a new tab

Volumetric rendering showing the azimuthal (ϕ) angles of segmented axons in the 7-week WT and NPC1 mouse OBs.

Finally, in the WT OB, the myelinated axon orientations mirrored those of the total axon population, while in the IPL and GCL of the NPC1 OB, the myelinated axons were strongly biased along the anterior-posterior axis (Fig. 5F–5G, fig. S21). This suggests that the remaining axons in these layers, mostly myelinated, project primarily to or from the olfactory cortex via the lateral olfactory tract (48).

Discussion

VIPS overcomes the working distance limits of high-resolution optical microscopy by repeated rounds of in-situ photochemical sectioning and imaging of a hydrogel-embedded sample. This non-contact process enables precise sectioning compatible with various hydrogel-expansion chemistries, and circumvents the distortion, sample loss, and protocol limitations of mechanical sectioning (Supplementary note 1). Furthermore, the process can be easily integrated into existing microscope setups, enabling continuous automated acquisition. When combined with LLSM imaging (33) and a high-performance computational pipeline (34, 36), VIPS effectively eliminates physical sample size as a barrier to mesoscale imaging of tissues at nanoscale resolution.

The large volumetric dynamic range of VIPS reveals anatomical heterogeneities, gradients, and biological insights that are undetectable when sampling smaller volumes lacking global context or spatial resolution (Supplementary note 2). Although only a single WT and NPC1 pair of OBs were compared in our study, a dedicated pipeline based on the sample preparation, imaging, and computation methods demonstrated here has the potential bandwidth to image multiple specimens to address quantitatively the degree of stereotypy and to explore different molecular markers to achieve more biologically meaningful comparisons and overall understanding.

Extending VIPS to the petabyte scale has exposed a series of technical challenges (Supplementary notes 3 and 4). First, labeling density and specificity proved limiting even at only twofold expansion, where our segmented count of thin unmyelinated axons in the lateral olfactory tract was only ~14% of that tallied in a prior EM study of the same pathway, albeit at different animal age (49). Dense and uniform labeling may prove even more challenging in larger specimens. Second, nonlinear warping observed at subvolume boundaries, which might be expected to increase at higher expansion factors, exposes the need for scalable, non-rigid registration pipelines capable of handling peta- to exavoxel data. Third, even with real-time compression, single samples may exceed a petabyte; storage costs, data processing, analysis, and manual curation may set practical limits on the number and size of specimens long before the limits of imaging are reached. Finally, 3D petabyte-scale data is far too vast to be mined effectively and comprehensively for novel insights by humans alone. AI systems will be needed to replace manual annotation and to extract quantitative information at scale under human queries. Progress in labeling chemistry, AI/GPU-accelerated non-rigid solvers and segmentation, compute costs, and AI vision are already narrowing these gaps.

Despite these challenges, by greatly extending the tissue volumes approachable with nanoscale fluorescence microscopy, VIPS takes a major step in advancing from studies that attempt to derive general insights from sparse subsampling of large tissues to quantitative measurement of biological stereotypy and variability and the derivation of fundamental insights from tissues in toto. Indeed, coupled with pan-membrane or protein labeling, VIPS might form the foundation for an optical pipeline to deliver synaptic-level dense connectomes of whole brains (50, 51).

Materials and methods are available in the supplementary materials.

Supplementary Material

Supplementary Materials
Movie S1
Download video file (42.4MB, mp4)
Movie S2
Download video file (37.2MB, mp4)

Materials and Methods

Supplementary Notes

Figs. S1 to S26

Table S1

Movies S1 to S2

References (52–64)

Acknowledgments:

We thank J. Zak for discussion on mouse olfactory system and assistance with data analysis. We thank S. Cologna for mouse sample preparation, A. Schantz and D. Keene for human sample collection, J. Sun, W. Cho, and L.-W. Gong for cell cultures, Y. Wang for assistance with the macroscale light-sheet photochemical sectioning experiment, and R. Torrez, J. Iwasa, and S. Kwon for assistance with animation and scientific illustration. We thank the Center for Advanced Microscopy & Nikon Imaging Center at Northwestern University and the Chicago Biomedical Consortium (CBC) for access to fluorescence microscopy facilities. We thank the NVIDIA IndeX team for sharing the NVIDIA Index software. We thank J. White for managing the ABC computing cluster. We gratefully acknowledge the support of this work by the Laboratory Directed Research and Development (LDRD) Program of Lawrence Berkeley National Laboratory under US Department of Energy contract No. DE-AC02- 05CH11231. This research used resources of the National Energy Research Scientific Computing Center (NERSC), a U.S. Department of Energy Office of Science User Facility located at Lawrence Berkeley National Laboratory, operated under Contract No. DE-AC02-05CH11231 using NERSC award DDR-ERCAP0025501.

Funding:

National Institutes of Health grant UG3MH126864 (RG, co-I)

University of Illinois Chicago Startup Fund (RG)

Philomathia Foundation Gift (SU, EB)

Lawrence Berkeley National Laboratory’s LDRD Program grants 7647437 & 7721359 (SU)

HHMI (EB)

Chan Zuckerberg Initiative Imaging Scientist program grants 2019-198142 & 2021-244163 (SU)

Chan Zuckerberg Biohub - San Francisco Investigator gift funding (SU)

Sloan Foundation grant G-2022-19390 (SU)

Footnotes

Competing interests: RG is a co-inventor of multiple patents related to expansion microscopy (U.S. Patents 11,873,374 and 12,233,184). The other authors declare that they have no competing interests. Portions of the technology described herein are covered by U.S. Patent 7,894,136 issued to EB, assigned to Lattice Light of Ashburn, VA, and licensed to Carl Zeiss Microscopy; U.S. Patents 8,711,211 and 9,477,074 issued to EB, assigned to HHMI, and licensed to Carl Zeiss Microscopy; and U.S. Patent 9,500,846 issued to EB. Portions of the technology described herein are filed for patent (U.S. App. No. 63/678,132).

Data and materials availability:

Custom-synthesized materials and reagents are available for researchers within the capacity of RG lab through a simple interinstitutional materials transfer (MTA) agreement (such as a UBMTA). Custom scripts used for petabyte-scale image data processing and analysis (“PetaVIPS”) are available on GitHub at https://github.com/abcucberkeley/PetaVIPS. The version associated with this manuscript is available on Zenodo (https://doi.org/10.5281/zenodo.16929810) (36). The uncompressed raw and processed data for this paper exceeds a petabyte in total and thus is not practical to be hosted on currently available public repositories. All data used in this manuscript are available upon request with a feasible data transfer and storage mechanism (such as Globus endpoint, or cloud buckets). All other data are available in the main text or the supplementary materials.

References and Notes

  • 1.Kirst C, Skriabine S, Vieites-Prado A, Topilko T, Bertin P, Gerschenfeld G, Verny F, Topilko P, Michalski N, Tessier-Lavigne M, Renier N, Mapping the fine-scale organization and plasticity of the brain vasculature. Cell 180, 780–795.e25 (2020). [DOI] [PubMed] [Google Scholar]
  • 2.Zhou C, Yang X, Wu S, Zhong Q, Luo T, Li A, Liu G, Sun Q, Luo P, Deng L, Ni H, Tan C, Yuan J, Luo Q, Hu X, Li X, Gong H, Continuous subcellular resolution three-dimensional imaging on intact macaque brain. Sci Bull (Beijing) 67, 85–96 (2022). [DOI] [PubMed] [Google Scholar]
  • 3.Richardson DS, Lichtman JW, Clarifying Tissue Clearing. Cell 162, 246–257 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ueda HR, Ertürk A, Chung K, Gradinaru V, Chédotal A, Tomancak P, Keller PJ, Tissue clearing and its applications in neuroscience. Nat Rev Neurosci 21, 61–79 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Chen F, Tillberg PW, Boyden ES, Expansion microscopy. Science 347, 543–548 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wang W, Chan YH, Kwon S, Tandukar J, Gao R, Nanoscale fluorescence imaging of biological ultrastructure via molecular anchoring and physical expansion. Nano Converg 9, 30 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Gradinaru V, Treweek J, Overton K, Deisseroth K, Hydrogel-tissue chemistry: Principles and applications. Annu Rev Biophys 47, 355–376 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Choi SW, Guan W, Chung K, Basic principles of hydrogel-based tissue transformation technologies and their applications. Cell 184, 4115–4136 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Gao R, Asano SM, Upadhyayula S, Pisarev I, Milkie DE, Liu T-L, Singh V, Graves A, Huynh GH, Zhao Y, Bogovic J, Colonell J, Ott CM, Zugates C, Tappan S, Rodriguez A, Mosaliganti KR, Sheu S-H, Pasolli HA, Pang S, Xu CS, Megason SG, Hess H, Lippincott-Schwartz J, Hantman A, Rubin GM, Kirchhausen T, Saalfeld S, Aso Y, Boyden ES, Betzig E, Cortical column and whole-brain imaging with molecular contrast and nanoscale resolution. Science 363, eaau8302 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Daetwyler S, Fiolka RP, Light-sheets and smart microscopy, an exciting future is dawning. Commun Biol 6, 502 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Lillvis JL, Otsuna H, Ding X, Pisarev I, Kawase T, Colonell J, Rokicki K, Goina C, Gao R, Hu A, Wang K, Bogovic J, Milkie DE, Meienberg L, Mensh BD, Boyden ES, Saalfeld S, Tillberg PW, Dickson BJ, Rapid reconstruction of neural circuits using tissue expansion and light sheet microscopy. Elife 11, e81248 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Tian X, Lin T-Y, Lin P-T, Tsai M-J, Chen H, Chen W-J, Lee C-M, Tu C-H, Hsu J-C, Hsieh T-H, Tung Y-C, Wang C-K, Lin S, Chu L-A, Tseng F-G, Hsueh Y-P, Lee C-H, Chen P, Chen B-C, Rapid lightsheet fluorescence imaging of whole Drosophila brains at nanoscale resolution by potassium acrylate-based expansion microscopy. Nat Commun 15, 10911 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Zhang Y, Gross H, Systematic design of microscope objectives. Part I: System review and analysis. 8, 313–347 (2019). [Google Scholar]
  • 14.Glaser A, Chandrashekar J, Vasquez J, Arshadi C, Ouellette N, Jiang X, Baka J, Kovacs G, Woodard M, Seshamani S, Cao K, Clack N, Recknagel A, Grim A, Balaram P, Turschak E, Liddell A, Rohde J, Hellevik A, Takasaki K, Barner LE, Logsdon M, Chronopoulos C, de Vries S, Ting J, Perlmutter S, Kalmbach B, Dembrow N, Reid RC, Feng D, Svoboda K, Expansion-assisted selective plane illumination microscopy for nanoscale imaging of centimeter-scale tissues. Elife 12, RP91979 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yin W, Brittain D, Borseth J, Scott ME, Williams D, Perkins J, Own C, Murfitt M, Torres RM, Kapner D, Bleckert A, Castelli D, Reid D, Lee W-CA, Graham BJ, Takeno M, Bumbarger DJ, Farrell C, Reid RC, da Costa NM, A petascale automated imaging pipeline for mapping neuronal circuits with high-throughput transmission electron microscopy. Nat Commun 11, 4949 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Scheffer LK, Xu CS, Januszewski M, Lu Z, Takemura S, Hayworth KJ, Huang GB, Shinomiya K, Maitlin-Shepard J, Berg S, Clements J, Hubbard PM, Katz WT, Umayam L, Zhao T, Ackerman D, Blakely T, Bogovic J, Dolafi T, Kainmueller D, Kawase T, Khairy KA, Leavitt L, Li PH, Lindsey L, Neubarth N, Olbris DJ, Otsuna H, Trautman ET, Ito M, Bates AS, Goldammer J, Wolff T, Svirskas R, Schlegel P, Neace E, Knecht CJ, Alvarado CX, Bailey DA, Ballinger S, Borycz JA, Canino BS, Cheatham N, Cook M, Dreher M, Duclos O, Eubanks B, Fairbanks K, Finley S, Forknall N, Francis A, Hopkins GP, Joyce EM, Kim S, Kirk NA, Kovalyak J, Lauchie S, Lohff A, Maldonado C, Manley EA, McLin S, Mooney C, Ndama M, Ogundeyi O, Okeoma N, Ordish C, Padilla N, Patrick CM, Paterson T, Phillips EE, Phillips EM, Rampally N, Ribeiro C, Robertson MK, Rymer JT, Ryan SM, Sammons M, Scott AK, Scott AL, Shinomiya A, Smith C, Smith K, Smith NL, Sobeski MA, Suleiman A, Swift J, Takemura S, Talebi I, Tarnogorska D, Tenshaw E, Tokhi T, Walsh JJ, Yang T, Horne JA, Li F, Parekh R, Rivlin PK, Jayaraman V, Costa M, Jefferis GSXE, Ito K, Saalfeld S, George R, Meinertzhagen I, Rubin GM, Hess HF, Jain V, Plaza SM, A connectome and analysis of the adult Drosophila central brain. Elife 9, e57443 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Peddie CJ, Genoud C, Kreshuk A, Meechan K, Micheva KD, Narayan K, Pape C, Parton RG, Schieber NL, Schwab Y, Titze B, Verkade P, Weigel A, Collinson LM, Volume electron microscopy. Nature Reviews Methods Primers 2, 51 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Shapson-Coe A, Januszewski M, Berger DR, Pope A, Wu Y, Blakely T, Schalek RL, Li PH, Wang S, Maitin-Shepard J, Karlupia N, Dorkenwald S, Sjostedt E, Leavitt L, Lee D, Troidl J, Collman F, Bailey L, Fitzmaurice A, Kar R, Field B, Wu H, Wagner-Carena J, Aley D, Lau J, Lin Z, Wei D, Pfister H, Peleg A, Jain V, Lichtman JW, A petavoxel fragment of human cerebral cortex reconstructed at nanoscale resolution. Science (1979) 384, eadk4858 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Gong H, Xu D, Yuan J, Li X, Guo C, Peng J, Li Y, Schwarz LA, Li A, Hu B, Xiong B, Sun Q, Zhang Y, Liu J, Zhong Q, Xu T, Zeng S, Luo Q, High-throughput dual-colour precision imaging for brain-wide connectome with cytoarchitectonic landmarks at the cellular level. Nat Commun 7, 12142 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Winnubst J, Bas E, Ferreira TA, Wu Z, Economo MN, Edson P, Arthur BJ, Bruns C, Rokicki K, Schauder D, Olbris DJ, Murphy SD, Ackerman DG, Arshadi C, Baldwin P, Blake R, Elsayed A, Hasan M, Ramirez D, Dos Santos B, Weldon M, Zafar A, Dudman JT, Gerfen CR, Hantman AW, Korff W, Sternson SM, Spruston N, Svoboda K, Chandrashekar J, Reconstruction of 1,000 projection neurons reveals new cell types and organization of long-range connectivity in the mouse brain. Cell 179, 268–281.e13 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yu W, Kang L, Tsang VTC, Zhang Y, Wong IHM, Wong TTW, Translational rapid ultraviolet-excited sectioning tomography for whole-organ multicolor imaging with real-time molecular staining. Elife 11, e81015 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yi Y, Li Y, Zhang S, Men Y, Wang Y, Jing D, Ding J, Zhu Q, Chen Z, Chen X, Li J-L, Wang Y, Wang J, Peng H, Zhang L, Luo W, Feng JQ, He Y, Ge W-P, Zhao H, Mapping of individual sensory nerve axons from digits to spinal cord with the transparent embedding solvent system. Cell Res 34, 124–139 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Xu F, Shen Y, Ding L, Yang C-Y, Tan H, Wang H, Zhu Q, Xu R, Wu F, Xiao Y, Xu C, Li Q, Su P, Zhang LI, Dong H-W, Desimone R, Xu F, Hu X, Lau P-M, Bi G-Q, High-throughput mapping of a whole rhesus monkey brain at micrometer resolution. Nat Biotechnol 39, 1521–1528 (2021). [DOI] [PubMed] [Google Scholar]
  • 24.Tsai PS, Friedman B, Ifarraguerri AI, Thompson BD, Lev-Ram V, Schaffer CB, Xiong Q, Tsien RY, Squier JA, Kleinfeld D, All-optical histology using ultrashort laser pulses. Neuron 39, 27–41 (2003). [DOI] [PubMed] [Google Scholar]
  • 25.Shi M-Y, Yao Y, Wang M, Yang Q, Ding L, Li R, Li Y, Huang H, Yang C-Y, Zhou Z, Zhu Z, Wen P, Dai F, Zeng X, Zhang K-M, Guo Y, Sun Z-A, Xia H, Ren Z, Cakmak YO, Zhang M, Xu F, Qu L, Zhu Q, Lau P-M, Xu C, Bi G-Q, High-speed mapping of whole-mouse peripheral nerves at subcellular resolution. Cell 188, 3897–3915.e20 (2025). [DOI] [PubMed] [Google Scholar]
  • 26.Chen R, Cheng X, Zhang Y, Yang X, Wang Y, Liu X, Zeng S, Expansion tomography for large volume tissue imaging with nanoscale resolution. Biomed Opt Express 12, 5614–5628 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Torres R, Takasaki K, Gliko O, Laughland C, Yu W-Q, Turschak E, Hellevik A, Balaram P, Perlman E, Sumbul U, Reid C, A scalable and modular computational pipeline for axonal connectomics: automated tracing and assembly of axons across serial sections. bioRxiv, 2024.06.11.598365 (2024). [Google Scholar]
  • 28.Park J, Wang J, Guan W, Gjesteby LA, Pollack D, Kamentsky L, Evans NB, Stirman J, Gu X, Zhao C, Marx S, Kim ME, Choi SW, Snyder M, Chavez D, Su-Arcaro C, Tian Y, Park CS, Zhang Q, Yun DH, Moukheiber M, Feng G, Yang XW, Keene CD, Hof PR, Ghosh SS, Frosch MP, Brattain LJ, Chung K, Integrated platform for multiscale molecular imaging and phenotyping of the human brain. Science (1979) 384, eadh9979 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kloxin AM, Kasko AM, Salinas CN, Anseth KS, Photodegradable hydrogels for dynamic tuning of physical and chemical properties. Science (1979) 324, 59–63 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Yanagawa F, Sugiura S, Takagi T, Sumaru K, Camci-Unal G, Patel A, Khademhosseini A, Kanamori T, Activated-ester-type photocleavable crosslinker for preparation of photodegradable hydrogels using a two-component mixing reaction. Adv Healthc Mater 4, 246–254 (2015). [DOI] [PubMed] [Google Scholar]
  • 31.Scheiger JM, Li S, Brehm M, Bartschat A, Theato P, Levkin PA, Inherently UV photodegradable poly(methacrylate) gels. Adv Funct Mater 31, 2105681 (2021). [Google Scholar]
  • 32.Günay KA, Chang T-L, Skillin NP, V Rao V, Macdougall LJ, Cutler AA, Silver JS, Brown TE, Zhang C, Yu C.-C. (Jay), Olwin BB, Boyden ES, Anseth KS, Photo-expansion microscopy enables super-resolution imaging of cells embedded in 3D hydrogels. Nat Mater, doi: 10.1038/s41563-023-01558-5 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Fu T-M, Liu G, Milkie DE, Ruan X, Görlitz F, Shi Y, Ferro V, Divekar NS, Wang W, York HM, Kilic V, Mueller M, Liang Y, Daugird TA, Gacha-Garay MJ, Larkin KA, Adikes RC, Harrison N, Shirazinejad C, Williams S, Nourse JL, Sheu S-H, Gao L, Li T, Mondal C, Achour K, Hercule W, Stabley D, Emmerich K, Dong P, Drubin D, Liu ZJ, Clapham D, Mumm JS, Koyama M, Killilea A, Bravo-Cordero JJ, Keene CD, Luo L, Kirchhausen T, Pathak MM, Arumugam S, Nunez JK, Gao R, Matus DQ, Martin BL, Swinburne IA, Betzig E, Legant WR, Upadhyayula S, A multimodal adaptive optical microscope for in vivo imaging from molecules to organisms. bioRxiv, 2025.06.02.657494 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Ruan X, Mueller M, Liu G, Görlitz F, Fu T-M, Milkie DE, Lillvis JL, Kuhn A, Gan Chong J, Hong JL, Herr CYA, Hercule W, Nienhaus M, Killilea AN, Betzig E, Upadhyayula S, Image processing tools for petabyte-scale light sheet microscopy data. Nat Methods 21, 2342–2352 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Tomassy GS, Berger DR, Chen HH, Kasthuri N, Hayworth KJ, Vercelli A, Seung HS, Lichtman JW, Arlotta P, Distinct profiles of myelin distribution along single axons of pyramidal neurons in the neocortex. Science 344, 319–324 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Ruan X, Betzig E, Upadhyayula S, abcucberkeley/PetaVIPS: v1.0.0 (2025). 10.5281/zenodo.16929810. [DOI]
  • 37.Imamura F, Ito A, LaFever BJ, Subpopulations of projection neurons in the olfactory bulb. Front Neural Circuits 14, 561822 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Yu T, Lieberman AP, Npc1 acting in neurons and glia is essential for the formation and maintenance of CNS myelin. PLoS Genet 9, e1003462- (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Hovakimyan M, Meyer A, Lukas J, Luo J, Gudziol V, Hummel T, Rolfs A, Wree A, Witt M, Olfactory deficits in Niemann-Pick Type C1 (NPC1) disease. PLoS One 8, e82216- (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Meyer A, Gläser A, Bräuer AU, Wree A, Strotmann J, Rolfs A, Witt M, Olfactory performance as an indicator for protective treatment effects in an animal model of neurodegeneration. Front Integr Neurosci 12 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Yan X, Lukas J, Witt M, Wree A, Hübner R, Frech M, Köhling R, Rolfs A, Luo J, Decreased expression of myelin gene regulatory factor in Niemann-Pick type C 1 mouse. Metab Brain Dis 26, 299–306 (2011). [DOI] [PubMed] [Google Scholar]
  • 42.Ong W-Y, Kumar U, Switzer RC, Sidhu A, Suresh G, Hu C-Y, Patel SC, Neurodegeneration in Niemann-Pick type C disease mice. Exp Brain Res 141, 218–231 (2001). [DOI] [PubMed] [Google Scholar]
  • 43.Takikita S, Fukuda T, Mohri I, Yagi T, Suzuki K, Perturbed myelination process of premyelinating oligodendrocyte in Niemann-PickType C mouse. J Neuropathol Exp Neurol 63, 660–673 (2004). [DOI] [PubMed] [Google Scholar]
  • 44.Pressey SNR, Smith DA, Wong AMS, Platt FM, Cooper JD, Early glial activation, synaptic changes and axonal pathology in the thalamocortical system of Niemann–Pick type C1 mice. Neurobiol Dis 45, 1086–1100 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Marshall KL, Farah MH, Axonal regeneration and sprouting as a potential therapeutic target for nervous system disorders. Neural Regen Res 16 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Patel SC, Suresh S, Kumar U, Hu CY, Cooney A, Blanchette-Mackie EJ, Neufeld EB, Patel RC, Brady RO, Patel YC, Pentchev PG, Ong W-Y, Localization of Niemann–Pick C1 protein in astrocytes: Implications for neuronal degeneration in Niemann–Pick type C disease. Proceedings of the National Academy of Sciences 96, 1657–1662 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Fünfschilling U, Supplie LM, Mahad D, Boretius S, Saab AS, Edgar J, Brinkmann BG, Kassmann CM, Tzvetanova ID, Möbius W, Diaz F, Meijer D, Suter U, Hamprecht B, Sereda MW, Moraes CT, Frahm J, Goebbels S, Nave K-A, Glycolytic oligodendrocytes maintain myelin and long-term axonal integrity. Nature 485, 517–521 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Brunert D, Rothermel M, Extrinsic neuromodulation in the rodent olfactory bulb. Cell Tissue Res 383, 507–524 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Brunjes PC, Kay RB, Arrivillaga JP, The mouse olfactory peduncle. Journal of Comparative Neurology 519, 2870–2886 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Karagiannis ED, Kang JS, Shin TW, Emenari A, Asano S, Lin L, Costa EK, Marblestone AH, Kasthuri N, Boyden ES, Expansion microscopy of lipid membranes. bioRxiv, 829903 (2019). [Google Scholar]
  • 51.Tavakoli MR, Lyudchik J, Januszewski M, Vistunou V, Agudelo Dueñas N, Vorlaufer J, Sommer C, Kreuzinger C, Oliveira B, Cenameri A, Novarino G, Jain V, Danzl JG, Light-microscopy-based connectomic reconstruction of mammalian brain tissue. Nature 642, 398–410 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Asano SM, Gao R, Wassie AT, Tillberg PW, Chen F, Boyden ES, Expansion Microscopy: Protocols for imaging proteins and RNA in cells and tissues. Curr Protoc Cell Biol 80, e56 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Damstra HGJ, Mohar B, Eddison M, Akhmanova A, Kapitein LC, Tillberg PW, Visualizing cellular and tissue ultrastructure using Ten-fold Robust Expansion Microscopy (TREx). Elife 11, e73775 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Chen BC, Legant WR, Wang K, Shao L, Milkie DE, Davidson MW, Janetopoulos C, Wu XS, Hammer JA, Liu Z, English BP, Mimori-Kiyosue Y, Romero DP, Ritter AT, Lippincott-Schwartz J, Fritz-Laylin L, Mullins RD, Mitchell DM, Bembenek JN, Reymann AC, Böhme R, Grill SW, Wang JT, Seydoux G, Tulu US, Kiehart DP, Betzig E, Lattice light-sheet microscopy: Imaging molecules to embryos at high spatiotemporal resolution. Science 346, 1257998 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Liu TL, Upadhyayula S, Milkie DE, Singh V, Wang K, Swinburne IA, Mosaliganti KR, Collins ZM, Hiscock TW, Shea J, Kohrman AQ, Medwig TN, Dambournet D, Forster R, Cunniff B, Ruan Y, Yashiro H, Scholpp S, Meyerowitz EM, Hockemeyer D, Drubin DG, Martin BL, Matus DQ, Koyama M, Megason SG, Kirchhausen T, Betzig E, Observing the cell in its native state: Imaging subcellular dynamics in multicellular organisms. Science 360, eaaq1392 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Liu G, Ruan X, Milkie DE, Görlitz F, Mueller M, Hercule W, Killilea A, Betzig E, Upadhyayula S, Characterization, comparison, and optimization of lattice light sheets. Sci Adv 9, eade6623 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Peng T, Thorn K, Schroeder T, Wang L, Theis FJ, Marr C, Navab N, A BaSiC tool for background and shading correction of optical microscopy images. Nat Commun 8, 14836 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Chan YH, Pathmasiri KC, Pierre-Jacques D, Hibbard MC, Tao N, Fischer JL, Yang E, Cologna SM, Gao R, Gel-assisted mass spectrometry imaging enables sub-micrometer spatial lipidomics. Nat Commun 15, 5036 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Tillberg PW, Chen F, Piatkevich KD, Zhao Y, Yu CC, English BP, Gao L, Martorell A, Suk HJ, Yoshida F, Degennaro EM, Roossien DH, Gong G, Seneviratne U, Tannenbaum SR, Desimone R, Cai D, Boyden ES, Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies. Nat Biotechnol 34, 987–992 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Nave K-A, Werner HB, Myelination of the nervous system: Mechanisms and functions. Annu Rev Cell Dev Biol 30, 503–533 (2014). [DOI] [PubMed] [Google Scholar]
  • 61.Cougnoux A, Drummond RA, Collar AL, Iben JR, Salman A, Westgarth H, Wassif CA, Cawley NX, Farhat NY, Ozato K, Lionakis MS, Porter FD, Microglia activation in Niemann–Pick disease, type C1 is amendable to therapeutic intervention. Hum Mol Genet 27, 2076–2089 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Demais V, Barthélémy A, Perraut M, Ungerer N, Keime C, Reibel S, Pfrieger FW, Reversal of pathologic lipid accumulation in NPC1-deficient neurons by drug-promoted release of LAMP1-coated lamellar inclusions. The Journal of Neuroscience 36, 8012 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Dorkenwald S, Matsliah A, Sterling AR, Schlegel P, Yu S, McKellar CE, Lin A, Costa M, Eichler K, Yin Y, Silversmith W, Schneider-Mizell C, Jordan CS, Brittain D, Halageri A, Kuehner K, Ogedengbe O, Morey R, Gager J, Kruk K, Perlman E, Yang R, Deutsch D, Bland D, Sorek M, Lu R, Macrina T, Lee K, Bae JA, Mu S, Nehoran B, Mitchell E, Popovych S, Wu J, Jia Z, Castro MA, Kemnitz N, Ih D, Bates AS, Eckstein N, Funke J, Collman F, Bock DD, Jefferis GSXE, Seung HS, Murthy M, Lenizo Z, Burke AT, Willie KP, Serafetinidis N, Hadjerol N, Willie R, Silverman B, Ocho JA, Bañez J, Candilada RA, Kristiansen A, Panes N, Yadav A, Tancontian R, Serona S, Dolorosa JI, Vinson KJ, Garner D, Salem R, Dagohoy A, Skelton J, Lopez M, Capdevila LS, Badalamente G, Stocks T, Pandey A, Akiatan DJ, Hebditch J, David C, Sapkal D, Monungolh SM, Sane V, Pielago ML, Albero M, Laude J, dos Santos M, Vohra Z, Wang K, Gogo AM, Kind E, Mandahay AJ, Martinez C, Asis JD, Nair C, Patel D, Manaytay M, Tamimi IFM, Lim CA, Ampo PL, Pantujan MD, Javier A, Bautista D, Rana R, Seguido J, Parmar B, Saguimpa JC, Moore M, Pleijzier MW, Larson M, Hsu J, Joshi I, Kakadiya D, Braun A, Pilapil C, Gkantia M, Parmar K, Vanderbeck Q, Salgarella I, Dunne C, Munnelly E, Kang CH, Lörsch L, Lee J, Kmecova L, Sancer G, Baker C, Joroff J, Calle S, Patel Y, Sato O, Fang S, Salocot J, Salman F, Molina-Obando S, Brooks P, Bui M, Lichtenberger M, Tamboboy E, Molloy K, Santana-Cruz AE, Hernandez A, Yu S, Diwan A, Patel M, Aiken TR, Morejohn S, Koskela S, Yang T, Lehmann D, Chojetzki J, Sisodiya S, Koolman S, Shiu PK, Cho S, Bast A, Reicher B, Blanquart M, Houghton L, Choi H, Ioannidou M, Collie M, Eckhardt J, Gorko B, Guo L, Zheng Z, Poh A, Lin M, Taisz I, Murfin W, Díez ÁS, Reinhard N, Gibb P, Patel N, Kumar S, Yun M, Wang M, Jones D, Encarnacion-Rivera L, Oswald A, Jadia A, Erginkaya M, Drummond N, Walter L, Tastekin I, Zhong X, Mabuchi Y, Figueroa Santiago FJ, Verma U, Byrne N, Kunze E, Crahan T, Margossian R, Kim H, Georgiev I, Szorenyi F, Adachi A, Bargeron B, Stürner T, Demarest D, Gür B, Becker AN, Turnbull R, Morren A, Sandoval A, Moreno-Sanchez A, Pacheco DA, Samara E, Croke H, Thomson A, Laughland C, Dutta SB, de Antón PGA, Huang B, Pujols P, Haber I, González-Segarra A, Choe DT, Lukyanova V, Mancini N, Liu Z, Okubo T, Flynn MA, Vitelli G, Laturney M, Li F, Cao S, Manyari-Diaz C, Yim H, Duc Le A, Maier K, Yu S, Nam Y, Bąba D, Abusaif A, Francis A, Gayk J, Huntress SS, Barajas R, Kim M, Cui X, Sterne GR, Li A, Park K, Dempsey G, Mathew A, Kim J, Kim T, Wu G, Dhawan S, Brotas M, Zhang C, Bailey S, Del Toro A, Yang R, Gerhard S, Champion A, Anderson DJ, Behnia R, Bidaye SS, Borst A, Chiappe E, Colodner KJ, Dacks A, Dickson B, Garcia D, Hampel S, Hartenstein V, Hassan B, Helfrich-Forster C, Huetteroth W, Kim J, Kim SS, Kim Y-J, Kwon JY, Lee W-C, Linneweber GA, Maimon G, Mann R, Noselli S, Pankratz M, Prieto-Godino L, Read J, Reiser M, von Reyn K, Ribeiro C, Scott K, Seeds AM, Selcho M, Silies M, Simpson J, Waddell S, Wernet MF, Wilson RI, Wolf FW, Yao Z, Yapici N, Zandawala M, Consortium TF, Neuronal wiring diagram of an adult brain. Nature 634, 124–138 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Materials
NIHMS2209202-supplement-Supplementary_Materials.pdf (10.4MB, pdf)
Movie S1
Download video file (42.4MB, mp4)
Movie S2
Download video file (37.2MB, mp4)

Data Availability Statement

Custom-synthesized materials and reagents are available for researchers within the capacity of RG lab through a simple interinstitutional materials transfer (MTA) agreement (such as a UBMTA). Custom scripts used for petabyte-scale image data processing and analysis (“PetaVIPS”) are available on GitHub at https://github.com/abcucberkeley/PetaVIPS. The version associated with this manuscript is available on Zenodo (https://doi.org/10.5281/zenodo.16929810) (36). The uncompressed raw and processed data for this paper exceeds a petabyte in total and thus is not practical to be hosted on currently available public repositories. All data used in this manuscript are available upon request with a feasible data transfer and storage mechanism (such as Globus endpoint, or cloud buckets). All other data are available in the main text or the supplementary materials.

RESOURCES