Sparse Cell Labeling Reveals Morphologic Diversity Among Mouse Scleral Fibroblasts

Abstract

Purpose

The purpose of this study was to investigate mouse scleral cell morphology in healthy and ocular hypertensive eyes.

Methods

B6.Cg-Ndor1Tg (UBC-cre/ERT2) 1Ejb/1J (UBC-Cre-ERT2) (Strain # 007001; Jackson Labs, Bar Harbor, ME, USA) were crossed with Gt(ROSA)26 Sor tm1(CAG-Brainbow2.1)Cle /J (R26R-Confetti) (Strain # 017492, Jackson Labs) to obtain hemizygous offsprings (UBC-Confetti). This enables tamoxifen-induced recombination of the Brainbow 2.1 cassette, inducing random expression of fluorescent proteins in all cells. Two weeks after tamoxifen treatment (intraperitoneal injection, n = 7 mice), anterior chamber bead injection raised the intraocular pressure (IOP). Eyes were enucleated 3 days after bead injection. YFP and GFP-labeled peripapillary scleral (PPS) and peripheral scleral cells were imaged, cell volume was measured, and individual cell morphology was highlighted using Imaris (version 9.3; 89 control and 136 glaucomatous cells).

Results

Imaging of YFP/GFP-labeled cells allowed visualization of individual scleral cells. Differences in the intensity of the Confetti YFP/GFP signal compared to nuclear and fibrillar actin (FA) labels enabled visualization of FA and nuclear structure within YFP/GFP-labeled cells. Under baseline conditions, cells were morphologically diverse, displayed migratory structures, and extensions into the surrounding extracellular matrix (ECM). Mean PPS cell volume (692 µm³, interquartile range [IQR] = 419.8–1304 µm³, n = 62 cells) was greater than peripheral cell volume (336 µm³, IQR = 217–488 µm³, n = 27 cells, P < 0.0001). Cells in glaucomatous eyes exhibited similar morphologic diversity and had reduced PPS cell volume compared with controls (277 µm³, IQR = 154–532, n = 83 cells, P < 0.0001).

Conclusions

Sparse cell labeling allowed visualization of individual cell morphology in the scleral environment. Cells demonstrated diverse morphologies, and PPS cells responded to IOP elevation.

Sclera, the tough, opaque outer layer that covers the eye, is essential for maintaining healthy vision. Sclera provides structural support, protects the eye from injury, and plays a key role in defining the eye's shape and size.¹ The sclera also plays a central role in emmetropization by regulating axial length in early childhood to properly focus light on the retina.² Axial length deviations result in hyperopic and myopic refractive errors requiring correction with glasses, contacts, or surgery. Hyperopic and myopic eyes also have an increased risk of angle-closure glaucoma and retinal detachment, respectively.^3–6

Localized scleral remodeling has been associated with the pathogenesis of distinct ocular diseases. Posterior staphylomas in highly myopic eyes are characterized by peripapillary scleral (PPS) thinning, collagen disorganization, and focal submacular thinning.^7,8 In glaucoma, structural changes of the PPS and lamina cribrosa are well-recognized hallmarks of disease progression.^9–11 Together, these conditions represent two of the most significant global causes of visual impairment and are expected to see their prevalence increase substantially in the coming decades.^12–14 Despite the sclera's central role in these disorders, the cellular and molecular pathways that govern its remodeling remain largely undefined.

Scleral fibroblasts are the most abundant cell type in the scleral stroma and are key effectors in responding to external stimuli and driving stromal remodeling.^15–18 Scleral fibroblasts rapidly respond to environmental factors, including cellular hypoxia and mechanical stress induced by fluctuations in intraocular pressure (IOP). This response triggers the differentiation of fibroblasts into myofibroblasts, the expression of proliferative markers, and the activation of key signaling pathways, including Rho-associated kinase and TGFβ.¹⁹ Under normal conditions, scleral fibroblasts exhibit functional diversity with distinct subpopulations identified through single-cell sequencing and immunolabeling.^18,20 However, in scleral pathologies like myopia and glaucoma, shifts in the proportions of these subsets contribute to tissue dysfunction and disease progression. To further our understanding of these vision-threatening conditions, it is critical to investigate how different scleral fibroblast subpopulations respond to pathological insults.

Fibroblast morphology is a key indicator of cellular function and behavior, providing valuable insights into how these cells respond to stimuli. The shape and structure of the fibroblast cytoskeleton, nucleus, organelles, and membrane vary depending on whether the cell is in a resting or activated state.^21–23 These morphological changes reflect underlying cellular processes such as migration, differentiation into myofibroblasts, and apoptosis, which precede tissue remodeling. Fibroblasts can undergo significant morphological alterations in response to mechanical stress or environmental cues before changes in their transcriptional profile.^24,25 These early changes in cell morphology and cytoskeletal organization occur rapidly prior to altered gene expression and are driven by post-transcriptional and post-translational mechanisms, such as Rho-GTPase and Src signaling.²⁶ Although the scleral fibroblast transcriptional and proteomic responses to elevated IOP have been characterized previously, the accompanying changes in single-cell morphology have remained largely unexplored because conventional imaging approaches cannot reliably resolve individual fibroblasts within the intact scleral network.

In tissues like the scleral stroma, visualizing individual fibroblast morphology can be particularly challenging. Quantitative analysis by Oglesby and colleagues estimated that the normal mouse sclera contains only approximately 4000 fibroblasts/mm³, or roughly 75,000 fibroblasts throughout the entire sclera, with the PPS exhibiting approximately 30% greater cellular density than more peripheral regions.¹⁹ Despite this relatively low cellular density, conventional labeling of all scleral fibroblasts produces a complex 3D meshwork of overlapping cell bodies and highly branched cellular processes.^27,28 Individual cells occupy large spatial territories and interdigitate extensively with neighboring fibroblasts, making it difficult to determine where one cell ends and another begins using standard histologic or immunofluorescent approaches. As a result, important features of single-cell morphology – including cell volume, process architecture, branching complexity, and regional heterogeneity – cannot be reliably resolved. However, understanding these morphological variations is essential for uncovering how different scleral fibroblast populations contribute to scleral remodeling in both health and disease. To overcome these limitations, we used sparse genetic labeling to isolate individual scleral fibroblasts within their native tissue environment, enabling comprehensive 3D reconstruction and quantitative morphometric analysis of single cells in situ.

To overcome these challenges, we use confetti mice in this study.^29,30 This genetic model enables the sparse labeling of individual cells in multiple distinct colors, allowing us to analyze cell morphology with high fidelity and subsequently providing powerful insights into cellular dynamics, lineage relationships, and interactions within the scleral tissue. By visualizing individual scleral fibroblasts under both baseline conditions and after IOP elevation, we can gain critical insights into cellular behavior and the heterogeneity of scleral populations in health and disease. We hypothesized that scleral fibroblasts exhibit distinct, quantifiable regional morphological variations, and differential responses to elevated IOP. We believe that these changes reflect underlying shifts in cellular behavior and region-specific heterogeneity within the scleral fibroblast population. This approach provides a new and valuable tool to study scleral remodeling at a level of detail previously difficult to achieve.

Materials and Methods

Mouse Strain Development

Animal experimentation was conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee of Johns Hopkins University and in accordance with the ARVO statement for the Use of Animals in Ophthalmic and Vision Research.³¹ B6.Cg-Ndor1^{Tg(UBC-cre/ERT2)1Ejb}/1J (Strain # 007001, Jackson Labs), called here UBC-Cre-ERT2, and Gt(ROSA)26 Sor ^{tm1(CAG-Brainbow2.1)Cle} /J (Strain # 017492, Jackson Labs), called here R26R-Confetti, were commercially obtained and hemizygous UBC-Cre-ERT2 were crossed with homozygous R26R-Confetti^Tg/Tg to obtain hemizygous UBC-Cre-ERT2 / R26R-Confetti^Tg/0 offspring that were ready for tamoxifen induction.

Tamoxifen Induction

Tamoxifen (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in corn oil (Sigma-Aldrich) to a final concentration of 20 mg/mL and stored at 4°C, protected from light. Tamoxifen was stored for no longer than 7 days prior to use. Mice (UBC-Confetti, 12–24 weeks, male and female mice) were injected intraperitoneally (IP) with 200 microliters (µL; 4 mg) of the solution through a 25-gauge needle for 3 consecutive days.

Bead Injection

Beads were injected into the mouse anterior chamber to raise IOP (n = 7 eyes of 7 mice).³² Following sedation with a mixture of ketamine (Fort Dodge Animal Health, Fort Dodge, IA, USA), xylazine (VedCo Inc., Saint Joseph, MO, USA), and acepromazine (Phoenix Pharmaceuticals, Burlingame, CA, USA), mice underwent unilateral intracameral bead injection. Briefly, a glass cannula with a 50-µm tip diameter connected to a Hamilton syringe (Hamilton, Inc., Reno, NV, USA) was used to inject 2 uL of each of 6 µm and 1 µm diameter Polybead Microspheres (Polysciences, Inc., Warrington, PA, USA) followed by 10 mg/mL sodium hyaluronate (Healon PRO; Advanced Medical Optics Inc., Santa Ana, CA, USA).

IOPs were measured under isoflurane using the TonoLab rebound tonometer (iCare, Vantaa, Finland) according to the manufacturer's instructions. IOPs were measured 3 days after bead injection, prior to enucleation. Final IOP measurements were calculated using the average of three separate IOP measurements for each eye. We selected the 3-day time point after microbead injection because it provides a consistent and reproducible elevation in IOP while capturing the earliest scleral fibroblast responses to ocular hypertension prior to detectable retinal ganglion cell loss.^19,33,34 This interval is sufficient to induce early transcriptional changes in scleral fibroblasts yet precedes substantial structural remodeling of the sclera.^19,33 Additionally, harvesting at this early stage minimizes the likelihood that elevated IOP or downstream cellular responses would alter YFP/GFP expression, thereby allowing fluorescent labeling to accurately reflect baseline cell morphology rather than pressure-induced changes in reporter expression.

Tissue Preparation for Imaging

Two weeks after tamoxifen injection, the mice were euthanized, and their eyes were enucleated. Extraocular tissue was removed under a dissecting microscope, and posterior eye cups were isolated. Eye cups were either prepared as fresh scleral flatmounts or fixed for 2 hours at 4°C in 4% paraformaldehyde in Sorenson's phosphate buffer. Afterward, fixed eye cups were either made into fixed sclera flatmounts (n = 6 normotensive eyes) or cryopreserved in a 2:1 mixture of sucrose and TissueTek optimum cutting temperature (O.C.T.) compound (Sakura Finetek USA, Inc., Torrance, CA, USA; n = 7 eyes with OHTN and 7 fellow control eyes). Directional positions (nasal versus temporal versus superior versus inferior) were not maintained. Cryopreserved scleral tissue was sectioned tangentially, parallel to the scleral surface. Sections were cut to a thickness of 12 to 14 µm. Previous studies have captured multiple layers of sheet-like scleral fibroblasts at similar thicknesses, as these cells lie flat between collagen layers in this orientation.³⁵

Sections were incubated in phosphate buffered saline (PBS) with 0.5% Triton X-100 (PBT) and 10% normal goat serum (NGS; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) overnight at 4°C. Slides were then washed with PBS, incubated with phalloidin Alexa Fluor 568 (A1280, Thermo Fisher Scientific, Waltham, MA, USA) for 45 minutes at room temperature, washed again with PBS, and mounted using ProLong Glass Antifade Mountant with NucBlue Stain (Thermo Fisher Scientific).

Confocal Microscopy and Imaging Analysis

Imaging was performed on a Zeiss LSM 710 confocal laser scanning microscope (Carl Zeiss Microscopy, Thornwood, NY, USA). Tiled Z-stack images were obtained using a 63× objective. NucBlue, GFP/YFP, and Phalloidin were imaged sequentially using 405 nm, 488 nm, and 561 nm lasers, respectively. UBC-Confetti RFP and CFP fluorescent labels had signal-to-noise ratios too low for imaging and analysis.

Images were analyzed using Imaris (version 9.3; Oxford Instruments, Zurich, Switzerland). Identification of PPS and peripheral sclera through serial sectioning was described previously.²⁷ Briefly, serial sections were examined starting at the optic nerve stump, progressing anteriorly until scleral tissue contiguous with the optic nerve was identified. The region contiguous with the optic nerve with circumferentially oriented actin fibrils was defined as PPS. Sclera outside this region containing a basket-weave pattern was defined as peripheral sclera. The number of cells imaged in each region was equally distributed between the eyes. GFP/YFP-positive cells within tiled Z-stacks were screened for images containing whole, non-truncated cells, which were then isolated by 3D cropping, and cell volumes were segmented using the surface function. Uniform parameters were used for each imaging session and were manually reviewed by a technician who was masked to bead-injection status. Truncation of cells within sections was often easily determined by visualizing a straight plane cut through an irregularly shaped cell. These cells were removed from any analysis. Additionally, truncated cells were identified using the “Distance to Image Border” filter set to 1 µm. To isolate the FA and nuclear structures of each cell, the segmented surface volume was used to create a region of interest (ROI), and all nuclei and FA signal outside the ROI were excluded from analysis.

Statistical Analysis

Data were summarized as mean ± standard deviation for normally distributed variables and as median with interquartile range (IQR) for non-normally distributed variables, as assessed by the Shapiro–Wilk test. Statistical significance was evaluated using the Mann–Whitney or Kruskal-Wallis test, with P < 0.05 considered significant. All analyses used GraphPad Prism version 10 (GraphPad Software, San Diego, CA, USA).

Results

Sparse Cell Labeling Enables Visualization of Individual Scleral Cell Morphology

To assess whether tamoxifen induction of UBC-Confetti mice enabled visualization of individual scleral fibroblast morphology, posterior eyecups were fixed and dissected 2 weeks after tamoxifen treatment, and fixed scleral flat mounts were created. Multichannel imaging revealed sparse labeling, with YFP+ and GFP+ cells distributed throughout the scleral stroma (Fig. 1). When combined with nuclear labeling (NucBlue) and fibrillar actin (FA) staining (phalloidin), we confirmed sparse GFP labeling in volumes associated with a single nucleus, indicating successful visualization of individual scleral fibroblasts. Imaging of fresh sclera flat mounts from UBC-Confetti mice also revealed sparsely labeled cells. However, image quality was less clear than in fixed tissues (Supplementary Fig. S1).

Figure 1.

Sparse cell labeling of scleral flat mounts. (A) Schematic of tamoxifen recombination in UBC-Confetti mice and (B) injection protocol that induces recombination in ubiquitin-expressing cells (above) and schematic of imaging GFP/YFP/CFP positive cells (below). (C) YFP-labelled scleral cells after tamoxifen induction. (D–F) Sparse labeling of GFP+ scleral fibroblasts reveals cell volume associated with a single nucleus (arrow). Scale bar = 15 µm.

Image Processing Highlights Distinct Nucleus and Fibrillar Actin Structures

To obtain higher-resolution images of individual scleral fibroblasts and to enable imaging of PPS scleral fibroblasts, UBC-Confetti eyes were harvested and fixed for cryosectioning. Scleral cryosections were stained for FA and mounted with NucBlue prior to 3D confocal imaging. Image processing using Imaris allowed clear integration of FA staining, nuclear position and shape with NucBlue, and GFP/YFP+ defined cell volume provided by UBC-Confetti to illustrate the complete morphology of single fibroblasts (Fig. 2).

Figure 2.

Sparse labeling and image processing. (A) Nuclear (blue) and fibrillar actin (red) labeling of the optic nerve (ON) and surrounding tissue, sectioned through a region including the anterior dural strut and posterior PPS (*). (B) GFP+ cell is highlighted (arrow). (C) Image processing to exclude all nuclei and FA outside the GFP+ cell volume with enlarged and rotated front (D1), side (D2), and top (D3) views of the cell. Scale bars = 15 µm A to C and 5 µm D.

We then imaged fibroblasts from the peripheral sclera, PPS, and dural struts (Fig. 3) and processed images to highlight each cell's FA and nuclear structures. Using this approach, we found diverse cell morphologies in the scleral stroma. Many fibroblasts did not conform to the classic spindle-shaped morphology with a centrally located nucleus, which is characteristic of stromal fibroblasts in cell culture. Instead, in situ scleral fibroblasts frequently had polarized nuclei and diverse morphologic features, including cytoplasmic extensions and migratory structures, such as strictures, lamellipodia, and dendrites.

Figure 3.

Isolated cells from PPS, peripheral sclera, and dural struts. Multiple views of PPS cells (A–D), peripheral scleral cells (E, F), and cells from the peripapillary dural struts (G). Scale bars = 7 µm A, 5 µm B, 5 µm C, 10 µm D, 3 µm E, 7 µm F1 and G1, and 5 µm F2.

Cell volume analysis demonstrated regional differences between the PPS and peripheral sclera (Fig. 4). PPS cell volume ranged from 171 to 3788 µm³ in contrast to peripheral cells that had a smaller range (107 to 975 µm³). Median volume of PPS fibroblasts (692 µm³, IQR = 419.8–1304 µm³, n = 62 cells) was significantly larger than peripheral fibroblasts (336 µm³, IQR = 217–488 µm³, n = 27 cells, P < 0.0001, Mann-Whitney U test) and 28% of PPS fibroblasts had volumes greater than 1000 µm³.

Figure 4.

PPS cell volume is greater than peripheral scleral cell volume. (**** = P < 0.0001).

IOP Elevation Causes Rapid Cell Morphology Changes

To investigate the effect of IOP elevation on cell morphology, we utilized the anterior bead injection (BI) model in tamoxifen-treated UBC-Confetti mice. Three days following bead injection, IOP was significantly higher in BI (20.9 ± 2.9 millimeters of mercury [mm Hg]) than in fellow untreated (10.6 ± 1.9 mm Hg, n = 7 mice, P < 0.0001, t-test). Eyes were enucleated 3 days after BI, fixed, cryosectioned, stained, and imaged. Cells retained extensions, diverse morphology, and migratory structures. However, cell volumes changed in the PPS and peripheral sclera following IOP elevation (Fig. 5). In the PPS, median cell volume in glaucomatous eyes was 277 µm³ (IQR = 154–532, n = 83 cells, 7 mice), a significant decrease from non-glaucomatous eyes (P < 0.0001, Mann-Whitney U test). In the peripheral sclera, median glaucomatous cell volume was 373 µm³ (IQR = 220–562, n = 53 cells, 7 mice), which was not significantly changed compared with control eyes (P = 0.6, Mann-Whitney U test).

Figure 5.

Increased IOP level alters PPS cell volume. (A) IOP levels in control (C) and bead-injected (BI) eyes were measured 3 days after bead injection (n = 7 mice). Peripheral (B) and PPS C cells after BI (scale bars = 10 µm). Cell volumes of control and bead-injected eyes in the PPS (D) and peripheral (E) sclera. (**** = P < 0.0001; NS = not significant).

Discussion

Here, we used sparse cell labeling in a mouse model to visualize individual scleral fibroblasts under normotensive and ocular hypertensive conditions. This approach enabled high-resolution assessment of scleral cell morphology using standard fluorescence microscopy and revealed several key findings. Scleral fibroblasts exhibited polarized nuclei, cytoplasmic extensions, cytoplasmic strictures, and demonstrated marked morphological heterogeneity. Cell volume varied regionally, with a significantly larger median volume in PPS cells compared with peripheral scleral fibroblasts. Moreover, cellular responses to elevated IOP were region-specific: PPS cell volume was significantly reduced after 3 days of IOP elevation, whereas peripheral scleral cell volume remained unchanged. Together, these findings underscore the structural complexity of scleral fibroblasts and reveal regionally distinct cellular response to IOP-mediated mechanical stress.

The complex architecture of scleral fibroblasts was noted previously. Elongated fibroblasts are interspersed between cell-free collagenous lamellae, forming a dense stromal reticular cellular network.^19,35 Although the sclera is often described as paucicellular, a readily identified population of cells occupies approximately 20% of the scleral thickness in mice, with a density of 4000 cells per mm³.³⁶ In both mouse and human tissue, fibroblasts are closely approximated to their surrounding extracellular matrix (ECM), and their cytoskeletal organization is strongly influenced by the orientation of adjacent collagen fibrils.^27,28 Using focus ion beam/scanning electron microscopy (FIB/SEM), Murata et al. imaged posterior and equatorial mouse scleral fibroblasts and identified polygonal cells compressed between collagenous lamellae with both horizontal and vertical contact points linking neighboring fibroblasts.³⁵ These findings support the hypothesis that scleral fibroblasts form a multicellular network via cell-to-cell contact points and adhesion proteins such as connexin-43.³⁷

Our approach complements these studies by further illustrating cellular morphology. Here, we imaged a larger number of eyes, a greater area of cells within each eye, and multiple scleral regions compared with prior TEM and FIB/SEM studies. We found marked morphologic diversity with many cells having polarized nuclei, cellular extensions, cytoplasmic structures, membrane blebbing, and structures resembling lamellae which are commonly associated with migratory or mechanically engaged cells in dense ECMs.³⁸ In addition to confirming morphological heterogeneity, we quantified cell volumes. We found a wide distribution of fibroblast sizes, with larger and more variable volumes in the PPS compared with more uniform volumes in peripheral regions. However, due to the sporadic labeling of a limited fraction of fibroblasts, we did not observe the extensive, interconnected fibroblast network. Imaging additional fluorophores in the Confetti reporter, such as CFP and RFP, as well using higher tamoxifen dosages to increase recombination frequency, would increase the likelihood of imaging two adjacent fibroblasts and their intercellular connectivity.

Our findings further highlight the PPS as a distinct scleral region with an enhanced response to mechanical stress. PPS has a specialized microarchitecture and material components when compared with other scleral regions. In contrast to the peripheral sclera, which contains isotropic collagen fibers arranged in a basket-weave pattern, PPS collagen fibers are highly aligned and circumferentially arranged around the optic nerve. This collagen architecture, along with a relatively high elastin content, protects the optic nerve from IOP-generated mechanical strain.¹ Glaucoma stiffens the PPS, and its highly anisotropic collagen microarchitecture becomes disorganized, thereby enhancing the translation of IOP-generated mechanical stress to the optic nerve and worsening of glaucoma severity.^9,36,39 Cell culture and single-cell RNA-seq showed unique transcriptomic signatures in PPS cells compared with peripheral scleral regions.^18,40 Here, we demonstrate additional differentiating features of PPS cells. PPS cell volumes differ from those of peripheral scleral cells and decrease after 3 days of IOP elevation. These findings support a model in which PPS cells have unique features that are IOP-responsive, underlie glaucomatous PPS remodeling, and affect glaucoma development and progression.

There are several possible explanations for the specific reduction in PPS fibroblast volume following BI. It could reflect increased mechanical compression of the cells and surrounding ECM. However, this explanation is difficult to reconcile with two observations: compression would be expected to resolve once the tissue was dissected and no longer under load, and peripheral scleral fibroblasts – which would experience comparable mechanical compression – did not show the same volume loss. Another possibility is that Confetti-labeled circulating monocytes entering the sclera upon IOP elevation could account for the observed decrease. Monocytes differentiate into activated macrophages upon tissue entry, and these cells adopt an ameboid morphology with a cell volume two to five times greater than that of scleral fibroblasts.⁴¹ If immune cell infiltration were driving the change, we would therefore expect measured cell volume to increase rather than decrease. Other potential explanations include cytoskeletal reorganization, cellular quiescence, autophagy, and regulatory volume decrease.^42–45 Future experiments – including cell-type-specific labeling to directly quantify immune cell proportion before and after IOP elevation – are necessary to determine the exact mechanism responsible for this change in PPS fibroblast cell volume.

There are limitations to this study. First, all studies were performed in mouse eyes. Many features of the mouse PPS are conserved in humans; however, the shape of the optic nerve head (ONH) differs between mouse and human eyes, and the mouse ONH lacks a collagenous lamina cribrosa. Second, UBC-CreERT2 is not expressed specifically in fibroblasts and could capture rare non-fibroblast cell types, including immune cells. However, studies of murine, rabbit, and human sclera have demonstrated that fibroblasts constitute the vast majority of scleral cells, providing confidence that most, if not all, of the cells imaged are fibroblasts.^32,35 Additionally, because fibroblasts were imaged in isolated tissue, some morphological changes may be lost once the eye is enucleated and the gauge pressure becomes 0 mm Hg. Finally, this current approach imaged only a subset of scleral cells and did not capture cell-to-cell contact points. Future studies will be able to take advantage of this mouse line by immunolabeling CFP- and RFP-positive cells to amplify the signal sufficiently for imaging and by using higher concentrations of tamoxifen to achieve higher recombination frequencies and visualize adjacent cells.

Conclusions

Here, we describe a method to visualize individual scleral cell morphology in mouse eyes and highlight regional differences in mean cell volume and response to elevated IOP. These findings further implicate a specialized role for peripapillary fibroblasts in glaucomatous ONH remodeling and pathogenesis.