CIRCOAST: a statistical hypothesis test for cellular colocalization with network structures.

Motivation: Colocalization of structures in biomedical images can lead to insights into biological behaviors. One class of colocalization problems is examining an annular structure (disk-shaped such as a cell, vesicle or molecule) interacting with a network structure (vascular, neuronal, cytoskeletal, organellar). Examining colocalization events across conditions is often complicated by changes in density of both structure types, confounding traditional statistical approaches since colocalization cannot be normalized to the density of both structure types simultaneously. We have developed a technique to measure colocalization independent of structure density and applied it to characterizing intercellular colocation with blood vessel networks. This technique could be used to analyze colocalization of any annular structure with an arbitrarily shaped network structure. Results: We present the circular colocalization affinity with network structures test (CIRCOAST), a novel statistical hypothesis test to probe for enriched network colocalization in 2D z-projected multichannel images by using agent-based Monte Carlo modeling and image processing to generate the pseudo-null distribution of random cell placement unique to each image. This hypothesis test was validated by confirming that adipose-derived stem cells (ASCs) exhibit enriched colocalization with endothelial cells forming arborized networks in culture and then applied to show that locally delivered ASCs have enriched colocalization with murine retinal microvasculature in a model of diabetic retinopathy. We demonstrate that the CIRCOAST test provides superior power and type I error rates in characterizing intercellular colocalization compared to generic approaches that are confounded by changes in cell or vessel density. Availability and implementation: CIRCOAST source code available at: https://github.com/uva-peirce-cottler-lab/ARCAS. Supplementary information: Supplementary data are available at Bioinformatics online.

Myh11+ microvascular mural cells and derived mesenchymal stem cells promote retinal fibrosis.

Retinal diseases are frequently characterized by the accumulation of excessive scar tissue found throughout the neural retina. However, the pathophysiology of retinal fibrosis remains poorly understood, and the cell types that contribute to the fibrotic response are incompletely defined. Here, we show that myofibroblast differentiation of mural cells contributes directly to retinal fibrosis. Using lineage tracing technology, we demonstrate that after chemical ocular injury, Myh11+ mural cells detach from the retinal microvasculature and differentiate into myofibroblasts to form an epiretinal membrane. Inhibition of TGFßR attenuates Myh11+ retinal mural cell myofibroblast differentiation, and diminishes the subsequent formation of scar tissue on the surface of the retina. We demonstrate retinal fibrosis within a murine model of oxygen-induced retinopathy resulting from the intravitreal injection of adipose Myh11-derived mesenchymal stem cells, with ensuing myofibroblast differentiation. In this model, inhibiting TGFßR signaling does not significantly alter myofibroblast differentiation and collagen secretion within the retina. This work shows the complexity of retinal fibrosis, where scar formation is regulated both by TGFßR and non-TGFßR dependent processes involving mural cells and derived mesenchymal stem cells. It also offers a cautionary note on the potential deleterious, pro-fibrotic effects of exogenous MSCs once intravitreally injected into clinical patients.

Pericyte Bridges in Homeostasis and Hyperglycemia.

Diabetic retinopathy is a potentially blinding eye disease that threatens the vision of one-ninth of patients with diabetes. Progression of the disease has long been attributed to an initial dropout of pericytes that enwrap the retinal microvasculature. Revealed through retinal vascular digests, a subsequent increase in basement membrane bridges was also observed. Using cell-specific markers, we demonstrate that pericytes rather than endothelial cells colocalize with these bridges. We show that the density of bridges transiently increases with elevation of Ang-2, PDGF-BB, and blood glucose; is rapidly reversed on a timescale of days; and is often associated with a pericyte cell body located off vessel. Cell-specific knockout of KLF4 in pericytes fully replicates this phenotype. In vivo imaging of limbal vessels demonstrates pericyte migration off vessel, with rapid pericyte filopodial-like process formation between adjacent vessels. Accounting for off-vessel and on-vessel pericytes, we observed no pericyte loss relative to nondiabetic control retina. These findings reveal the possibility that pericyte perturbations in location and process formation may play a role in the development of pathological vascular remodeling in diabetic retinopathy.

Myh11 Lineage Corneal Endothelial Cells and ASCs Populate Corneal Endothelium.

Purpose: To establish Myh11 as a marker of a subset of corneal endothelial cells (CECs), and to demonstrate the feasibility of restoring the corneal endothelium with Myh11-lineage (Myh11-Lin[+]) adipose-derived stromal cells (ASCs). Methods: Intraperitoneal administration of tamoxifen and (Z)-4-hydroxytamoxifen eyedrops were used to trace the lineage of Myh11-expressing cells with the Myh11-Cre-ERT2-flox-tdTomato mouse model. Immunostaining and Western blot characterized marker expression and spatial distribution of Myh11-Lin(+) cells in the cornea, and administration of 5-ethynyl-2'-deoxyuridine labeled proliferating cells. ASCs were isolated from epididymal adipose Myh11+ mural cells and treated with cornea differentiation media to evaluate corneal endothelial differentiation potential. Differentiated ASCs were injected into the anterior chamber to test for incorporation into corneal endothelium following scratch injury. Results: A subset of CECs express Myh11, a marker previously thought restricted to only mural cells. Myh11-Lin(+) CECs marked a stable subpopulation of cells in the cornea endothelium. Myh11-Lin(+) ASCs undergo CEC differentiation in vitro and incorporate into injured corneal endothelium. Conclusions: Dystrophy and dysfunction of the corneal endothelium accounts for almost half of all corneal transplants, the maintenance of the cornea endothelium is poorly understood, and there are a lack of mouse models to study specific CEC populations. We establish a mouse model that can trace the cell fate of a subpopulation of CECs based on Myh11 expression. A subset of ASCs that share this Myh11 transcriptional lineage are capable of differentiating into CECs that can incorporate into injured corneal endothelium, revealing a potential cell source for creating engineered transplant material.

Adipose-derived stem cells from diabetic mice show impaired vascular stabilization in a murine model of diabetic retinopathy.

Diabetic retinopathy is characterized by progressive vascular dropout with subsequent vision loss. We have recently shown that an intravitreal injection of adipose-derived stem cells (ASCs) can stabilize the retinal microvasculature, enabling repair and regeneration of damaged capillary beds in vivo. Because an understanding of ASC status from healthy versus diseased donors will be important as autologous cellular therapies are developed for unmet clinical needs, we took advantage of the hyperglycemic Akimba mouse as a preclinical in vivo model of diabetic retinopathy in an effort aimed at evaluating therapeutic efficacy of adipose-derived stem cells (mASCs) derived either from healthy, nondiabetic or from diabetic mice. To these ends, Akimba mice received intravitreal injections of media conditioned by mASCs or mASCs themselves, subsequent to development of substantial retinal capillary dropout. mASCs from healthy mice were more effective than diabetic mASCs in protecting the diabetic retina from further vascular dropout. Engrafted ASCs were found to preferentially associate with the retinal vasculature. Conditioned medium was unable to recapitulate the vasoprotection seen with injected ASCs. In vitro diabetic ASCs showed decreased proliferation and increased apoptosis compared with healthy mASCs. Diabetic ASCs also secreted less vasoprotective factors than healthy mASCs, as determined by high-throughput enzyme-linked immunosorbent assay. Our findings suggest that diabetic ASCs are functionally impaired compared with healthy ASCs and support the utility of an allogeneic injection of ASCs versus autologous or conditioned media approaches in the treatment of diabetic retinopathy.