Diabetic retinopathy is fundamentally a vascular disease. Compelling evidence suggests that defects in mural cells--smooth muscle cells (SMC) and pericytes, may play an important role in this disease process. The loss of pericytes is the earliest morphological lesion observed in diabetic retinopathy. Furthermore, mice null for genes required for mural cell recruitment develop microaneurysms and leaky vessels, the hallmarks of early phase diabetic retinopathy. My long-term goal is to develop therapeutic strategies to treat this disease, based on a more complete understanding of mural cell biology. Specifically, I am interested in the following questions: how are mural cells recruited to the newly formed vessels? How can the recruitment of mural cells be manipulated? What is the nature of mural cell/endothelial cell interactions? What are the effects of mural cell degeneration on adult vasculature? Can this degeneration process be slowed or even reversed? Mouse retinal vasculature is an ideal system to study these questions because of its well-defined structure, accessibility to experimental intervention, potential for transgenic manipulation and availability of many mutations. Thus, our immediate goal has been to develop animal models and tools to facilitate our studies of questions listed above. To this end, we have been doing the following: (1) Creating transgenic mice whose mural cells express GFP in vivo. (2) Creating transgenic mice that allow conditional ablation of mural cells in vivo. (3) Identifying specific molecular markers for mouse retinal pericytes. Finished work: (1) Creating SMAA-GFP mice whose retinas contain mural cells expressing GFP in vivo. In adult SMAA-GFP mice, vascular smooth muscle cells are brightly fluorescent while pericytes exhibit variable levels of GFP expression. All pericytes associated with choroidal capillaries are positive for GFP, whereas only a subset of pericytes associated with inner-retinal capillaries demonstrate GFP fluorescence. New pericytes are recruited to capillaries during normal vascular development and under conditions of pathological neovascularization. In contrast to the situation seen in mature capillaries, all of these newly recruited pericytes show intense GFP fluorescence. Thus, these mice provide a convenient tool to investigate the process of pericyte recruitment during neovascularization. In one line of SMAA-GFP mice, the RPE cells also expressed GFP at high level. The GFP signal of RPE cells cultured from these mice can be significantly suppressed by short interfering RNAs (siRNAs) duplex against GFP. Since pericytes and RPE cells are involved in various blinding diseases such as diabetic retinopathy and macular degeneration, and RNA interference (RNAi) is a powerful tool for blocking gene expression, the SMAA-GFP mice are a useful animal model to explore RNAi based therapeutic strategies targeting these two cell types. (2) Creating transgenic mice for conditional ablation of mural cells In vivo studies show that lens cells and Schwann cells degenerate in galactose-fed mice over-expressing AR in those cells. Since the endogenous AR activity of wild type mice is very low, the galactose-induced toxicity will be limited to cells in which AR is expressed transgenically. A construct (SMAA-hAR) containing human AR (hAR) cDNA under the control of SMAA promoter was used to produce three lines of SMAA-HAR mice. Using antibodies against hAR, hAR is found in mural cells of retina vessels as well as in a subset of cone bipolar cells and RPE cells. Currently, we are feeding these animals with galactose rich diet to ablate cells expressing high level of HAR. (3) Identification of NG2 as a marker for mouse retinal pericytes One major problem hampering the progress of mural cells research, particularly pericytes, is the lack of definitive in vivo markers. Conflicting reports abound in literature. For example, although smooth muscle alpha actin (SMAA) is the marker of choice for cultured pericytes, its presence in pericytes in vivo has remained unclear. NG2 is a transmembrane proteoglycan and was originally reported to localize to vascular endothelial cells in CNS tissues and mural cells in blood vessels of non-CNS tissues. However, using immunohistochemistry on flat mount and elastase digest retina preparation, The NG2 is highly expressed in vasculature and retinal pigment epithelial (RPE) cells in mouse retinas of all ages. The vascular NG2 immunoreactivity is primarily associated with pericytes in capillaries and smooth muscle cells in larger vessels. In RPE cells, NG2 is localized to the apical plasma membrane. NG2 is the best marker available for identifying retinal pericytes in adult mice. The apical localization of NG2 in RPE cells makes it an useful tool to study the molecular mechanisms underlying the polarization of protein localization in RPE cells. Furthermore, the available NG2 binding peptides provide the possibility of targeting therapeutics to RPE cells and pericytes.
| Sakurai, Shinichi; Alam, Shahabuddin; Pagan-Mercado, Glorivee et al. (2002) Retinal capillary pericyte proliferation and c-Fos mRNA induction by prostaglandin D2 through the cAMP response element. Invest Ophthalmol Vis Sci 43:2774-81 |
