Based on these SB431542 findings, it was suggested that aspects of human AMD might be treated by CCR2 inhibition. This theoretically should inhibit MP recruitment and reduce VEGF production from these cells, while avoiding the side effects resulting from systemic MP depletion. In addition to recruited MP, many resident eye cells have been reported to produce VEGF, including retinal astrocytes, Mu¨ller cells under hypoxic conditions and the RPE. RPE cells cultured with activated mGlia produce pro-inflammatory, chemotactic and pro-angiogenic molecules, including CCL2 and VEGF. The quantitative contribution of different cell types of the eye to intraocular VEGF production and its impact on CNV is unknown. To answer these questions, we used a murine laser-induced CNV model and analysed VEGF production by flow cytometry. This technique allows simultaneous co-staining for several cell type markers, allowing to classify VEGF-producing cells, as we have reported previously. Furthermore, we employed CCR2deficient mice to clarify which of the VEGF-producing cells of the eye are dependent on this chemokine receptor. We here report that only one cell type in the eye, BU 4061T Proteasome inhibitor namely MP, upregulate VEGF after laser injury in a CCR2-dependent manner. However, their influence on CNV is only transient, arguing against the use of CCR2 inhibitors in AMD. The intraocular production of VEGF is critical for neovascularisation in human AMD and in laser-induced CNV, which often serves as a murine model of AMD. The present study to our knowledge is the first to employ flow-cytometry to identify the cellular sources of VEGF in this model. A major advantage of flow-cytometry is the ability to perform multicolor analysis, which allows the concurrent use of several additional markers to identify VEGF+ cells. This allowed demonstrating that ocular endothelial cells, vimentin cells, and several myeloid immune cells, namely MPs, DCs and microglia cells, but not neutrophil granulocytes, contained VEGF. VEGF-content in the steady state was highest in endothelial cells, and in the retina also in microglia cells. RPE cells have been previously described to produce VEGF, but were not identifiable by flow cytometry, which might be due to their loss during cell isolation or due to physical properties like the dark pigmentation of these cells. Some cells were not classifiable with our technique, and may include for example fibroblasts, which have been shown to produce VEGF in vitro after stimulation with pro-inflammatory cytokines. The intracellular detection of VEGF may indicate cell-intrinsic production or uptake of VEGF from other sources after binding to VEGF receptors on the cell surface. ECs belong to the prime target cells of VEGF; they express particularly strong levels of VEGF receptors, and can use them to internalize this cytokine. Consequently, the VEGF that we detected within ECs might not or only partially be produced by the ECs themselves. Macrophages express lower levels of VEGF receptors, and therefore should be less capable of capturing exogenous VEGF. On the other hand, numerous publications have established macrophages as potent producers of VEGF by analyzing their VEGF mRNA content. Especially in tumors, a role of macrophage-derived VEGF was suggested. VEGF supports tumor angiogenesis and leads to formation of vessels that are similarly irregular and leaky as those in CNV.