Diabetic nephropathy (DN) is the most common cause of chronic kidney failure and end-stage renal disease in the Western world. Studies from diabetic animal models and clinical trials have shown that inhibition of the renin-angiotensin system delays the progression of advanced DN. However, a recent large-scale clinical trial has revealed that inhibition of renin-angiotensin system in early phases of DN does not slow the decline of renal function or the development of morphological lesions, suggesting that different mechanism(s) may be involved in the different stages of DN. The role of epithelial-mesenchymal transition in renal fibrosis has been intensively investigated. Recently, endothelial-mesenchymal transition, or endothelial-myofibroblast transition (EndoMT) has emerged as another mechanism involved in both developmental and pathological processes. The essential role of EndoMT in cardiac development has been thoroughly studied. EndoMT also exists and contributes to the development and progression of cardiac fibrosis, lung fibrosis, liver fibrosis and corneal fibrosis. EndoMT is a specific form of epithelial-mesenchymal transition. During EndoMT, endothelial cells lose endothelial markers and obtain mesenchymal markers. Recent evidence from our laboratory and others suggests that EndoMT plays an important role in the development of renal fibrosis in several pathological settings, including experimental DN. This review considers the evidence supporting the occurrence of EndoMT in normal development and in pathology, as well as the latest findings suggesting EndoMT contributes to fibrosis in DN. Whether experimental findings of EndoMT will be reproduced in human studies remains to be determined.