PURPOSE: Tissue inhibitors of metalloproteinases (TIMPs) are crucial for the maintenance of retinal extracellular matrix such as interphotoreceptor matrix and internal limiting membrane. This study is to examine whether retinal cells respond to mechanical stretching and produce TIMPs. METHODS: Chick retinal adherent cells in near confluency were exposed to mechanical stretching of the bottom of a 6-cm Petri dish at the maximum magnitude of 4,500 microstrain and at a cycle of 30 s for 72 h. TIMP-1 and TIMP-2 levels in the medium at 24, 48 and 72 h after the beginning of stretching were measured by enzyme immunoassay, and their expression was examined by immunohistochemistry. RESULTS: The number of retinal cells during the 72-hour period of stretching did not change significantly both in the stretched group and in the nonstretched control group. Retinal cells in the stretched group produced significantly larger amounts of TIMP-1 and TIMP-2 at 48 h after stretching, compared with nonstretched controls (p = 0.0163 and p = 0.047, respectively, Mann-Whitney U test). Immunohistochemically, a large part of retinal cells in nonstretched Petri dishes was positive for glial fibrillary acidic protein, indicative of glial cells, while some small foci of cells were positive for neuron-specific enolase, indicative of neurons. Fluorescent double labeling demonstrated that both glial cells and neurons were positive for TIMP-1 and TIMP-2. CONCLUSION: Chick retinal cells, most of which were glial cells mixed with a small number of neurons, produced TIMP-1 and TIMP-2. Their production was enhanced by cyclic mechanical stretching. Copyright 2001 S. Karger AG, Basel
PURPOSE: Tissue inhibitors of metalloproteinases (TIMPs) are crucial for the maintenance of retinal extracellular matrix such as interphotoreceptor matrix and internal limiting membrane. This study is to examine whether retinal cells respond to mechanical stretching and produce TIMPs. METHODS:Chick retinal adherent cells in near confluency were exposed to mechanical stretching of the bottom of a 6-cm Petri dish at the maximum magnitude of 4,500 microstrain and at a cycle of 30 s for 72 h. TIMP-1 and TIMP-2 levels in the medium at 24, 48 and 72 h after the beginning of stretching were measured by enzyme immunoassay, and their expression was examined by immunohistochemistry. RESULTS: The number of retinal cells during the 72-hour period of stretching did not change significantly both in the stretched group and in the nonstretched control group. Retinal cells in the stretched group produced significantly larger amounts of TIMP-1 and TIMP-2 at 48 h after stretching, compared with nonstretched controls (p = 0.0163 and p = 0.047, respectively, Mann-Whitney U test). Immunohistochemically, a large part of retinal cells in nonstretched Petri dishes was positive for glial fibrillary acidic protein, indicative of glial cells, while some small foci of cells were positive for neuron-specific enolase, indicative of neurons. Fluorescent double labeling demonstrated that both glial cells and neurons were positive for TIMP-1 and TIMP-2. CONCLUSION:Chick retinal cells, most of which were glial cells mixed with a small number of neurons, produced TIMP-1 and TIMP-2. Their production was enhanced by cyclic mechanical stretching. Copyright 2001 S. Karger AG, Basel