Normal rat retina lacks significant numbers of ED2-positive macrophages.

We want to correct some statements that we published in an earlier volume of IOVS. We described microglia and macrophages in the normal rat retina and the response of these cells during the endotoxin-induced uveitis model (EIU). Retinal wholemounts from normal rats and at various time points after endotoxin injection were stained with the monoclonal antibodies ED1 and ED2. We reported the staining pattern and densities (196–271 cells/mm ) of these two antibodies to be essentially similar and described a massive influx of macrophages 24 to 96 hours after endotoxin injection. We reported that these cells were both ED11 and ED21. Since then we have repeated the experiments and performed similar protocols in collaboration with Dr. Paul McMenamin, who has extensive experience with these monoclonal antibodies, during a sabbatical visit he paid to our Institute. We experienced some difficulties in the staining of ED2, which caused us to review the material originally published in IOVS. With the aid of Dr. McMenamin, we performed extensive tests with the monoclonal antibody batches used in the original experiments, compared them with new batches, and concluded that retinal microglia, i.e., the macrophages within the retinal parenchyma, are weakly ED11 but ED22. The original findings that these cells were ED21 was in fact erroneous and appeared to be due to an error in labeling of antibodies at the time of aliquoting. It is important that this issue be clarified, because a number of studies of normal rat retinal tissue have failed to find significant numbers of ED21 macrophages in the retina. Furthermore, no influx of ED21 macrophages was noted in EIU by Pouvreau et al. We apologize for any inconvenience caused to our colleagues as a result of this error.