Erythropoietin (EPO) is a hematopoietic growth factor that stimulates proliferation and differentiation of erythroid precursor cells and is also known to exert neurotrophic activity in the central nervous system (CNS). However, little is known about expression of EPO and EPO receptor (EPOR) in human CNS tissues. In the present study, we investigated the effects of proinflammatory cytokines on EPO and EPOR expression in highly purified cultures of human neurons, astrocytes, microglia, and oligodendrocytes using reverse transcription-polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA). EPO mRNA was demonstrated only in human astrocytes, while EPOR expression was found in human neurons, astrocytes, and microglia. Neither EPO nor EPOR expression was found in oligodendrocytes. In human astrocytes, EPO mRNA and secreted EPO protein levels were downregulated after exposure to proinflammatory cytokines (IL-1β, IL-6, or TNF-α). In human neurons, TNF-α treatment markedly increased EPOR expression. These results suggest that proinflammatory cytokines regulate expression of EPO and EPOR in human neurons, astrocytes, and microglia and further facilitate interactions among different cell types in the human CNS.

Background and Purpose—: Cerebral microbleeds (MBs) are frequently detected in patients with stroke, especially those who experience intracerebral hemorrhage. However, the clinical significance of MBs in subjects without cerebrovascular disease is still unclear. We performed a prospective study to determine whether the presence of MBs provides useful prognostic information in healthy elderly individuals. Methods—: We tracked 2102 subjects (mean age, 62.1 years) over a mean interval of 3.6 years after they voluntarily participated in the brain checkup system at the Shimane Institute of Health Science. An initial assessment was performed to document the presence of MBs and silent ischemic brain lesions and to map the location of the MBs. During the follow-up period, we obtained information about stroke events that occurred in each subject. Results—: MBs were detected in 93 of the 2102 subjects (4.4%). Strokes occurred in 44 subjects (2.1%) during the follow-up period. They were significantly more common among subjects with MBs. Age and hypertension were independent risk factors for MBs. The presence of MBs was more strongly associated with a deep brain hemorrhage (hazard ratio, 50.2; 95% CI, 16.7 to 150.9) than ischemic stroke (hazard ratio, 4.48; 95% CI, 2.20 to 12.2). All hemorrhagic strokes occurred in deep brain regions, and they were associated with MBs located in the deep brain region. Conclusions—: This longitudinal study demonstrated that the presence of MBs can be used to predict hemorrhagic and ischemic stroke, even in healthy elderly individuals.