SELECTIVE CHEMOKINE MESSENGER-RNA EXPRESSION FOLLOWING BRAIN INJURY

Injury in non-neuronal tissues stimulates chemokine expression leading to recruitment of inflammatory cells responsible for orchestration of repair processes. The signals involved in directing repair of damage to the brain are less well understood. We hypothesized that following brain injury, chemokines are expressed and regulate the rate and patte rn of inflammatory cell accumulation. The two chemokine subfamilies ar e alpha(alpha)-chemokines, which primarily function as neutrophil chem oattractants, and the beta(beta)chemokines, which function primarily a s monocyte chemoattractants. We assessed alpha and beta chemokine mRNA expression patterns and leukocyte accumulation following a cerebral c ortical lesion. Cortical lesions were produced with and without additi on of endotoxin, Escherichia coli lipopolysaccharide (LPS), which stim ulates cytokine expression. We studied the expression of the beta-chem okines: monocyte chemoattractant protein (gene product JE; MCP-1/JE), macrophage inflammatory protein-1 alpha and beta (MIP-1 alpha and MIP- 1 beta), and the regulated upon activation normal T expressed and secr eted chemokine (RANTES) as well as the alpha-chemokines: interferon-ga mma-inducible protein (IP-10) and N51/KC (KC; a murine homologue of MI P-2). Changes in gene expression were analyzed by northern analysis at different time points following injury. Leukocyte and macrophage dens ities were analyzed by immunohistochemistry at the same time intervals . All chemokines were elevated following cortical injury/endotoxin. MC P-1 and MIP-1 alpha were elevated at 2 h and peaked 6 h, MIP-1 beta pe aked at 6 h, but declined more rapidly than MCP-1 or MIP-1 alpha, and IP-10 peaked at 6 h and showed the most rapid decline. KC was elevated at 1 h, and peaked at 6 h following LPS. RANTES was elevated at 1 h a nd achieved a plateau level between 6 and 18 h, then declined. In cont rast, sterile injuries produced in the absence of endotoxin only induc ed the mRNA of the beta-chemokine MCP-1, and its expression was delaye d compared to the cortical injury/endotoxin group. The presence of che mokine message as early as 1 h indicates that expression of this class of molecules is an early response in the repair process following tra umatic brain injury. Macrophage/microglia accumulation occurred more r apidly, activated microglia further from the lesion border, and more c ells accumulated in cortical injury/endotoxin than in cortical lesions produced under sterile conditions. Thus, there was a positive correla tion between beta-chemokine expression and the number of beta-chemokin e responsive cells (i.e. microglia) accumulating in injury sites. This is the first comprehensive study using a panel of chemokine probes an d specific marcophage/microglial markers to study in vivo activation o f the brain following injury. Our data show that the brain is capable of expression of multiple chemokine genes upon appropriate stimulation (e.g. LPS-treatment). The gradient of microglial activation is consis tent with physical damage stimulating release of chemokines that diffu se from the injury site. These data strongly suggest that chemokines a re instrumental in the initiation of repair processes following brain injury. (C) 1998 Elsevier Science B.V.