Introduction
Tumor necrosis factor-alpha (TNFa) is a pleotrophic cytokine suspected to enhance or deter cellular survival by activation of signal transduction cascades. Brain tissue levels of TNFa are increased after middle cerebral artery occlusion (MCAO). We describe a new model of ischemic brain injury in a transgenic (Tg) rat overexpressing the murine TNFa gene. We tested the hypothesis that chronic elevation of TNFa in the brain of the Tg rat will expand infarct volume after MCAO and cause impairment of spatial discrimination compared to wild type (WT) controls, without affecting post-ischemic reperfusion.
Methods
We constructed a Tg rat overexpressing the murine TNFa gene with its own promoter. Levels of TNFa protein were quantified in tissue homogenates prepared from Tg rats and WT controls. Parallel groups of Tg and WT rats underwent MCAO for 1 h before assessment of percent infarct volume 24 h or 7 d afterward. Animals rendered ischemic and sustained for 7 d underwent a spatial discrimination task in a Morris water maze before measurement of infarct volume. Cortical perfusion was monitored by laser-Doppler flowmetry (LDF) during and after MCAO.
Results
TNFa protein in brain was fivefold higher in non-ischemic Tg vs. WT rats (57±12 [SD] vs. 10±3 pg/100 mg protein; n=3 per group; p≤0.01; unpaired t test). At 24 h, mean %infarct volume was greater in Tg animals (32±11%; n=7) than in WT controls (17±10%; n=9; p≤0.01; unpaired t test). Both groups learned a spatial discrimination task before ischemia (decrease in escape latencies over 5 trial blocks: WT 34.4±16.3 to 8.7±7.2 sec [n=10]; p=0.0002; Tg 44.3±10.9 to 25.8±22 sec [n=6]; p=0.0012; ANOVA /Fisher's test). No significant differences between non-ischemic groups were detected except in the fifth trial block (p=0.02). Post-ischemic WT rats (n=7) showed differential search strategy by spending more than 25% of allotted time in the goal quadrant (39.9±14%; p=0.03; t-test). Tg rats (n=6) failed to demonstrate preference for the goal quadrant that exceeded chance. No significant difference was detected in mean percent infarct volume measured at 7 d between post-ischemic WT (18±8%) and Tg (21±7%) rats. There were no significant differences between Tg and WT rats (n=10 per group) in mean cortical perfusion measured by LDF at pre-ischemic baseline, during 1 h of ischemia, or during 30 min of post-ischemic reperfusion.
Conclusion
TNFa protein was elevated selectively in brain of Tg rats and remained low in serum and external organs. Overexpression of TNFa in brain had no effect on cortical perfusion during and after MCAO, but probably contributed to greater observed infarct volume in Tg animals at 24 h. The difference in infarct volume between WT and Tg animals observed after 24 h was not apparent after 7 d of post-ischemic reperfusion. This may indicate time-related amelioration of acute inflammatory reactions. Both groups of non-ischemic animals performed successfully in cognitive assessment tasks. Ischemic WT rats maintained spatial discriminatory skills while Tg rats failed to persist in the goal quadrant appropriately, perhaps due to the effect of chronic TNFa expression on synaptic architecture.