Abstracts

Over the past decade there has been great interest in the potential use of brain cooling as a treatment for neurological diseases. Our previous work has shown that focal cooling to 20^oC with a thermoelectric (Peltier) device can rapidly terminate neocortical seizures and produces no cortical injury. Because there have been relatively few studies examining the possibility of brain injury caused by cooling, we decided to determine whether cooling to temperatures close to freezing would lead to any detectable structural alterations in either intact cortex or hippocampal slices. [italic]In vivo[/italic] experiments were performed on halothane-anesthetized rats. A craniotomy was created over the left hemisphere. Cortical cooling was accomplished with a Peltier device, which made direct contact with the pial surface and maintained a temperature of 5^oC on the brain surface. The cooling system was activated by a computer program every 2 minutes for 30 seconds, over 2 hours. The rats were sacrificed 3 day post operation. Paraffin sections of the brain were prepared and examined using Nissl and TUNEL staining for necrotic injury, neuronal loss, and apoptotic death. For [italic]in vitro[/italic] studies, 500 [sub]^[mu][/sub]m hippocampal slices were prepared from 2-3 month old green-fluorescent protein (GFP) expressing transgenic mice (line M). Slices were studied in a submerged chamber maintained at 33^oC and cooled by perfusion with chilled ACSF. The temperature declined by 5^oC/min. In each slice, one or two GFP-positive, distal CA1 dendrites were imaged using a Zeiss LSM 510 multiphoton microscope. We acquired Z-stacks of the same dendrites and associated spines every 15-30 minutes for 2 hours. The images were analyzed off-line by a blinded observer to assess changes in spine number (gain or loss) at different temperatures. In the [italic]in vivo[/italic] experiments there were no evidence of necrosis and apoptosis from Nissl and TUNEL staining on the cooling regions. Quantitative cell counts showed 1075[plusmn]103 neurons/mm² in control cortex and 1098[plusmn]171 neurons/mm² in cooled cortex (p[gt]0.05). When hippocampal slices were cooled to 10^oC, there was no loss of dendritic spines over the 120 minutes. However, in slices that were cooled to 5^oC, we saw a loss of spines and obvious beading of the dendritic shaft. With rewarming, the beading resolved and the spines reappeared. We did not identify new spine formation with either prolonged cooling or rewarming. Our results suggest that cooling to as low as 5^oC is well tolerated by the [italic]in vivo [/italic]and[italic] in vitro [/italic]mammalian central nervous system. The results also showed that there is a large margin of safety between the range of temperatures required for proposed therapeutic uses of hypothermia and the temperatures capable of inducing even transient neuronal alterations. (Supported by the Alafi Family Foundation, the NINDS (R01 NS42936, R21 NS045652, and P01 NS32636), and Citizens United for Research in Epilepsy, Inc.)