Quantitative Imaging of the Heterogeneity of Brain Potassium Depletion in Experimental Focal Ischemia

With few exceptions, pathological progression in ischemic stroke is presumed to occur uniformly within the ischemic core region. These exceptions include edema formation, brain tissue [Na+] increase, and the qualitative visually-observed decrease of brain tissue [K+], [K+]br, all of which occur in peripheral regions of the ischemic core. We hypothesize that [K+]br within these peripheral regions are heterogeneous (with lower [K+]br in the peripheral compared to the central ischemic core) and are not associated with neuronal degradation. Permanent focal ischemia in 13 rats was produced for 2.5-5 h. Brain sections were quantitatively stained for K+ to assess [K+]br variations between the peripheral and central ischemic core. Regions within the cortical ribbon were used to explore differing rates of K+-depletion expressed as the slopes of [K+]br vs. time relations. Adjacent sections were observed for reflective change and stained for microtubule-associated protein 2 (MAP2) to identify the ischemic region and to relate neuronal pathology to [K+]br variations. The mean value of normal cortex (NC) [K+]br was 96 mEq/kg and of K+-depletion in all ischemic regions over time was 12.2 mEq/kg/h, consistent with measurements from other studies. Exaggerated K+-depletion occurred in 56% of the peripheral ischemic core regions classed as depleted peripheral ischemic core (ICp-DP) regions. These were clearly separated (p<0.001) from the non-depleted peripheral ischemic core (ICp-ND) regions. The normal cortex (NC) regions show stability of [K+]br with a slope near zero. However, the 13.6 mEq/kg/h slopes of the central ischemic core (ICc) and ICp-ND regions were similar (p=0.99) and showed a significant decrease over time. The 6.2 mEq/kg/h slope of the ICp-DP regions was significantly different from that of the ICc (p=0.010) and the ICp-ND (p=0.0071). This lower slope of the ICp-DP curve 2.5 h after stroke onset is due to the accelerated K+-efflux from 0 to 2.5 h, as its value at stroke onset must be [~]100 mEq/kg. However, these differential K+ losses were not reflected in the homogeneous peripheral ischemic core MAP2 immunoreactivity losses. Unlike [K+]br, there was no difference between the MAP2 immunoreactivity in K+-depleted and non-K+-depleted peripheral ischemic core regions (ICp-ND vs ICp-DP, ICp-ND vs ICp-DP, unpaired t-test, p=0.83, p=0.16, respectively). While confirming previous results of quantitative regional losses of [K+]br in the ischemic core, we show that K+ dynamics within the peripheral and the central ischemic core are heterogeneous and not related to MAP2-assessed neuronal structural integrity: the K+-depleted regions in the peripheral ischemic core regions are presumably closer to glymphatic system and other K+-efflux pathways. Such differing K+ dynamics at the edge of the ischemic core in the hyper-acute period in first hours after ischemic onset possibly relate to the spreading depolarization-mediated expansion of the infarct during the period of secondary brain injury. Peripheral ischemic core regions with less K+ might limit spreading depolarization initiation and propagation if there is insufficient K+ for depolarization to occur and make restoration of parenchymal membrane potential improbable even if the functionality of the Na+,K+-ATPase is restored. Further study of differing K+-dynamics within the ischemic core might lead to a better understanding of ischemic stroke pathophysiology.