Regulation of phosphorylation of c-jun N-terminal kinases (JNKs) in neonatal mouse hypoxic ischemic encephalopathy model

Abstract

Jun N-terminal kinases (JNKs) can be phosphorylated and activated in response to diverse stress signals in vitro, such as UV irradiation, hyperosmotic shock, and have been implicated in the pathogenesis of glutamate excitotoxicity and cerebral ischemia in adult animal models. The three isoforms (JNK1, 2 and 3) also have differential roles in regulating developmental neuronal apoptosis. However, their role(s) in postnatal brain injury has not been addressed. We adopted neonatal mouse hypoxic ischemic encephalopathy model to study their role in mediating delayed cell death after injury. Western blot analysis with phosphor-specific JNK antibody (recognizes all isoforms of JNK) revealed high basal level of phosphorylated JNK (p-JNK) in the mouse brain hippocampal lysates on postnatal day 7. At the conclusion of the HI experiments (permanent unilateral common carotid artery ligation plus 30 min of hypoxia – designated time 0 h), the ipsilateral p-JNK level was reduced to 1/10 that of contralateral (which was reduced to about 80% of control animals). Such reduction persisted for 30 min., then dramatically rose from 45 min to a peak at 3 h and a secondary fall from 6 to 18 h after HI. Given that there could be differential changes of individual p-JNK isoforms, further work is in progress to define the relative proportion of these isoforms throughout the course. The spatial and cellular changes in p-JNK were investigated with immunohistochemical study. At 1 h post HI, p-JNK staining was not detectable in the majority of cells, likely reflecting that the high basal p-JNK on Western blot was contributed by many neural cells. However, positive cells were evident in the hippocampus with the temporal change compatible with that revealed by Western blot. Specifically, in the ipsilateral hippocampus at 1 h after HI, cytoplasmic p-JNK was detected in neurons in the pyramidal cell layer, and the granule cell layer of dentate gyrus. At 3 and 6 h after HI, the number of positive neurons increased, some with condensed nuclei and strong p-JNK immunoreactivity throughout the whole cell. At 24 h after HI, the staining intensity and cell number reduced, except for some hippocampal neurons retaining the strong cytosolic p-JNK immunoreactivity. Parallel immunostaining with anti-phosphorylated-c-Jun antibody (phospho-Serine-73) showed a similar pattern of changes after HI. In the ipsilateral hippocampus at 3 h after HI, increased nuclear immunoreactivity in the pyramidal cell layer of hippocampus and the granule cell layer of dentate gyrus, with much lower intensity over the contralateral side. At 6 and 24 h after HI, the staining intensity decreased. To investigate if JNK phosphorylation may be associated with caspase activation, double immunostaining for phospho-JNK and active caspase-3 was performed. At 6 h post HI, the positive active caspase-3 immunoreactivity present in some ipsilateral hippocampal pyramidal neurons and the granule neuons of dentate gyrus, with condensed and/or apoptotic nuclei, did not colocalise with p-JNK staining. By 24 h after HI, the lack of colocalization between p-JNK and active caspase-3 persisted in the pyramidal cell layer of hippocampus. However, some of these p-JNK positive cells had condensed/apoptotic nuclei. In contrast, almost all neurons in granule cell layer of dentate gyrus were double stained for p-JNK and active caspase-3 staining. Thus, while our data support that JNK phosphorylation (activation) may lead to caspase-3 activation in dentate gyrus, but not in pyramidal cell layer, it remains possible that other caspase-dependent or even caspase-independent apoptotic death pathway could be differentially recruited by the various p-JNK isoforms. Complicating such theory is the fact that some strongly p-JNK positive neurons maintained healthy nuclear morphologies, casting an additional possibility of specific JNK isoform phosphorylation being pro-survival for selective neurons after HI insult. Along this line JNK1 and 2 have been shown to be pro-survival during neuronal development while JNK-3 being pro-apoptotic. We therefore conclude that further detailed dissection of the specific involvement of different JNK isoforms is crucial in defining the roles of JNK pathways in neonatal HI insult