EXPRESSION OF FOS AND MICROGLIAL REACTION FOLLOWING OCCLUSION OF THE MIDDLE CEREBRAL ARTERY IN RATS

Abstract

Occlusion of the middle cerebral artery (MCA) in adult rats leading to consistent infarctions in the areas it supplies is a useful experimental model of focal cerebral ischaemia with characteristics of stroke in human. Most of the histopathological studies of focal cerebral ischaemia have focused on neuronal damage, induced expression of Fos-like immunoreactivity (Fos-LI) in neurons and microglial reaction in the primary infarction area as well as in the associated retrograde thalamic degeneration secondary to the cortical infarction. It remains to be ascertained whether neurons and glial cells in the spinal cord far removed from the primary infarct would also respond to or be activated by the cerebral ischaemia. In fact, some authors have pointed out the difficulty and potential pitfalls in detecting ischaemic damages in areas other than the site of primary infarct (Yamamoto et al., 1987; Iizuka et al., 1989a; Nagasawa and Kogure 1990). The present study was therefore carried out to investigate the expression of Fos-LI and glial reaction in areas other than the primary site of ischaemic infarct with special reference to the spinal cord, a topic which seems to have eluded the attention of many studies in MCA occlusion. By using tetrazoolium blue chloride (TTC), Nissl and Fink-Heimer stainings and immunocytochemistry, the present study has largely confirmed previous results of the ischaemic lesion in the ipsilateral cerebral cortex following MCA occlusion. TTC and Nissl staining have shown that the ischaemic damage in the present study affected specifically the hindlimb area in the ipsilateral cerebral cortex. Fink-Heimer staining has demonstrated unequivocal degeneration of axons in the corticospinal tract along its pathways in the spinal cord at various time intervals following MCA occlusion. Degenerating nerve fibres were traced as distal as the dorsal and ventral horns at lumbar levels contralateral to the arterial occlusion. By OX-42 immunocytochemistry combined with immunoelection microcopy, drastic microglial reaction was observed in the ipsilateral cerebral cortex and thalamus at 1 and 3 days after MCA occlusion. Induced expression of Fos-LI was confined mainly to neurons in layers II and IV, and to a lesser extent, layers I,Ill and V in piriform, frontoparietal and cingulate cortex on the operated side. Fos-like immunoreactive cells were sparsely distributed in the cerebral cortex on the contralateral side and in sham-operated animals. Fos-LI in the above-mentioned regions was first detected at 1 hour after MCA occlusion, being most vigorously expressed at 2-4 hours, but was diminished by 2 days after ischaemia. In the core territory delineated by the characteristic "ischaemic penumbra", Fos-Ll was absent at all time points after the arterial occlusion. Following MCA occlusion, induced expression of Fos-LI was also detected in some hypothalamic, medullary and thoracic spinal cord neurons. Thus, at 1 and 2 h, especially in the latter time interval after MCA occlusion, Fos-LI confined to the cell nucleus, was detected bilaterally in cells of the supraoptic nucleus and the paraventricular nucleus of the hypothalamus, the nucleus of the solitary tract, the area postrema and ventrolateral medulla. A few Fos-like immunoreactive neurons were observed in the nucleus raphe pallidus and obscurus, and in the intermediolateral nucleus of the thoracic spinal cord. In the corresponding areas in sham-operated animals, Fos-like immunoreactive neurons were sparsely distributed or absent. Colocalization study showed that a variable number of the Fos-Iike immunoreactive neurons in the nucleus of the solitary tract and ventrolateral medulla coexpressed tyrosine-hydroxylase (TH) immunoreactivity . Such double labelled neurons appeared to be more common in the latter site. It is suggested that the induction of Fos-LI in neurons of the hypothalamus, medulla and thoracic spinal cord was linked to cardiovascular regulation following the middle cerebral artery occlusion. A striking feature after MCA occlusion was the induction of Fos-LI in both the dorsal and ventral horn neurons of the spinal cord at the lumbar segment. The labelling was often bilateral but generally more substantial contralaterally. In the ventral horn, some of the Fos positive neurons were confirmed to be somatic motoneurons innervating the tibialis anterior muscle of the lower extremity contralateral to MCA occlusion as shown by their retrograde labelling with horseradish peroxidase injected into the muscle. Fos-LI was absent in the ventral horn of the spinal cord at cervical, thoracic and sacral segments both in experimental and sham-operated rats. These findings suggest that the expression of c-fos may be used as a sensitive transneuronal marker for the study of neuronal activity in the spinal cord elicited by brain damage, viz. focal cerebral ischaemia. When combined with horseradish peroxidase as a retrograde tracer, the method has been shown to be a useful technique for studying the connectivity of the corticospinal motor system. Another remarkable feature following MCA occlusion was the response of microglia and astrocytes, as detected immunohistochemically by the monoclonal antibody OX-42 and anti-glial fibrillary acidic protein, respectively, in the lumbar spinal cord. At 1 and 2 days after MCA occlusion, OX-42 immunoreactivity of microglia in both the contralateral dorsal and ventral horns of the lumbar spinal cord was moderately increased compared with cells of the ipsilateral side. The microglial reaction was progressive with some cells transformed into amoeboidic form considered to be macrophages at day 3. By 5 days, many of the reactive microglia notably in the ventral horn appeared to encircle the soma of motoneurons. At 7 days, microglial reaction had subsided while astrocytes in the same area now became hypertrophied to replace the perineuronal microglia. The microglial response in both the cerebral cortex and lumbar spinal cord was effectively reduced by the NMDA receptor antagonist MK-801, suggesting that it may be mediated by glutamate. It is speculated that excessive glutamate may be released at the terminals of ischaemic corticospinal neurons which would act through NMDA receptors on the postsynaptic spinal cord neurons. Immunoelectron microscopy confirmed the degeneration of axons in the dorsal horn. The most dramatic feature,however, was the occurrence of some neurons undergoing degeneration or total lysis which appeared to be engulfed by reactive microglia at 3 and 5 days after MCA occlusion. It is suggested that as with microglial reactivity mentioned above the neuronal degeneration may also be attributable attributed to excessive release of glutamate from the axonal terminals of corticospinal neurons affected by the cerebral ischaemic insult. Degeneration of dorsal horn neurons could also have resulted from their deafferentation following ischaemic lesion of the corticospinal fibres which are the main source of afferent inputs. Compared with those in the dorsal horn, the ventral horn neurons remained ultrastructurally intact. Though appeared morphologically activated, microglial cells associated with the neuronal soma were not phagocytic suggesting that their activation may be neuroprotective in function. How this may be effected would be scope for future study, although from a speculative point of view, this may be mediated through secretion of nerve growth factors beneficial for the ventral horn neurons. A better understanding of the functional relationship between neurons and microglia at the spinal cord level after MCA occlusion would help design strategies for potential therapeutic interventions. The present study has demonstrated that prolonged treatment with the xanthine derivative propentofylline (PPF) (1-(5 '-oxohexyl)-3-methyl-7 propylxanthine) can interfere with the potentially neurotoxic cell properties acquired by the pathological activation of microglial cells. The results showed that daily treatment of PPF beginning at 1 day after MCA occlusion for two or four consecutive days markedly suppressed the microglial response as well as prevention of formation of amoeboidic microglia. This has greatly amplified the potentiality of PPF used as a neuroprotective drug against microglia-related neuron damage induced by cerebral ischaemia. This effect could be of particular clinical significance since it was achieved by treatment after pathological cell activation had already been initiated. Thus, chronic treatment with PPF may provide a neuroprotective therapy for those diseases in which an ongoing activation of microglial cells is thought to be relevant for neuronal damage.