Trans-neuronal regulation of cortical apoptosis in the adult rat olfactory system

Abstract

Previous work has identified a population of neurons within the anterior piriform cortex that undergo rapid apoptosis following de-afferentation by olfactory bulbectomy in adult rats. The specific initiation signal for apoptosis in this paradigm is unknown, but may include an activity-dependent trans-neuronal cascade. The present report examined the effect of adult-onset unilateral naris occlusion, which reduces olfactory bulb afferent excitation of piriform cortex, on apoptosis (terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling [TUNEL]) in the rat anterior piriform cortex. Adult Long-Evans hooded rats received unilateral naris occlusion or a control manipulation and were sacrificed after 1, 5, 7, 10 or 20 days later. For comparison, a second group of rats received a unilateral bulbectomy and were sacrificed 24 h later. Counts of TUNEL-stained cell profiles were performed for layers I/II and layer III of the anterior piriform cortex ipsilateral and contralateral to the manipulation. The results confirmed that unilateral bulbectomy produced a dramatic increase in TUNEL labeling in layers I/II of the ipsilateral piriform cortex 24 h after bulbectomy. Unilateral naris closure also produced enhanced TUNEL labeling, although the magnitude of the effect was less than that produced by bulbectomy, and enhanced TUNEL labeling was apparent both ipsilateral and contralateral to the sealed naris compared to controls. Deprivation-induced TUNEL labeling was detectable by 24 h post-closure, peaked at 5 days and was no different from controls by 20 days post-closure. Neither bulbectomy nor naris closure affected TUNEL labeling in layer III. Together, these results suggest that there is a population of superficial cells in piriform cortex whose survival is tightly regulated by sensory input.

Introduction

The mammalian olfactory system has been shown to be an outstanding model for understanding the role of experience in development and maintenance of neural circuits. Rates of neurogenesis and apoptosis [9], [11], [12], [21], [32], [36], neurochemical expression [5], [8], [30], [40], dendritic elaboration and axonal projection patterns [6], [28], [38], as well as sensory and synaptic physiology [3], [18], [34], [39] have all been shown to be regulated by sensory input, often even in adult animals. For example, olfactory receptor neurons, whose axons form the olfactory nerve input to the olfactory bulb, are continually replaced throughout life. Reductions in airflow and/or odor stimulation due to external naris occlusion results in a significant reduction in the rate of neurogenesis of receptor neurons [11]. Furthermore, receptor neuron activity is required for normal receptor axon targeting of the glomerular layer [41]. In the olfactory bulb, which receives exclusive input from the ipsilateral olfactory epithelium, reduced olfactory stimulation reduces rates of adult neurogenesis of granule cell interneurons [9], [12], while enriched odor stimulation enhances neurogenesis [36].

Olfactory bulb mitral cells project directly to the piriform (olfactory) cortex, where they synapse onto several types of target neurons including superficial and deep pyramidal cells and semilunar pyramidal cells [20]. The superficial and deep pyramidal neurons, with somas located in cortical layers II and III, respectively, receive direct input from the ipsilateral olfactory bulb, as well as indirect input from the contralateral olfactory bulb via commissural fibers. Many of these cells in turn project back to the ipsilateral olfactory bulb. Semilunar cells, with somas located in cortical layer IIa, are unique in that (1) they have large dendritic spines primarily within the termination zone of the lateral olfactory tract suggesting they may be primarily responsive to ipsilateral input, and (2) they do not send axons back to the olfactory bulb [20]. In accordance with this potential reliance on ipsilateral input, semilunar cells have been shown to be especially sensitive to perturbation of the ipsilateral olfactory bulb. For example, in adults, complete removal of the cortical afferent by unilateral bulbectomy produces a rapid, pronounced wave of apoptosis selective to semilunar cells [7], [14], [21].

This remarkable dependence of cortical cell survival on afferent input could be due to a variety of mechanisms, which may include a role for normal patterns of afferent (glutamatergic mitral/tufted cell) activity. In fact, naris occlusion, which reduces overall olfactory bulb activity [18], [23] and disrupts the normal respiration entrained activity of mitral/tufted cells [34], produces a decrease in semilunar cell dendritic elaboration during development [38]. The present study examined the role of sensory input on piriform cortex cell survival in adult rats using terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) as a marker of apoptosis, and compared the effects of sensory deprivation with olfactory bulbectomy. The results suggest that unilateral naris occlusion produces a wave of cortical apoptosis that appears within 24 h of naris occlusion, peaks around 5 days and is complete within 3 weeks, and demonstrate that reductions in sensory input can lead to increased apoptosis in adult primary sensory cortex.