The cricket exhibits a highly unusual pattern of compensatory dendritic growth in the adult central nervous system, and serves as an excellent animal model for the study of successful neuronal recovery following injury. In normal crickets, neurons that receive auditory input from one cricket “ear” project up to the brain unilaterally, without crossing the cricket’s “midline” (Figure 1A). Removal of an ear by amputation of a foreleg, however, results in neurons crossing over to form new connections with the existing auditory neurons on the opposite side of the body (Figure 1B), leading, in turn, to recovery of neuronal function. The Horch lab previously correlated upregulation of sema2a—the gene that codes for a diffusible semaphorin protein in invertebrates--with this compensatory growth. Semaphorins are a class of proteins that have been shown to play a role in dendritic growth in a wide variety of species, working as either a chemoattractant or chemorepellant in guiding growth of dendrites during development. As a result, this study tested the hypothesis that sema2a’s upregulation is necessary for the compensatory growth seen following deafferentation.
We first designed a protocol for RNA interference (RNAi) to induce knockdown of sema2a. Double-stranded RNA corresponding to the sema2a gene was injected into juvenile crickets to induce knockdown of sema-2a in the adult cricket’s prothoracic ganglion. Using both semi-quantitative-reverse transcriptase PCR and Quantitative Real-Time PCR to assess the resulting levels of sema2a expression, we confirmed that our method of RNAi significantly decreased sema2a levels (Figure 1A-B). Further analysis using QPCR demonstrated that the knockdown effect was induced by sema2a dsRNA specifically, and lasted at least 7 days post-deafferentation. In addition, sema1a levels were unaffected, demonstrating the high specificity of RNAi (not shown). Together, these results provided strong evidence of large, consistent and specific sema2a knockdown.
Next, we compared dendritic and axonal growth in these sema-knockdown crickets following deafferentation to that seen in control deafferents. Because the compensatory growth response in deafferented crickets has been shown to occur over the course of several days, we chose to assess neuronal morphology seven days post-deafferentation. Interestingly, deafferented AN2 dendrites appeared to cross the midline in sema-injected deafferents as they do in non-injected controls (not shown). In contrast, N5 axons in sema-injected deafferents exhibited an abrupt “wall”-like formation at the midline compared to controls (Fig. 2A-C). Though further quantification of these results is needed, we conclude that sema2a may influence N5 morphology in deafferents while having no direct influence on AN2 morphology.
Faculty Mentor: Hadley Horch
Funded by the Arnold and Mabel Beckman Foundation