| dc.description.abstract | Habituation is defined as a reduced behavioral response to a stimulus after repeated or continuous exposure. Long-term habituation (LTH) is a long-lasting form of non-associative memory which is dependent on de novo protein synthesis.
Activity-dependent translation of mRNA in neurons is fundamental not only at the cell-wide level but specifically at synaptic level, where translational regulation is dependent on ribonucleo protein granules RNPs.
RNPs are membraneless organelles which allow cytoplasmic compartmentalization of mRNA and proteins. RNPs are formed by intrinsically disordered domains (IDR) of RNA-binding proteins (RBPs), to allow mRNA localization and to enable spatial-temporal regulation of protein synthesis.
The aim of this thesis is to investigate spatial and temporal requirements of RNP-dependent de novo protein synthesis in LTH formation in Drosophila melanogaster.
The RBP Ataxin-2 (Atx2), and in particular its C-terminal IDR (cIDR), has been shown to influence RNP assembly in vivo, hence protein synthesis, and LTH formation. However, the particular spatial and temporal requirements of Atx2-dependent LTH formation are not well understood. To address this question, I established and characterized a new behavioral paradigm to assess olfactory avoidance LTH.
This new set up allows for an automated way to study group behavior. I have thoroughly optimized training and testing conditions, confirming that the paradigm can be used with different habituation odors on different fly strains. I have also assessed three of the defining characteristics of habituation: spontaneous recovery, specificity, and dishabituation. These characteristics are important to prove that decrease in response after long exposure to a stimulus is not due to motor fatigue or sensitization, but to habituation.
Lastly, I decide to evaluate the molecular and circuit characteristics of LTH. The rutabaga (rut) gene is one of the most well-known and well-studied memory mutant in Drosophila.
In accordance with the literature, I have shown that rut hypomorph mutant flies show impaired LTH and that the inhibitory local interneurons (LNs) are required for LTH formation. These results show that the new developed method to assess LTH can be reliably used to investigate habituation.
Deletion of Atx2 cIDR impairs both RNPs and LTH formation. Based on these findings, I designed and developed atx2 flip lines to investigate the spatial and temporal requirements of Atx2 cIDR in LTH. These lines allow the conditional exclusion or inclusion of Atx2 cIDR and allow cell-specific control of Atx2 cIDR presence. I also developed an atx2 flip line which allows the conditional exclusion of the entire atx2 locus, to compare the effects of the absence of the cIDR domain versus the absence of the whole Atx2 protein.
I designed in silico and made the atx2 flip plasmids using several cloning techniques. These plasmids were then used to replace the atx2 locus in flies using CRISPR/Cas9 homologous recombination technology. Following, I validated the designed lines using sequencing, morphological and behavioral analyses. However, these lines could not be used for following behavioral experiments, due to low eclosure rate of the experimental animals and the aims of the project were adjusted.
Two separate lines of research have been pursued: 1) to further dissect the role of Atx2 and its cIDR in RNPs formation in the ovaries; 2) to investigate the interaction of Atx2 with Imp as molecular partner for LTH formation.
RNPs are very well characterized in the ovaries, which makes them an exceptional model to study the role of cIDR in molecular interactions. Thus, I investigated the role of Atx2 and its cIDR in the assembly of known ovarian RNP granules proteins, such as: Me31b and Cup. I found that exclusion of the atx2 locus in the female germline completely abolish Me31b and Cup RNPs formation. However, excluding just Atx2 cIDR or knocking down the protein using RNAi, impairs but does not abolish RNP formation. These results posit Atx2 and its cIDR as regulators of RNP granules in the female germline.
Furthermore, I decided to investigate the possible interaction between Imp and Atx2 in LTH. I decided to investigate the Imp-Atx2 interaction because: Imp is found in RNP granules in the brain, Imp colocalizes with Me31b in neuronal RNP; additionally Imp has a disordered domain which deletion causes a specific defect in long-term but not short-term memory. Atx2 and Imp null trans-heterozygotes mutants show LTH impairment, which can be rescued with genomic expression of Atx2. Memory impairment in trans-heterozygotes assay indicates a trans dominant phenotypic interaction between the two proteins, suggesting that Atx2 and Imp work in the same molecular pathway for LTH formation.
Following, I decided to investigate whether Imp alone could influence LTH. For this reason, Imp requirement was tested in projection neurons (PNs) using RNAi. Imp k/d significantly decreases LTH performance but does not abolish LTH formation. This is consistent with previous research showing that Imp is a ?shell? component of RNPs, and it is not found at the RNPs core. Hence, Imp k/d will affect the RNP external shell leaving its core unperturbed and similarly will impair but not block LTH.
This work provides novel insights not only into the molecular regulation of LTH, but onto the mechanism of action of Atx2, improving our understanding of fundamental biological processes of localized protein synthesis, with implications in development and neuroscience at large. | en |
