Kausik Si, Ph.D - US grants

DESCRIPTION (provided by applicant): Although a large number of experiences produce transient memory only some of them form stable long-term memory. The value of the experience and internal state of the organism plays an important role in determining this transition from short-term to long-term memory. However, how external experience and the internal state of the organism interact at a molecular level to influence memory storage is largely unknown. The synthesis of new proteins is required to form stable long-term memory. CPEB, a family of RNA-binding proteins, regulate activity-dependent protein synthesis in the neuron. In Aplysia, a specific CPEB family member, ApCPEB, is required for the persistence of activity-dependent increase in synaptic strength beyond 24 hours. In Drosophila, a CPEB family member Orb2 is required to stabilize memory beyond 12 hours. In mice and human a CPEB family member CPEB3 has been linked to episodic memory formation. These observations suggest that CPEB family members play evolutionarily conserved role in the persistence of memory. The Drosophila Orb2 gene produces two distinct primary transcripts from two different transcription start sites, encoding the Orb2A and Orb2B proteins. Mutations that selectively inactivate Orb2A protein specifically destabilize long-term memory. We have found that the Orb2A mRNA has two forms; an intron containing unspliced form that does not code for Orb2A protein (Orb2A-np) and a spliced form that codes for full length Orb2A protein (Orb2A-pc). Surprisingly, the Orb2A-np mRNA is the prevalent isoform in the adult brain and exposure to cues that form associative long-term memory in flies, such as exposure of hungry flies to sucrose and odor increases the level of protein coding spliced Orb2A-pc transcript. The Drosophila orthologue of mammalian splicing regulator Nova binds Orb2A transcript and regulates the formation of Orb2A-pc mRNA. In this proposal we intend to dissect the behavioral cues and Nova-dependent molecular mechanism that regulate the level of protein coding Orb2A transcript. We also intend to address when and how long Orb2 proteins are required for persistence of memory. We hope that an understanding of the interaction of experiences and molecules that stabilize memory would be useful to understand addictions and post traumatic syndromes, in which a single experience forms extremely stable memory. Also, we hope to gain insight into the role of regulated mRNA processing in long-lasting memory.

Project Summary Our efforts to delineate the biochemical processes that convert a transient experience to a persistent memory led to the discovery that functional aggregation of a Drosophila RNA-binding protein Orb2A is critical for the animals? ability to form and retain a memory. During the previous grant period, in our effort to decipher regulation of Orb2A aggregation, we discovered that the expression of Orb2A protein in the adult brain is controlled via a unique mechanism. The fully transcribed and polyadenylated Orb2A transcripts retain an intron and are stably expressed in the fly brain. The intron-retained Orb2A mRNA does not code for a protein, owing to multiple in frame stop codons. However, behavioral training that produces long-term but not short-term memory transiently increases the spliced protein-coding mRNA level. This splicing event, important for long-term memory, is mediated by a specific isoform of the splicing regulator Pasilla, psL. Intron retention (IR), a poorly studied form of alternative splicing, was thought to modulate gene expression primarily by degrading the mRNA, trapping it in the nucleus and thereby reducing protein expression. Our findings suggest that IR can be used in a novel way: to hold mRNA in a ?poised state?, which can allow spatial and temporal integration of various signals to acutely increase protein expression to regulate memory. These observations led us to ask how prevalent IR is in the adult Drosophila brain, and what is the molecular basis of IR in the brain and how experience interacts with the intron-retention mechanisms to control protein expression and memory. We have performed poly(A) RNA-sequencing from wild type fly brains. The analysis provided some surprising results. First, 8.9% of all detectable introns, corresponding to 3031 genes, show ? 25% retention in wildtype fly brain. These includes 55 genes with role in learning and memory. A recent genome-wide study revealed ~6% of fully transcribed and polyadenylated transcripts in the mouse brain also retains an intron. Remarkably, the mouse orthologues of some the Drosophila genes, including Orb2 orthologue CPEB2, also retains intron. Second, like Orb2, aggregate-prone prion-like proteins are significantly overrepresented in the intron retained genes. This includes a novel prion-like phosphatase we have recently identified. From adult brain, we have identified proteins that bind to a retained intron and interacts with psL. In the current proposal we intend to investigate how memory-related stimuli activate p38a, a putative kinase of psL, how behavioral stimuli effect psL interaction with its binding partners, what sequence motifs in a retained motif influence intron retention, how intron retention controls expression of aggregate-prone proteins and how perturbation of each of these processes effects animals ability to form and retain memory. These studies would provide insight into the IR-mediated control of protein expression, a novel gene regulatory mechanism, as well as the molecular basis of memory and the regulation of aggregate-prone proteins.

Project Summary The aggregate failure of pre-clinical and clinical trials in AD has demonstrated that an improved fundamental understanding of memory and how the molecular components of memory are altered in the AD disease process is necessary to develop effective treatment. The broad objective of the project is to identify the biochemical substrates of long-lasting memories in mammals. The current proposal focuses on a family of RNA-binding, cytoplasmic polyadenylation element binding protein (CPEB), that stabilizes memory in invertebrates and mice. Remarkably, CPEB family protein forms non-disease-causing amyloidogenic aggregates and aggregation of CPEB is necessary to stabilize memory. As amyloids are typically linked to disease states, the question remains how similarly structured A?42 or Tau proteins can have opposing effects on memory. Therefore, to develop a better understanding of the relationship between amyloids that support memory and amyloids that disrupt memory, we will use a variety of techniques to solve the structure and function of the CPEB family members, CPEB2 and CPEB3, in human and mice. In Aim 1, we will use cryo- electron microscopy to solve the structure of CPEB aggregates from fresh human frontotemporal lobe tissue collected from 25-50-year-old human subjects undergoing tissue removal under the standard of care for their disease. These tissues would have been otherwise discarded. In Aim 2, mice lacking the ability to form aggregates of CPEB2 and CPEB3 will be trained and tested in a one-trial inhibitory avoidance task to assay their ability to form, maintain, and recall memory. In Aim 2 we will also investigate the consequence of CPEB2 and CPEB3 aggregation in translation of mRNA encoding synaptic proteins. The results would be the first to provide direct structural analysis of a functional amyloid linked to memory in mammals, the structural distinctions, if any, between functional and toxic amyloid in the human brain, and precisely link CPEB2 and CPEB3 aggregation and activity to animals? ability to form or stabilize memory. This knowledge would provide the foundation to investigate in the future how toxic amyloids of A?42 or Tau specifically perturb memory.