Denise J. Cai - US grants

DESCRIPTION (provided by applicant): Numerous studies have demonstrated the importance of CREB in memory consolidation. However, recent findings suggest that CREB also plays a key role in memory allocation. Increasing the levels of CREB increases the probability that a given neuron in the amygdala will be recruited into a memory trace, while decreasing the levels of CREB has the opposite effect. Previous results also suggest that CREB affects memory allocation by altering neuronal excitability, whereby high levels of CREB increases the intrinsic excitability of neurons in the amygdala which biases these cells towards being included in the memory trace. While there is convincing evidence that CREB and neuronal excitability are important for memory allocation in the amygdala, it is unclear whether these principles generalize to other brain regions important for memory, such as the hippocampus. I propose to directly test this hypothesis by manipulating a subpopulation of CA1 neurons with viral CREB to see if the representation is biased towards those neurons with increased CREB. Additionally, I will examine the temporal dynamics of CREB activation and neuronal excitability in CA1 following acquisition of a context memory using Western blot analysis and whole cell patch recording, respectively. Lastly, I will examine the co-allocation of two memories with transgenic TetTag mice, which can label activated neurons at 2 different time points. My prediction is that if CREB biases memory allocation, then increased CREB activation induced by one hippocampus-dependent memory should bias the allocation of a second memory to many of the same neurons recruited to store the first memory. Our previous studies in the lateral amygdala pioneered the field of memory allocation. While this field has been very exciting, it has been exclusively limited to the amygdala and for this field to move forward, it is essential that we can generalize our findings to other brains structures and memory types. The results from these proposed studies will give us a better idea of how memory allocation processes affect the integration and storage of information. With this knowledge, we can better develop targeted treatments for memory disorders, including Alzheimer¿s disease, as we already know there is aberrant CREB-mediated gene regulation in this population. PUBLIC HEALTH RELEVANCE: Alzheimer's disease (AD) is the most common form of dementia in the aging population and memory impairment is the principle defining feature. There is strong evidence that these patients have an aberrant regulation of CREB in the brain, which could be related to the memory loss. The results from these proposed studies will give us a better idea of how CREB affects the integration and storage of information. With this knowledge, we can better develop targeted treatments for memory disorders, including Alzheimer's disease.

PROJECT SUMMARY/ABSTRACT The proposed research project will address two fundamental research questions about information processing in the biological brain. We will begin addressing this question by focusing on how the hippocampus accumulates different spatial maps (i.e. linear tracks) across a lifetime by a novel i imaging technique, the wire- free miniature microscope (Miniscope), for long-term in vivo calcium imaging in untethered, freely behaving animals. We will investigate the question of orthogonal vs. integrated coding of spatial maps and their impact on memory capacity and efficiency in young adult mice. Does learning new (more) maps increase the efficiency of neural encoding, such that there are fewer cells encoding redundant spatial information? Is a decreasing number of cells required to encode each subsequent map to prevent the hippocampus from reaching capacity? Can we calculate the memory capacity of a particular network or of the hippocampus overall? Next, to explore the question of how these processes may change across a lifetime, we will investigate whether or not the same rules for coding spatial maps apply in middle aged mice with varying number of memories for different linear tracks previously accumulated. As we age, is there simply less ?available space? in our brains as many of the neuronal resources have been taken up by the previous memories accumulated across a lifetime and, consequently, more competition for the neural resources required to encode new information? Is generalization, at least in part, a consequence of a change in storage strategy, where details are dropped in order to maximize efficiency. We will use state of the art modeling techniques as well as causal biological manipulations to probe these questions. Answering these questions regarding how the brain optimizes storage capacity for information will contribute to better understanding a broad spectrum of brain disorders that have problems with relational memory, including Schizophrenia, Depression, Dementia, Alzheimer's disease. Furthermore, by understanding the rules the brain uses to optimize storage capacity and perform efficiently, these rules and principles can be applied to developing neuroprosthetics that can treat patients with brain damage, including stroke, epilepsy, traumatic brain injury.

PROJECT SUMMARY/ABSTRACT Post-traumatic stress disorder (PTSD) symptoms are often triggered by environmental stimuli that were not directly present during trauma, but which nevertheless elicit strong fear responses. Thus, understanding how fear spreads to non-trauma associated stimuli is paramount to PTSD treatment. Recently, I found that different contextual memories encoded close in time (i.e., 5 hours) are linked by sharing an overlapping neural ensemble, making recall of one memory more likely to trigger recall of another memory encoded close in time. My lab has preliminary evidence that increasing the negative valence of a memory extends the temporal window of retrospective linking, such that the fear of an aversive context is linked with neutral stimuli experienced days prior. Linking aversive experiences to past events is ecologically valuable, as the past has predictive value for the future. However, overlinking traumatic experiences to everyday memories could be maladaptive and may promote PTSD symptoms. To observe dynamic ensemble activity across long time scales, I have co-developed a wire-free Miniscope which allows in vivo calcium imaging in untethered, freely behaving mice. Using the Miniscope, we found that neutral and aversive memories are linked by an increased overlap in their hippocampal neural ensembles, and that this overlap emerges sometime after learning. We will therefore test the hypothesis that the enhancement of memory-linking by aversive experience emerges through co-reactivation of memory representations during an offline consolidation period. We will test the sufficiency and necessity of ensemble reactivation, as well as synaptic plasticity, in the linking of aversive memories and safe memories. This work could therefore shed fundamental light on the spread of fear in PTSD and how it might be clinically targeted.