PROJECT SUMMARY This proposal aims to record and store neural activity into DNA/RNA strands with high spatial and temporal resolution, enabling neural activity recording densely within local circuits, yet broadly across large scale neural circuits, and potentially across the entire brain. Current technologies require tradeoffs in coverage, spatial resolution, or temporal resolution. Recording with full coverage across extended circuits – and potentially even the whole brain -- requires distributed, high bandwidth capabilities that cannot be met with current optical or electrical measurements. To address this recording limitation, we propose to use DNA and RNA polymerases to record neural activity and store it as sequence information in the neuron, to be recovered and/or analyzed after the experiment. DNA and RNA polymerases extend their substrates with multiple incorporations per second, giving the desired temporal resolution, comparable to calcium imaging, but potentially far more scalable in the sense that volumes much bigger than can feasibly imaged could be captured. The sequence information can be stored locally, avoiding the need for invasive probes that cause brain damage and/or gliosis, and the sequence can be recovered with single molecule resolution by next-gen sequencing or in situ sequencing. In this proposal, we will evaluate a unique class of DNA and RNA polymerases that do not use a template to determine base addition and whose base addition preference can be made sensitive to Ca2+ levels by protein engineering. We hypothesize two strategies may enable this base-specific incorporation based on Ca2+ levels: (a) terminal deoxynucleotidyl transferase (TdT) adds bases in a random process that is biased toward dGTP by the local Ca2+ level, or (b) polyU RNA polymerases (an analog to the more common polyA polymerase) can be made to interact with RNA strands in a Ca2+ dependent manner by creating polyU polymerase-CaM fusion and a M13-fused RNA-binding domain. In this way, high Ca2+ will cause the polyU polymerase to localize to the mRNA and preferentially add UTPs. In both strategies, regions of an extended strand that are overrespresented in dGTP or UTP will correspond to times when Ca2+ levels were high. If either of these approaches are successful, it would enable a genetically encoded neural recorder capable of monitoring every neuron in the brain. To evaluate the viability of template-independent polymerase as neural recorders, we have devised a pipeline approach that enables rapid end-to-end evaluation of different protein engineering strategies. This pipeline approach allows us to rapidly explore a large engineering space, where positive results can be rapidly pursued, and iterative strategies can be used to overcome obstacles. At the end of the project, we will have demonstrated proof-of-concept, and will be well positioned to continue onward with optimizing robustness, simplifying the process for end-users, and disseminating the nano-recording technology. With members of our team having collectively distributed molecular tools to thousands of labs, we are excited to lead a fast-paced project potentially capable of radically accelerating neuroscience progress.

PROJECT NARRATIVE Treatment of a range of neurological and mental health problems is hampered by an insufficient understanding of high resolution neural activity in the brain. This proposal will develop DNA/RNA based recording techniques that will record Ca2+ timecourses (associated with neuron firing) and store that information in DNA or RNA strands to be sequenced later. This technique can provide future researchers with far denser and broader neural activity maps than previously possible, revealing mechanisms underlying brain diseases.