Darcy S. Peterka, Ph.D. - US grants

? DESCRIPTION (provided by applicant): Improving our understanding of the functional circuitry of the brain has important and manifold implications for our understanding of mental health, as well as fields like consciousness and computing. In the last decade, optical techniques have arisen that allow both recording and control of targeted neurons for brain mapping, and many of the best of these techniques employ multiphoton microscopes with spatial light modulator (SLM) technology and complex algorithms to shape the light and analyze increasingly large neural microcircuits in three-dimensions (3D). SLMs can arbitrarily shape the wavefront of light to create multiple independently targeted beams in 3D to control groups of neurons, with the maximum number of studied neurons being limited primarily by the laser power on the SLM. Because the SLM can mimic nearly any optical element, these versatile tools also provide additional capabilities when incorporated into microscopes, such as adaptive aberration correction and remote focusing. Despite the potential for SLMs to revolutionize the microscopes used in neuroscience, their adoption remains limited by the difficulty in incorporating the SLM into the expensive multiphoton microscope platforms used by investigators and by the complexity of integrating SLM control into the microscopy software. In this Phase II effort, Boulder Nonlinear Systems (BNS) and Dr. Darcy Peterka and the Yuste laboratory at Columbia University will address this barrier by developing a user-friendly bolt-on SLM module for existing multiphoton microscopes along with full software integration of the SLM into both open-source and commercial microscopy software. This work will leverage knowledge gained during the Phase I development of the Pocketscope, a portable and low-cost SLM microscope for simple in vitro neuroscience studies, and integrate close feedback from a range of industry partners and leaders in neuroscience. As part of this work, BNS will also improve the speed, power handling, and reliability of the SLMs and utilize their strategic commercial partner, Meadowlark Optics, to bring down SLM cost and improve software integration. Successful completion of this project will result in the new SLM-based microscope module, platform- indpendent software integration, and improved SLM joining the Phase I Pocketscope to provide a suite of powerful tools, each with their own impact and commercial niche, capable of transforming the optical exploration of neural networks.

Optimal calcium imaging with shaped excitation Understanding information flow in the brain is dependent on simultaneously recording the activity of large neuronal populations. It seems impossible to interrogate neurons serially, and still image large populations of neurons with high temporal resolution and high signal to noise. This is linked to the inverse relationship between volume scanned, and the signal collected per voxel, at fixed spatial and temporal resolution. However, this is not a hard limit. The goal of most functional imaging is to recover and assign activity signals from neurons; here we demonstrate that nearly all past approaches have dramatically oversampled spatially to create human-interpretable images. This is not necessary ? once the spatial footprints of the observed neurons are known, constrained non-negative matrix factorization methods can extract highly accurate temporal activity signals from very low spatial resolution movies. Reducing the number of samples required in imaging allows us to significantly speed up acquisition. In this proposal we introduce a new fast computational imaging method, leveraging modern computational demixing methods with simple optical hardware to increase imaging speeds by an order of magnitude. We proposed to use modern spatial light modulator systems to provide a flexible and powerful tool for optically implementing our proposed spatial downsampling approach, while taking better advantage of laser power and avoiding standard problems with diffraction- limited imaging caused by limited dwell times on the sample. The resulting combined hardware- software solution will be inexpensive, easy to implement and maintain, and widely applicable in the hundreds of labs currently using multi-photon imaging methods. Thus, the proposed approach will enable a critical leap towards achieving the goals of the BRAIN initiative.

The brain remains one of the most mysterious systems in the universe ? one critical challenge in gaining understanding has been that its constituent elements have features that span across many length scales, in three-dimensions, from the microscopic to the microscopic. In the last few years, new developments in tissue clearing methods and microscopy promise to allow the exploration of the brain over its natural range of scales. We propose to acquire a high performance multi-wavelength lightsheet microscope for a multi-user cellular imaging core that is capable of imaging centimeter size brain and tissue sections while maintaining sub-cellular resolution. This versatile instrument can be used with all currently developed tissue clearing agents, as well as with live tissue, and will be applied to myriad systems that span critical Neuroscience questions, from the role of molecules and genes in development, to the role of experience and learning, in shaping neural circuits.

Summary/Abstract (30 lines) The mission of the Advanced Imaging and Instrumentation Core will be to deliver the necessary technical skill, training and resources to enable the complex in-vivo research projects proposed. The current U19 proposal is only possible thanks to very recent developments in in-vivo imaging, recording and manipulation techniques that have enabled real-time studies in awake behaving mice. In particular, it is essential to leverage in-vivo, awake measurement techniques to study motor control since one must be able to measure associated motor output and task performance. Our Advanced Imaging and Instrumentation Core is composed of key experts in the development and deployment of novel technologies, 2 of them BRAIN Initiative grant awardees. However, it is important to note that the technologies to be deployed here will not require significant high-risk innovation, but rather represent the integration and implementation of recently demonstrate breakthrough techniques for rigorous application to studying motor circuits. These include 3-photon in-vivo functional microscopy, Adaptive optics assisted photo manipulation, wide-field optical mapping and meso-scale two-photon microscopy. These imaging methods will be combined with innovative surgical preparations, tasks, electrode recordings and closed-loop strategies. Importantly, none of these technologies is available commercially, and thus the core will need to ensure that these systems are quickly and effectively established and supported throughout the duration of the project. Documentation of this development will be accompanied by adoption of ever evolving new technologies as they become available. The Advanced Imaging and Instrumentation Core will also play an important role in training and enabling researcher to adopt complex techniques, while working closely with the Data Science Resource core on all aspects of data standardization, archiving, analysis and dissemination. Overall, this core will represent a deep collaboration between experts in the development and deployment of novel technologies, in collaboration with those leveraging the power of these important systems.

PROJECT SUMMARY This Phase II Lab-to-Marketplace proposal aims to commercialize a new remote focusing technique that can change the focus of a microscope by as much as 500 ?m in less than 40 ?s, 3 orders of magnitude faster than other discrete focus change techniques. Our initial market is neuroscience imaging, where the ability of researchers to step between focal planes at the millisecond timescale of neuronal circuits is limited by the speed and/or complexity of current remote focusing techniques. Piezo translated objectives and liquid electrically tunable lenses have fairly long settling times, on the order of 10-20 ms, which lowers the effective duty cycle at high frame rate imaging. When these devices are operated in resonant mode, duty cycles are higher, but there are still long delays between accessing disparate axial regions. Our remote focusing device uses thin liquid crystal (LC) switches and liquid crystal polarization gratings (LCPGs) to create dynamic lenses. We originally introduced LCPGs as linear gratings for nonmechanical multiangle beamsteering, but realized they can also be leveraged for extremely high speed focusing. In Phase I, we demonstrated the first use of LC switches and circularly-patterned LCPG lenses for changing the focus of a multiphoton microscope system. We were able to shift the focus by ~300 micrometers in < 40 ?s; the settling time is independent of the device?s diameter or of the distance shifted. Axial focusing at these deeply submillisecond timescales is crucial in particular for imaging 3D neural circuits, but will also find applications in other areas where speed and/or hysteresis-free reproducibility is important. In Phase II we plan to bring the LCPG remote focusing lens stack to market with a target price of $1000 and an initial target application of optogenetics research. To reach this target price, we will undertake a systematic process development effort to increase yield, similar to techniques we have used in the LC microdisplay industry. We will also develop an in-house custom LC switch controller for greatly reduced cost and increased robustness and ease of use. With a new grating recording setup we will be able to record LCPGs with 50 mm diameter, and also address wavefront error. With our collaborators at Columbia University, we will characterize the PSF, magnification, dispersion, and wavelength-dependent signal-to-noise ratio in multiple commercial and homebuilt multiphoton microscopes, and with multiple microscope objectives. We will perform interlaminar, intralaminar, and multiplane imaging in live, behaving mice as demonstrations of the new capabilities enabled by this fast remote focusing device.

PROJECT SUMMARY To understand brain function, it is essential to identify how information is represented in neuronal population activity and how it is transformed by individual neurons as it flows through microcircuits. ?Two-photon (2P) microscopy is a core tool for this because it enables neuronal activity to be monitored at high spatial resolution deep within brain tissue in behaving animals?. ?However, ?t?he temporal resolution of conventional galvanometer-based 2P microscopy severely limits measurements of fast signaling in 3D neuronal circuits. Acousto-optic lens (AOL) microscopy, which enables fast focussing and selective imaging of regions of interest distributed within the imaging volume, has substantially improved the temporal resolution of 3D 2P microscopy. But current AOL microscopes, which rely on ?linear acoustic drive waveforms, suffer from limitations that make them ine?fficient to monitor signaling in structures that project in the Z dimension. ?Each change in the focus requires a 24 ?µ?s ?dead time? to refill the AOL aperture and continuous line scanning is restricted to the selected X-Y focal plane, limiting imaging rates for 3D dendritic trees to a few Hz, rather than the 100-1000 Hz required for monitoring neurotransmitter reporters and voltage indicators. ?The main aim of this project is to optimize and disseminate ?nonlinear ?AOL 3D microscopy, a technology we have invented to overcome these limitations by enabling ultra-fast line scanning (up to 40 kHz) in any arbitrary direction in X, Y and Z. By developing a prototype ?nonlinear AOL 2P microscope with real time correction of brain movement, we have demonstrated the performance of this technology for high-speed multiscale 3D imaging of neural circuits in awake behaving animals. We will build on these results by optimizing ?nonlinear AOL microscopy for imaging entire 3D dendritic trees and the surrounding neuronal population at unprecedented speeds. We will develop variants of this dendritic ?arboreal imaging? approach to provide low spatial resolution, ultra-high-speed 3D imaging (up to 1 kHz) by combining the fast scanning and adaptive optics properties of ?nonlinear ?AOLs. We will also extend the real time FPGA analysis used in our closed loop 3D movement correction to enable ?attentional imaging? where active regions of a dendritic tree, or circuit, are rapidly detected and imaged at higher spatio-temporal resolution. These applications ?will provide the temporal resolution required for monitoring voltage across the entire 3D dendritic tree of pyramidal cells in awake animals for the first time. Moreover, attentional imaging will enable neurotransmitter release to be mapped at high spatiotemporal resolution. Low cost dissemination of this powerful new technology will be achieved by providing US labs and an imaging facility with compact ?nonlinear AOL modules that will be added to their existing conventional 2P microscopes. By extending our open source microscope GUI software, standardizing data formats with NWB2 and refining automated analysis pipelines, we will also deliver reliable user-friendly microscope control and a semiautomated data analysis framework for the collaborators to carry out experiments on a range of different neural circuits.