The Picower Institute for Medical Research

New camera chip design allows for optimizing each pixel's timing to maximize signal to noise ratio when tracking real-time visual indicator of neural voltage. Neurons communicate electrically so to understand how they produce brain functions such as memory, neuroscientists must track how their voltage changes -- sometimes subtly -- on the timescale of milliseconds. In a new paper in Nature Communications, MIT researchers describe a novel image sensor with the capability to substantially increase that ability. The invention led by Jie Zhang, a postdoctoral scholar in The Picower Institute for Learning and Memory lab of Sherman Fairchild Professor Matt Wilson, is a new take on the standard "CMOS" technology used in scientific imaging. In that standard approach, all pixels turn on and off at the same time -- a configuration with an inherent trade-off in which fast sampling means capturing less light. The new chip enables each pixel's timing to be controlled individually. That arrangement provides a "best of both worlds" in which neighboring pixels can essentially complement each other to capture all the available light without sacrificing speed. In experiments described in the study, Zhang and Wilson's team demonstrates how "pixelwise" programmability enabled them to improve visualization of neural voltage "spikes," which are the signals neurons use to communicate with each other, and even the more subtle, momentary fluctuations in their voltage that constantly occur between those spiking events. "Measuring with single-spike resolution is really important as part of our research approach," said senior author Wilson, a Professor in MIT's Departments of Biology and Brain and Cognitive Sciences (BCS), whose lab studies how the brain encodes and refines spatial memories both during wakeful exploration and during sleep. "Thinking about the encoding processes within the brain, single spikes and the timing of those spikes is important in understanding how the brain processes information." For decades Wilson has helped to drive innovations in the use of electrodes to tap into neural electrical signals in real-time, but like many researchers he has also sought visual readouts of electrical activity because they can highlight large areas of tissue and still show which exact neurons are electrically active at any given moment. Being able to identify which neurons are active can enable researchers to learn which types of neurons are participating in memory processes, providing important clues about how brain circuits work. In recent years, neuroscientists including co-senior author Ed Boyden, Y. Eva Tan Professor of Neurotechnology in BCS and The McGovern Institute for Brain Research and a Picower Institute affiliate, have worked to meet that need by inventing "genetically encoded voltage indicators" (GEVIs), that make cells glow as their voltage changes in real-time. But as Zhang and Wilson have tried to employ GEVIs in their research, they've found that conventional CMOS image sensors were missing a lot of the action. If they operated too fast, they wouldn't gather enough light. If they operated too slow, they'd miss rapid changes. But image sensors have such fine resolution that many pixels are really looking at essentially the same place on the scale of a whole neuron, Wilson said. Recognizing that there was resolution to spare, Zhang applied his expertise in sensor design to invent an image sensor chip that would enable neighboring pixels to each have their own timing. Faster ones could capture rapid changes. Slower-working ones could gather more light. No action or photons would be missed. Zhang also cleverly engineered the required control electronics so they barely cut into the space available for light-sensitive elements on a pixels. This ensured the sensor's high sensitivity under low light conditions, Zhang said. Two demos In the study the researchers demonstrated two ways in which the chip improved imaging of voltage activity of mouse hippocampus neurons cultured in a dish. They ran their sensor head-to-head against an industry standard scientific CMOS image sensor chip. In the first set of experiments the team sought to image the fast dynamics of neural voltage. On the conventional CMOS chip, each pixel had a zippy 1.25 millisecond exposure time. On the pixel-wise sensor each pixel in neighboring groups of four stayed on for 5 milliseconds, but their start times were staggered so that each one turned on and off 1.25 seconds later than the next. In the study, the team shows that each pixel, because it was on longer, gathered more light but because each one was capturing a new view every 1.25 milliseconds, it was equivalent to simply having a fast temporal resolution. The result was a doubling of the signal-to-noise ratio for the pixelwise chip. This achieves high temporal resolution at a fraction of the sampling rate compared to conventional CMOS chips, Zhang said. Moreover, the pixelwise chip detected neural spiking activities that the conventional sensor missed. And when the researchers compared the performance of each kind of sensor against the electrical readings made with a traditional patch clamp electrode, they found that the staggered pixelwise measurements better matched that of the patch clamp. In the second set of experiments, the team sought to demonstrate that the pixelwise chip could capture both the fast dynamics and also the slower, more subtle "subthreshold" voltage variances neurons exhibit. To do so they varied the exposure durations of neighboring pixels in the pixelwise chip, ranging from 15.4 milliseconds down to just 1.9 milliseconds. In this way, fast pixels sampled every quick change (albeit faintly), while slower pixels integrated enough light over time to track even subtle slower fluctuations. By integrating the data from each pixel, the chip was indeed able to capture both fast spiking and slower subthreshold changes, the researchers reported. The experiments with small clusters of neurons in a dish was only a proof-of concept, Wilson said. His lab's ultimate goal is to conduct brain-wide, real-time measurements of activity in distinct types of neurons in animals even as they are freely moving about and learning how to navigate mazes. The development of GEVIs and of image sensors like the pixelwise chip that can successfully take advantage of what they show is crucial to making that goal feasible. "That's the idea of everything we want to put together: large-scale voltage imaging of genetically tagged neurons in freely behaving animals," Wilson said. To achieve this, Zhang added, "We are already working on the next iteration of chips with lower noise, higher pixel counts, time-resolution of multiple kHz, and small form factors for imaging in freely behaving animals." The research is advancing pixel by pixel. In addition to Zhang, Wilson and Boyden the paper's other authors are Jonathan Newman, Zeguan Wang, Yong Qian, Pedro Feliciano-Ramos, Wei Guo, Takato Honda, Zhe Sage Chen, Changyang Linghu, Ralph-Etienne Cummings, and Eric Fossum. The Picower Institute for Learning and Memory, The JPB Foundation, the Alana Foundation, The Louis B. Thalheimer Fund for Translational Research, the National Institutes of Health, HHMI, Lisa Yang and John Doerr provided support for the research.

Single-cell gene expression patterns in the brain motor and frontal cortex, and evidence from follow-up experiments, reveal many shared cellular and molecular similarities that could be targeted for potential treatment. On the surface, the movement disorder amyotrophic lateral sclerosis (ALS), also known as Lou Gehrig's disease, and the cognitive disorder frontotemporal lobar degeneration (FTLD), which underlies frontotemporal dementia, manifest in very different ways. In addition, they are known to primarily affect very different regions of the brain. However, doctors and scientists have noted several similarities over the years, and a new study in Cell reveals that the diseases have remarkable overlaps at the cellular and molecular levels, revealing potential targets that could yield therapies applicable to both disorders. The new paper, led by scientists at MIT and the Mayo Clinic, tracked RNA expression patterns in 620,000 cells spanning 44 different cell types across motor cortex and prefrontal cortex from postmortem brain samples of 73 donors diagnosed with ALS, FTLD, or who were neurologically unaffected. "We focused on two brain regions that we expected would be differentially affected between the two disorders," said Manolis Kellis, co-senior author of the paper and professor in the Computer Science and Artificial Intelligence Laboratory at MIT. "It turns out that at the molecular and cellular level, the changes we found were nearly identical in the two disorders, and affected nearly identical subsets of cell types between the two regions." Indeed, one of the most prominent findings of the study revealed that in both diseases the most vulnerable neurons were almost identical both in the genes that they express, and in how these genes changed in expression in each disease. "These similarities were quite striking, suggesting that therapeutics for ALS may also apply to FTLD and vice versa," said lead corresponding author Myriam Heiman, associate professor in The Picower Institute for Learning and Memory and the Department of Brain and Cognitive Sciences at MIT. "Our study can help guide therapeutic programs that would likely be effective for both diseases." Heiman and Kellis collaborated with co-senior author Veronique Belzil, then associate professor of neuroscience at the Mayo Clinic Florida, now director of the ALS Research Center at Vanderbilt University. Another key realization from the study is that brain donors with inherited vs. sporadic forms of the disease showed similarly altered gene expression changes, even though these were previously thought to have different causes. That suggests that similar molecular processes could be going awry downstream of the diseases' origins. "The molecular similarity between the familial (monogenic) form and the sporadic (polygenic) forms of these disorders suggests that convergence of diverse etiologies into common pathways," Kellis said. "This has important implications for both understanding patient heterogeneity and understanding complex and rare disorders more broadly." 'Practically indistinguishable' profiles This overlap was especially evident, the study found, when looking at the most affected cells. In ALS, known to cause progressive paralysis and ultimately death, the most endangered cells in the brain are upper motor neurons (UMN) in layer 5 of the motor cortex. Meanwhile in behavioral variant frontotemporal dementia (bvFTD), the most common type of FTLD that is characterized instead by changes to personality and behavior, the most vulnerable neurons are spindle neurons, or von Economo cells, found in layer 5 of more frontal brain regions. The new study shows that while the cells look different under the microscope, and make distinct connections in brain circuits, their gene expression in health and disease is nevertheless strikingly similar. "UMNs and spindle neurons look nothing alike and live in very different areas of the brain" said Sebastian Pineda, lead author of the study, and a graduate student jointly supervised by Heiman and Kellis. "It was remarkable to see that they appear practically indistinguishable at the molecular level and respond very similarly to disease." The researchers found many of the genes involved in the two diseases implicated primary cilia, tiny antenna-like structures on the cell's surface that sense chemical changes in the cell's surrounding environment. Cilia are necessary for guiding the growth of axons, or long nerve fibers that neurons extend to connect with other neurons. Cells that are more dependent on this process, typically those with the longest projections, were found to be more vulnerable in each disease. The analysis also found another type of neuron, which highly expresses the gene SCN4B and which was not previously associated with either disease, also shared many of these same characteristics and showed similar disruptions. "It may be that changes to this poorly characterized cell population underlie various clinically-relevant disease phenomena," Heiman said. The study also found that the most vulnerable cells expressed genes known to be genetically-associated with each disease, providing a potential mechanistic basis for some of these genetic associations. This pattern is not always the case in neurodegenerative conditions, Heiman said. For example, Huntington's disease is caused by a well-known mutation in the huntingtin gene, but the most highly affected neurons don't express huntingtin more than other cells, and the same is true for some genes associated with Alzheimer's disease. Looking beyond neurons, the study characterized gene expression differences in many other brain cell types. Notably, researchers saw several signs of trouble in the brain's circulatory system. The blood-brain barrier (BBB), a filtering system that tightly regulates which molecules can go into or come out of the brain through blood vessels, is believed to be compromised in both disorders. Building on their previous characterization of human brain vasculature and its changes in Huntington's and Alzheimer's disease by Heiman, Kellis, and collaborators including Picower Institute Director Li-Huei Tsai, the researchers found that proteins needed to maintain blood vessel integrity are reduced or misplaced in neurodegeneration. They also found a reduction of HLA-E, a molecule thought to inhibit BBB degradation by the immune system. Given the many molecular and mechanistic similarities in ALS and FTLD, Heiman and Kellis said they are curious why some patients present with ALS and others with FTLD, and others with both but in different orders. While the present study examined "upper" motor neurons in the brain, Heiman and Kellis are now seeking to also characterize connected "lower" motor neurons in the spinal cord, also in collaboration with Belzil. "Our single-cell analyses have revealed many shared biological pathways across ALS, FTLD, Huntington's, Alzheimer's, vascular dementia, Lewy body dementia, and several other rare neurodegenerative disorders," says Kellis. "These common hallmarks can pave the path for a new modular approach for precision and personalized therapeutic development, which can bring much-needed new insights and hope." In addition to Pineda, Belzil, Kellis and Heiman, the study's other authors are Hyeseung Lee, Maria Ulloa-Navas, Raleigh Linville, Francisco Garcia, Kyriaktisa Galani, Erica Engelberg-Cook, Monica Castanedes, Brent Fitzwalter, Luc Pregent, Mahammad Gardashli, Michael DeTure, Diana Vera-Garcia, Andre Hucke, Bjorn Oskarsson, Melissa Murray and Dennis Dickson. Support for the study came from the National Institutes of Health, Mitsubishi Tanabe Pharma Holdings, The JPB Foundation, The Picower Institute for Learning and Memory, the Robert Packard Center for ALS Research at Johns Hopkins, The LiveLikeLou Foundation, the Gerstner Family Foundation, The Mayo Clinic Center for Individualized Medicine, and the Cure Alzheimer's Fund.

A potential new Alzheimer's drug represses the harmful inflammatory response of the brain's immune cells, reducing disease pathology, preserving neurons and improving cognition in preclinical tests. Though drug developers have achieved some progress in treating Alzheimer's disease with medicines that reduce amyloid-beta protein, other problems of the disease including inflammation, continue unchecked. In a new study, scientists at The Picower Institute for Learning and Memory at MIT describe a candidate drug that in human cell cultures and Alzheimer's mouse models reduced inflammation and improved memory. The target of the new "A11" molecule is a genetic transcription factor called PU.1. Prior research has shown that amid Alzheimer's disease, PU.1 becomes an overzealous director of inflammatory gene expression in the brain's microglia immune cells. A11 suppresses this problematic PU.1 activity, the new research shows, by recruiting other proteins that repress the inflammatory genes PU.1 works to express. But because A11 concentrates mostly in the brain and does not reduce PU.1 levels, it does not appear to disrupt PU.1's other job, which is to ensure the production of a wide variety of blood cells. "Inflammation is a major component of Alzheimer's disease pathology that has been especially hard to treat," said study senior author Li-Huei Tsai, Picower Professor of Neuroscience at MIT and director of The Picower Institute and MIT's Aging Brain Initiative. "This preclinical study demonstrates that A11 reduces inflammation in human microglia-like cells as well as in multiple mouse models of Alzheimer's disease and significantly improves cognition in the mice. We believe A11 therefore merits further development and testing." Tsai and Elizabeta Gjoneska of the National Institutes of Health are co-corresponding authors of the study published in the Journal of Experimental Medicine. As a postdoc, Gjoneska co-led a 2015 study that implicated PU.1 as a regulator of errant microglia inflammation in a mouse model of Alzheimer's disease. That research was a collaboration between Tsai's lab and that of MIT Computer Science Professor Manolis Kellis, co-led by former postdoc Andreas Pfenning, now a faculty member at Carnegie Mellon University. Ever since then, Tsai has been seeking a safe way to restore PU.1 activity to healthier levels. The work described in the new paper, led by Picower Institute research scientist William Ralvenius, starts with experiments to further validate that PU.1 would be a therapeutically meaningful target. To do that the scientists compared gene expression in immune cells of postmortem brain samples from Alzheimer's patients and mouse models and matching non-Alzheimer's controls. The comparisons showed that Alzheimer's effects major changes in microglial gene expression and that an increase in PU.1 binding to inflammatory gene targets was a significant component of that change. Moreover, they showed that reducing PU.1 activity in a mouse model of Alzheimer's reduced inflammation and neurodegeneration, the death of neurons. Screening success Genetically knocking down PU.1 in the body is not a viable therapeutic strategy given its importance in normal healthy function. The team therefore screened more than 58,000 small molecules from libraries of FDA-approved drugs and novel chemicals to see if any could safely and significantly reduce key inflammation and Alzheimer's related genes regulated by by PU.1 in cell cultures. After several rounds of increasingly stringent screening, they narrowed the field down to six chemicals. A11 was by far the most potent among them. They tested the effects of A11 doses on the function of human microglia-like cells cultured from patient stem cells. When they exposed the microglia-like cells to immune molecules that typically trigger inflammation, cells dosed with A11 exhibited reduced expression and secretion of inflammatory cytokines and less of the cell body shape changes associated with microglia inflammatory responses. The cells also showed less accumulation of lipid molecules, another sign of inflammatory activation. Looking at gene expression patterns, the scientists observed that A11-treated cells exposed to inflammatory triggers behaved much like unperturbed microglia, suggesting that A11 helps prevent microglia from overreacting to inflammatory cues. Two more lab tests aimed at understanding how A11 exerts its effects revealed that it doesn't change PU.1 levels. Instead it counteracts PU.1 activity by recruiting several proteins including MECP2, HDAC1, SIN3A and DMNT3A, known to repress expression of its targets. Essentially amid Alzheimer's disease, A11 tamps down what PU.1 amps up. "A11 represents a first-in-class molecule that converts PU.1 from a transcriptional activator to a transcriptional repressor, resulting in a controlled state of microglial inflammation," the authors wrote. Mice in mazes Having established that A11 reduced inflammatory activity in microglia and determined how that happens, the team focused on whether it worked as a medicine in mouse models of Alzheimer's disease. Pharmacological tests indicated that A11 is readily cleared from tissues and is capable of reaching brain cells. Moreover, in healthy mice the chemical successfully crossed the blood-brain barrier and remained in brain cells much longer than anywhere else. Finally the team tested the effects of the drugs on Alzheimer's disease pathology and symptoms in three mouse strains that each model different aspects of Alzheimer's disease: CK-p25 mice (severe neurodegeneration), Tau P301S transgenic mice (tauopathy), and 5XFAD mice (amyloid pathology). Male and female CK-p25 mice dosed with A11 showed less inflammatory response among microglia and astrocyte cells and lost fewer neurons than untreated controls. TauP301S Tg mice responded similarly, also exhibiting a significant reduction of phosphorylated tau protein in the hippocampus region of the brain, which is an essential area for memory. In 5XFAD mice, amyloid was significantly reduced. The team subjected the Tau P301S Tg and CK-p25 mice to mazes designed to test their short-term working memory and longer-term learning. In both models and on both tests, A11-treated mice performed significantly better than untreated controls. For example, in the "Morris Water Maze," where mice have to learn the location of a submerged platform that allows them to rest, treated CK-p25 mice learned much faster than untreated ones. Much more testing needs to be done before A11 could become an approved medicine, Tsai said, but she noted that it could complement the new treatments that target amyloid. "Given that A11 acts via a distinct mechanism from existing AD therapeutics, A11 could be used alone or in combination with approved therapeutics to provide improved treatment options for neurodegenerative diseases," the authors concluded. In addition to Tsai, Gjoneska and Ralvenius, the paper's other authors are Alison E. Mungenast, Hannah Woolf, Margaret M. Huston, Tyler Z. Gillingham, Stephen K. Godin, Jay Penney, Hugh P. Cam, Fan Gao, Celia G. Fernandez, Barbara Czako, Yaima Lightfoot, William J. Ray, Adrian Beckmann, Alison M. Goate, Edoardo Marcora, Carmen Romero-Molina, Pinar Ayata, and Anne Schaefer. The Robert A. and Renee E. Belfer Family Foundation and the National Institutes of Health funded the research. Additional support came from The JPB Foundation and The Picower Institute for Learning and Memory, The Halis Family Foundation, Lester A. Gimpelson and Jay L. and Caroll Miller.