Lawrence B. Cohen - US grants

All of the objectives of this project involve the use of optical methods for monitoring neuron activity. These methods are based on the finding that certain dyes, when bound to membranes, change their optical properties in response to changes in membrane potential. In invertebrate ganglia optical monitoring detects the individual action-potentials in individual cell bodies. Using a 124 element photodiode array positioned in a microscope image plane we have been able to monitor the activity of many neurons of Aplysia or Pleurobranchaea buccal ganglia. We think we are approaching the capability of monitoring most of the spike activity in such preparations. This would provide a powerful tool for determining all of the neurons active during particular behaviors and for testing hypothese about neuronal control of simple behaviours. We are particularly interested in the nuronal basis of simple forms of learning. Additional behavioral experiments and optical measurements in semi-intact preparations are proposed for the immediate future. These would be followed by attempts to determine how the interactions between neurons changed to accomplish the behavior. In vertebrate nervous systems optical monitoring does not yet have single cell resolution. The signals represent average changes in membrane potential of the neurons and processes imaged onto the individual detectors. At present it is possible to measure changes in activity in rat sensory cortex in response to whisker movement or light flashes. We plan experiments to improve the signal-to-noise ratio in these experiments, to develop the use of optical measurements in cat visual cortex, and to determine if there is odor dependent localization of activity in the salamander olfactory bulb. Neurosurgeons could more accurately locate primary epileptogenic areas of cortex if they could identify epileptic foci in the inter-siezure interval. Since neurons in the epileptic foci are said to be somewhat depolarized during the intersiezure interval, voltage sensitive dyes might be used to locate these cells. We plan experiments on model systems to test the feasibility of two possible approaches toward localization.

There are two long-term objectives of the proposal. First, to understand in neuronal detail how simpler nervous systems are able to generate behaviors and how these nervous systems are able to accomplish simple forms of learning. The second is to understand what happens in cortex during epileptic seizures. In order to better accomplish these scientific goals, we also have the methodological goal of improving optical techniques for monitoring neuron activity. We have been using optical methods to monitor neuron activity in the Aplysia abdominal ganglion during the gill-withdrawal reflex. Our preliminary results suggest that between 250 and 420 neurons in the ganglion are activated by a mechanical stimulus to the siphon. However this estimate comes from a recording that is perhaps 35% complete. We propose experiments to improve the completeness of the recording, experiments to allow preliminary identification of many of these neurons, and analyses to allow us to follow the activity of individual neurons during a series of trials involving habituation and sensitization. We also plan experiments to try to see if the same behavioral plasticity can be achieved with fewer neurons. If this is successful we propose to make a model of the neuron interactions that generate the behavior. We have been able to monitor activity in bicuculline induced inter-ictal epileptiform discharges in rat somatosensory cortex. Our results were surprising in two regards. First, they indicated that the optical measurements and the ball electrode measurements could give very disparate results. Second, they indicated that there were qualitative differences between discharges induced by sensory stimulation and spontaneous discharges. Our first aim is to confirm these result. Then we plan a local application of bicuculline to study the spread (or absence of spread) of the epileptic activity away from the treated region. Finally we plan a simultaneous measurement from both the bicuculline treated and the untreated contralateral somatosensory cortex. These experiments are meant to lead toward measurements in more sophisticated models (primates) of the human disease. These experiment may also lead to the use of optical monitoring techniques for determining the location of epileptic foci in surgical patients.

The long term goal of the proposal is to explain the neuronal bases of simple forms of behavior at the level of signal processing in all of the elements of the functional neuronal network. Toward this goal, we would first try to achieve simultaneous recording of the action potentials and subthreshold signals in both cell bodies and processes in a restricted number of neurons using preparations in which simultaneous recording of action potentials from most neuronal cell bodies (but not processes)is already possible. This study would be based on experiments using multi-site optical recording techniques. Optical recording of subthreshold potential changes from small neuronal processes is still exceedingly difficult because of the small signal size relative to noise. At present it can only be used under most favorable and thus very limited experimental conditions. The improvement in sensitivity that we hope to obtain will be based on two approaches. First we would exploit the idea of limiting the application of impermeant voltage-sensitive dyes to a restricted number of neurons in the network and not stain the whole structure. This may be accomplished by intracellular application of dyes into cell bodies. Intracellular staining would improve spatial resolution of optical recording as well as signal-to-noise ratio. Second, we propose to continue the search for new dyes that would give better optical signals after either extra- or intracellular application. We propose to test analogs of most successful dyes from different categories of voltage sensitive compounds and to use several different invertebrate preparations that are especially suitable for optical recording studies of neuronal correlates of behavior. With the best dye/preparation combination we then propose to develop the recording and analysis of signal integration at the level of neuronal processes of individual cells within a functional network. The knowledge we obtain about how normal behavior is generated in these primitive animals will help us understand mental disorders such as depression, drug addiction or schizophrenia.

The Children's Clinical Research Center (CCRC) consists of five inpatient beds, outpatient facilities of two treatment rooms and reception area, office space, skilled research nurses and other personnel to support these resources. The Children's Clinical Research Center, which has been operating since 1963, permits the faculty of the Departments of Pediatrics, Child Study Center, Psychiatry, Surgery, Human Genetics and Internal Medicine, their postdoctoral fellows and medical students, to conduct clinical investigation in children. The Center provides the environment for studies of normal and abnormal body function, and the cause, progression, prevention, control and cure of childhood disease, as well as the psychosocial implications of organic disease in this age group. Major areas of investigation include developmental changes in insulin secretion and insulin action in childhood health and disease, regulation of substrate supply and brain metabolism in children, alterations in cardiorespiratory function in neonates and children, natural history and responses to experimental drug therapy in pediatric AIDS, factors that influence language development and cognitive function in healthy children and children with dyslexia and attention deficit disorders, the pheonomenology and neurobiology of Tourette's syndrome and other neuropsychiatric disorders, state-of-the-art advances in the treatment of the critically ill neonate. Our center also provides the clinical research infrastructure for investigators participating in NIH supported clinical trials in diabetes, oncology, neurological and neonatal disorders. The CCRC is an ideal setting in which advances in scientific knowledge can be translated into new or improved methods of health care for children.

DESCRIPTION (provided by applicant): The pattern of activity in the circuits of the brain and their experience-dependent changes underlie the processing of sensory information, perception, and motor control. Much has been learned about the anatomical wiring of brain circuits and about the properties of individual neurons in the intact brain, but considerable mystery remains about how the properties of individual neurons emerge from their connectivity and how multiple groups of neurons are activated during behaviors. Part of the problem has been that high precision electrical recording is usually obtained from only one or a few neurons at a time, when salient events are actually processed by large assemblies of neurons. To provide for a fast high-resolution recording from mammalian neurons, this proposal seeks to improve fluorescent protein (FP) based voltage sensors. These probes will be self-contained, not requiring any exogenous factors to function, and thus will be genetically-encodable. We are seeking probes that are readily expressed on the cell's surface, show maximum changes in intensity with membrane potential alterations, respond rapidly to changes in membrane potential, and are minimally disruptive to cells. Members of the project have been involved in the development of first generation FP-voltage sensors, including Fluorescent Shaker (FlaSh), Voltage-Sensitive Fluorescent Protein (VSFP) and Sodium channel Protein Activity Reporting Construct (SPARC). These constructs have demonstrated the feasibility of creating channel-FP constructs that alter fluorescence intensity with changes in cell membrane potential. Significant improvements in the response characteristics may come from a pseudo-saturating examination of the ion channel/transporter and fluorescent protein space. This proposal will create large libraries of membrane protein / FP fusion constructs varying the membrane protein, the location of the inserted FP and the isoform of the inserted FP. These libraries will be created by a novel transposon-based FP insertion process. Constructs will be screened for surface expression in hippocampal neurons and tested for voltage-dependent fluorescence changes using fast fluorescence measurements combined with voltage-clamp electrophysiology. We will express the most promising FP-voltage sensors in brain slices and in vivo using viral infection followed by the production of transgenic mice.

This proposal aims to develop better tools for analyzing brain cells and circuits and for large-scale recordings of brain activity. The currently available tools are relatively primitive in terms of sensitivity and speed. One major function of a neuron is to process electrical signals. Thus a tool that is of particular significance is high speed membrane potential imaging. Genetically encoded fluorescent protein voltage indicators (GEVI's) are a obvious strategic approach for ?visualizing the brain in action?. Genetically encoded sensors are especially interesting to neuroscientists because, as proteins, they can be expressed in individual cell types in the mammalian brain. Because each brain region has up to 100 different cell types, sensor expression in a specific cell type is essential for imaging the activity of that cell type. Recently, there has been a dramatic improvement in the signal size of GEVIs (i.e. ArcLight, 40%/100 mv) but ArcLight has a relatively slow response time constant (?=10 msec). There are now several faster GEVIs (?=0.3 to 2.0 msec) but they have smaller signal sizes (~10%/100mv). One goal of this proposal is developing a GEVI with both large and fast responses to membrane potential changes. Several probe characteristics other than size and speed are also critically important. One is the wavelength range of excitation and emission. At present many sensors and activators are based on GFP and its analogues. Thus, probes with red excitation and emission spectra would allow simultaneous dual function measurements. One aim is to develop useful red GEVIs. Second, many GEVIs have a sigmoidal fluorescence-voltage relationship. The position of this relationship along the voltage axis can be adjusted via mutations in the voltage sensitive domain of the GEVI. Thus GEVIs can be selective reporters of different ranges of the neuron membrane potential and thereby selective for action potential activity versus subthreshold activity. This selectivity depends on both the voltage at half-maximal activation as well as the steepness of the sigmoidal curve. Lastly, it will be important to target the GEVIs to specific regions of the neuron including cell body, dendritic post-synaptic zones, and presynaptic terminals. All of the probes in the proposal are based on the voltage sensitive domain of a membrane protein (a phosphatase from Ciona or zebrafish) with one or two fluorescent proteins inserted into the N- or C-terminal region. The development of improved GEVIs will involve molecular biology for generating novel probes (bigger, faster, red, targeted anatomically and physiologically) followed by testing in cultured HEK293 cells, acutely dissociated neurons, and zebrafish embryos. Probes that function well in these initial screens will then be incorporated into virus particles for in vivo transfection and measurements in the mammalian brain. Improved and selective GEVIs would be useful to the community carrying out optical measurements of brain activity.