Karel Svoboda - US grants

Neuronal dendrites exist in an amazing variety of shapes. They consume most of the brain's energy and account for most of the neuronal surface area. The vast majority of synapses are located on dendrites. Over the last decade it has become clear that neuronal dendrites also contain a rich variety of Na+, Ca2+ and K+ channels that can support Na+ and Ca2+ action potentials. The "back-propagating" Na+ action potential in particular has aroused considerable interest primarily because it could provide a natural mechanism for a feedback signal from the soma to distant synapses about the state of the soma. Back-propagating action potentials could therefore play an important role in facilitating Hebbian synaptic plasticity. Dendritic action potentials could in addition be involved in many physiological and pathological phenomena, including amplification of synaptic currents, learning and memory, ischemia, trauma, epilepsy, and neurodegenerative disorders. But despite many attractive proposals, the function of dendritic action potentials has remained unclear. Bridging the gap between dendritic physiology, brain function and behavior will require experimentation in vivo. For this purpose we recently developed the application of Two- Photon Laser Scanning Microscopy (TPLSM) to in vivo [Ca2+] imaging. TPLSM allows functional imaging at micrometer spatial resolution up to 600 mum deep into the brain. Using TPLSM together with intracellular somatic membrane potential measurements, working in the barrel cortex of the rat, we have previously shown that Na+ action potentials produce [Ca2+] transients in the proximal dendrites of neocortical pyramidal cells. In preliminary studies we developed techniques to measure the dendritic membrane potential in vivo, opening up the study of dendritic excitability in the intact neocortex at a level of detail that was previously reserved for experiments in cell culture and brain slices. As part of the Specific Aims of this proposal we will determine the: mechanisms of Na+ action potential back-propagation in layer 2/3 dendrites; mechanisms and function of excitability in the dendrites of layer 5 pyramidal cells; mechanisms of modulation of dendritic excitability by patterned activity, inhibition, and neuromodulatory systems. These studies will be helped by improvements to existing instrumentation for TPLSM-based physiology in vivo.

[unreadable] DESCRIPTION (provided by applicant): Over the last decade the availability of new laser light sources and modern data acquisition engines has dramatically increased the range of applications of laser scanning microscopy (LSM). Today, LSM includes some of the most important optical imaging modalities in biology, including confocal and 2-photonmicroscopy. LSM is central to research in neurobiology, cancer research, immunology, pathology, developmental biology and other fields of research. Experimental systems amenable to LSM range from single biomolecules to whole living animals. Although numerous commercial LSMs exist, the most interesting and challenging applications of LSM demand custom hardware and software. However, few investigators have attempted this approach, mainly because of the difficulty of writing the complex software required to run a laser scanning microscope. Open source software for laser scanning microscopy is not available. In our laboratory we have developed a powerful program to control a laser scanning microscope. This program, Scanlmage, performs all of the typical functions a LSM performs with important advantages, i) Scanlmage is almost entirely written in Matlab, thereby fusing image acquisition with one of the most sophisticated image analysis environments; ii) Scanlmage uses a novel approach to signal conditioning and integration, and does not require complex custom electronics, removing significant obstacles to custom design; iii) Scanlmage miswritten in a modular and object-oriented style, making it easily expandable to accommodate emerging technologies. We propose to produce on-line help and documentation for Scanlmage and make the software available open source. We further propose to develop new modules for Scanlmage to address the need to integrate LSM with new imaging technologies. These new modules will include random access laser scanning, the simultaneous operation of two scanners (e.g. one for imaging and the other for photolytic release of bioactive substances), and the combination of laser scanning microscopy and fluorescence lifetime imaging. We believe that Scanlmage will facilitate exciting biological discovery in numerous laboratories. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Understanding the connectivity of neural networks and its relationships to nervous system function and dysfunction remains a major challenge. Typical approaches to this problem include lesions or focal pharmacology; these methods are crude and non-specific. Recently, the mouse genome project and largescale expression studies, together with emerging transgenic technologies, have begun to allow selection of specific neuronal populations for intervention. But what should one target to intervene in neuronal function? Ideally a mechanism for intervention should be conditional, i.e. only interfere with function when prompted by the experimenter. Such induction should be rapid and rapidly reversible. Intervention should also be delicate, not requiring, for example, major surgical intervention. Finally, the intervention should be specific, in the sense that it interferes with a well-defined aspect of neuronal function without a myriad of secondary effects. We will design molecular genetic tools that allow conditional perturbation of synaptic transmission. Triggered by administration of pharmacological agents without endogenous targets, synaptic vesicles and/or proteins important for exocytosis will be mislocalized or immobilized, interfering with synaptic transmission. This will be accomplished in three steps. 1.) We will engineer modified versions of proteins involved in exocytosis by introducing domains (FK506 binding protein and target of rapamycin binding domain) that, while by themselves do not interfere with protein function, will confer on the fusions the ability to interact with small chemical crosslinkers (derivatives of FK506 and rapamycin). Addition of such cell-permeable dimerizers will induce protein-protein interactions that will sequester essential synaptic vesicle components away from their natural partners and/or the site of function, resulting in loss of synaptic transmission. 2.) Systems for inactivating synaptic transmission will be tested in cultured neurons and brain slices using optical techniques to monitor vesicle cycling and electrophysiology to measure synaptic currents. 3.) Well-characterized systems will be introduced into subpopulations of neurons in mice using viruses and transgenic approaches. Systems will be tested in vivo using intrinsic signal imaging and electrophysiological techniques.