|
|
Laboratory of Synapse Structure and Function | |||||
| About | Anatomy | Links | Publications | Search | Tools | |
|
About the Work in Our Laboratory...Knowing the dimensions and connectivity of synapses is fundamental to understanding function. In the brain, more than 90% of synapses occur on dendritic spines. These tiny protrusions from the surface of dendrites measure about 1 micrometer in length. Dendritic spine structure is clearly important for normal brain function because when brain function is impaired, such as in conditions of mental retardation, epilepsy, and stroke, the dendritic spines are either gone, or have highly distorted shapes and sizes.
Dendritic spines and their synaptic components are too small to measure accurately with light microscopy so we have developed and standardized computer-assisted approaches to analyze them in three-dimensions through serial section electron microscopy. We have established the rat hippocampal slice as a model system for structural synaptic plasticity by developing a microwave-enhanced procedure that produces rapid fixation to the center of the slice within a minute after the last physiological recording. We also collaborate with Dr. Sergei Kirov to use two-photon and confocal microscopy to visualize global effects of different levels of synaptic activity on spine density along dendrites, prior to ultrastructural analyses. Recently, we have received a Javits Merit award to extend our studies regarding the functional ultrastructure of LTP into the hippocampus of the awake rat in collaboration with Dr. Wickcliffe Abraham at the University of Otago, Dunedin, New Zealand. Our research has revealed contrasting effects of synapse activation on spine structure and formation in the immature and mature rat hippocampus. Prior to postnatal day (PN) 11, hippocampal CA1 synapses are located on the dendritic shafts or along dendritic filopodia, often piled up around the base of a filopodium. By PN15, shorter dendritic spines have emerged with enlarged heads each of which hosts one synapse, filopodia are rare, and shaft synapses have diminished almost to their low (<5%) adult level. In parallel, we have shown that LTP develops during PN 11 to 15 after dendritic spines have begun to emerge. Recent studies from our lab show that if LTP is induced at PN15, polyribosomes shift from dendritic shafts into dendritic spines and those spines containing the polyribosomes have enlarged synapses by 2 hours after induction, relative to synapses at control sites in the same slices. Our recent findings show that there is a robust increase in the number of dendritic spines with synapses, nonsynaptic filopodia, and polyribosomes throughout the dendrites by 30 minutes after induction of LTP. Spine number returns to control levels by 2 hours suggesting a rapid competition and elimination for synaptic sites during 1-2 hours after induction of LTP. These findings provide strong evidence for local changes in protein synthesis at a subpopulation of synapses on developing hippocampal neurons during LTP. Like PN15 mature hippocampus shows no net change in synapse number or size at 2 hours after induction of LTP. The new results from PN15 have motivated our re-examination of polyribosomes in mature hippocampus to determine whether they are in a subpopulation of enlarged spines during LTP. With Dr. Abraham we will determine how soon, and for how long after induction of LTP the synapses remain altered providing a stable mechanism of enduring LTP. We have measured dendritic spines and their presynaptic partners in several mature brain regions. We established that interneurons do not synapse with hippocampal dendritic spines, thereby removing differences in presynaptic input types as a potential source of the large (>10 fold) variation in spine dimensions along even a short 5-micron segment of dendrite. Our findings show that larger spines have proportionately larger and more complex subcellular organelles, and postsynaptic receptive surfaces, and more presynaptic vesicles. We have shown that spine necks are constricted just enough to allow the heads to be relatively isolated biochemical compartments near the synapses, without choking off transmission of electrotonic signals to the postsynaptic dendrite. Furthermore, our recent studies show that different subsets of spines contain smooth endoplasmic reticulum, or endosomal compartments suggesting the independence of these organelles in modulating synaptic efficacy. These findings demonstrate the power of serial EM to capture and illustrate dynamic processes on the ultrastructural scale by comparing across ages, conditions, times, dendrites, and brain regions. Studies using confocal and two-photon microscopy in collaboration with Dr. Kirov show that blocking synaptic transmission on mature hippocampal neurons results in a prodigious up regulation in the number of dendritic spines. These findings suggest that a homeostatic mechanism was triggered to maintain a constant level of synaptic input on mature neurons. In contrast, if synaptic transmission is blocked at ages younger than postnatal day 21, there is no net effect on spine number. It appears that the manifestation of this homeostatic mechanism requires a substantial period of postnatal development. Recent three-dimensional studies at the ultrastructural level demonstrate a recapitulation of development in the formation of new synapses during conditions of blocked synaptic transmission in the adult. Astrocytic processes are heavily endowed with glutamate transporters; therefore, measuring the degree to which they surround and intervene between synapses is integral to understanding their role in synaptic transmission and plasticity. We find that about 50% of hippocampal synapses have astrocytic processes surrounding a portion of their perimeters. This structural arrangement differs from cerebellum, for example, wherein nearly 100% of the synapses have a complete glial ensheathment. In the absence of glia, neurons grown in culture are 12 fold less active than neurons grown with glia. Efforts are underway to determine whether the glial processes grow towards or away from hippocampal synapses under the influence of LTP, LTD, and/or homeostatic mechanisms regulating total synaptic transmission. Areas in which I plan to expand our research include models of learning and memory and mental retardation in vivo, the role of sleep and circadian rhythms in modulating synapse number and structure, and mapping the location of key molecules in three dimensions. In addition, I am working to understand the role of protein synthesis in structural synaptic plasticity. In earlier work, I showed that LTP has a circadian cycle, being more likely to occur at a higher magnitude during the rat’s sleep cycle for hippocampal area CA1 and during the rat’s active wake cycle in hippocampal area dentata. One goal will be to determine whether circadian and sleep rhythms in the mature nervous system provide quiet periods when new dendritic spines and synapses are formed which can be subsequently preserved or eliminated during learning and memory. I also plan new efforts towards producing accurate three-dimensional maps of relevant molecules at the synapse, along dendrites, presynaptically, and in relationship to astroglial processes. In addition, I plan to expand our database of synapses (see http://synapse-web.org).
Selected publications: Harris KM, Teyler TJ. (1984) Developmental onset of long-term potentiation in area CA1 of the rat hippocampus. J. Physiol., 346:27-48 (352K PDF). Jackson PS, Suppes T, Harris KM (1993) Stereotypical changes in the pattern and duration of long-term potentiation at postnatal days 11 and 15 in the rat hippocampus. J.Neurophysiol. 70:1412-1419 (3,493K PDF). Fiala JC, Allwardt B, and Harris KM (2002) Dendritic spines do not split during hippocampal LTP or maturation. Nat Neurosci. 5(4): 297-8(311K PDF). Ostroff LE, Fiala JC, Allwardt B, Harris KM (2002) Polyribosomes redistribute from dendritic shafts into spines with enlarged synapses during LTP in developing rat hippocampal slices. Neuron. 35(3):535-545(552K PDF). Harris KM, Fiala JC, Ostroff L. (2003) Structural changes at dendritic spine synapses during long-term potentiation. Philos Trans R Soc Lond B Biol Sci. 358:745-8(1.56MB PDF). Spacek, J and Harris, KM (2004) Trans-endocytosis via Spinules in Adult Rat Hippocampus. J Neuroscience 24(17):4233-41; and featured in “This week in the journal.” Email requests for this pub to kharris@mail.clm.utexas.edu. Sorra, KE, Mishra, A, Kirov SA and Harris, KM (2006) Dense core vesicles resemble active-zone transport vesicles and are diminished following synaptogenesis in mature hippocampal slices. Neuroscience, In Press. Witcher MR, Kirov,SA and Harris, KM (2006) Plasticity of perisynaptic astroglia during synaptogenesis in the mature rat hippocampus. Glia, In Press. |
Our goal is to elucidate structural components involved in the cell biology of learning and memory. We study long-term potentiation (LTP) and its complement, long-term depression (LTD), in the developing and mature hippocampus because these phenomena have many of the physiological characteristics that are expected for learning and memory in the brain. Our working hypothesis is synaptic plasticity that serves to modify synapses in the creation of new memories competes with homeostatic mechanisms that serve to prevent saturation of synaptic strength and neuropathology. Our focus has been on dendritic spines because they are the major postsynaptic targets of excitatory axons throughout the brain and because their structure and composition serve both synaptic plasticity and stabilizing homeostatic mechanisms.
