The M.R. Bauer Distinguished Lecturer
Kristen M. Harris, Ph.D.
Professor of Neurobiology
Fellow, Center for Learning and Memory
University of Texas at Austin
Week of September 20, 2010

Structural Substrate of Synaptic Scaling During LTP

 

In many ways, the work of Dr. Kristen Harris epitomizes this year's theme of the structure-function relationship. In her formal public lecture, one of the features of our week-long visitors, she highlighted how structural changes in dendritic spines, the small protrusions through which neurons receive information across synapses, will change over time in response to incoming signals. Work from the Harris lab has shown that dendritic spines will change their shape in response to different stimuli, thus altering their function.

The Harris Lab is interested in changes in synaptic structure that accompany and support learning and memory. Her lab studies long-term potentiation (LTP) in the developing and mature hippocampus, because LTP has many of the physiological characteristics that are expected to occur during learning and memory in the brain. The lab's working hypothesis is that structural synaptic plasticity serves to modify synapses in the creation of new memories, which competes with homeostatic mechanisms that serve to prevent saturation of synaptic strength and neuropathology. To probe this issue, Dr. Harris has focused on dendritic spines, which are the major postsynaptic targets of excitatory axons throughout the brain. Using cutting-edge imaging techniques, members of the Harris lab can measure dendritic spines and their presynaptic axonal partners using 3D reconstruction approaches. Use of these techniques has shown that neighboring spines can operate as independent compartments, and that dendritic spines can also share resources along a dendritic segment, which may serve to regulate the total amount of synaptic input that can be supported by a dendrite.

In addition to her formal public lecture, Dr. Harris also gave more focused talks to smaller audiences and led a neuroscience class for upper-division undergraduates and first-year graduate students. In her formal talk, titled "Structural substrate of synaptic scaling during LTP along mature hippocampal dendrites," she presented research demonstrating that LTP induced by a naturalistic pattern of stimulation results in several ultrastructural changes with time following the induction of LTP in hippocampal brain slices from mature rats. First she described in detail the new methods her laboratory has developed: (1) to rapidly fix brain slices to preserve synaptic and dendritic structure; (2) to provide unbiased sampling of the dendritic population that underwent LTP; and (3) to ensure that alteration in synapse number and size that occurs during slice preparation does not obscure the experimental effects.
Dr. Harris then went on to show new results that offered evidence for synaptogenesis through elevation in shaft and stubby spines, as well as filopodia, during the first 30 minutes following LTP. Later, by two hours after induction of LTP, small spines and their synapses were eliminated, while those that remained were significantly enlarged relative to spine synapses receiving only control stimulation. Presynaptic axonal boutons associated with the small spines were also eliminated. Only the spines with enlarged synapses contained polyribosomes, suggesting that local protein synthesis was needed to sustain the synapse enlargement. Furthermore, the presynaptic axonal boutons associated with synapses that were enlarged during LTP had fewer docked and reserve pool vesicles, suggesting that more vesicles were released, but that machinery needed to replenish them at the enlarged synapses had not yet been recruited to the presynaptic boutons by two hours after induction of LTP. Dr. Harris next described how synapse elimination was perfectly counterbalanced by synapse enlargement, providing strong evidence for structural synaptic scaling following synaptic plasticity. The Harris lab is extending these studies into the intact brain of behaving animals, and preliminary work from the lab suggests that LTP at the medial perforant pathway results in the sharing of resources even beyond the region of synaptic activation (and thus affecting pathways elsewhere along the dendritic arbor).

In a second lecture, Dr. Harris described the process of synaptogenesis and the formation of dendritic spines during development in the hippocampus using their methods of 3D reconstruction from serial electron microscopy at different ages. Animals are not born with dendritic spines, but instead have virtually spine-free dendrites that develop filopodia and shaft synapses during the first postnatal week. By about postnatal day 12, the first mature dendritic spines are beginning to form, and as animals mature more spines emerge. Dr. Harris linked the ontogeny of post-tetanic potentiation (PTP), which lasts only a few seconds, to the beginning of synaptogenesis. Short-term potentiation (STP), lasting about two hours, then takes over at postnatal day 10, before spines have formed. Enduring LTP lasting longer than three hours was linked to the formation of dendritic spines. Dr. Harris presented a new induction paradigm that will allow researchers to test whether stimulation that induces enduring LTP after postnatal day 12 can be used to shift from STP to enduring LTP when given twice. This new paradigm should provide new insights into how experience shapes and prepares dendrites and their synapses for experiences as the animals mature and face new challenges in their environments.