The overall goal of this work is to develop anatomical, functional, and molecular magnetic resonance imaging (MRI) techniques that allow non-invasive assessment of brain function and apply these tools to study plasticity and learning in the rodent brain. MRI techniques are having a broad impact on understanding brain. Anatomical based MRI has been very useful for separating gray and white matter and detecting numerous brain disorders. Functional MRI techniques enable detection of regions of the brain that are active during a task. Molecular MRI is an emerging area, whose major goal is to image a large variety of processes in tissues. The goal of this project is to translate MRI developments in all these areas to study system level changes that occur in the rodent brain during plasticity and learning.
Aim 1 : Over the past few years, we have completed studies in the rodent brain that acquired very high temporal and spatial resolution functional MRI (fMRI) to monitor changes in hemodynamics as a surrogate marker of electrical activity during forepaw stimulation. Over the past year we have demonstrated that fMRI from single arterioles from deep cortex can be effectively imaged using blood volume based MRI techniques. fMRI has long been able to detect individual draining vessels but now we have demonstrated the ability to detect individual vessels in deep tissue. The ability to detect single arterioles complements our previous work detecting single venuoles. In the coming year we will do a detailed analysis of time courses in the different vessel compartments. In addition, a one dimensional imaging technique has been developed that enables us to achieve 50 micron spatial resolution through the cortex and 50 msec temporal resolution. In somatosensory cortex, fMRI signals start in layer 4 at about 600-800 msec consistent with our previous work. In motor cortex fMRI onset corresponds to the neural input in mid-cortical areas. In a model where neural input into somatosensory cortex switches to beginning in layer 2/3 and or layer 5 rather than layer 4, the fMRI onset also switches to layer 2/3 and layer 5. This work is consistent with the hypothesis that the onset of fMRI enables extracting information about the onset of neural activity in a brain region. In the coming year we will verify this idea as well as consider beginning studies on the human brain to look at onset dynamics at high spatial temporal resolution.
Aim 2 : Over the past several years we have demonstrated that manganese chloride enables MRI contrast that defines neural architecture, can monitor activity, and can be used to trace neural connections. Over the last couple of years we have completed the assignment of cortical layers detected using manganese enhanced MRI by comparison to histology and have demonstrated that functional anatomy of several cortical regions of the rodent brain can be defined in individual animals. The ability to detect layers has been applied to a mouse model of neurodegeneration in the olfactory bulb demonstrating the MRI at his level of resolution can detect layer specific degeneration. In addition, we have completed studies that trace the laminar inputs of the olfactory pathway from the olfactory bulb to rodent frontal cortex using manganese enhanced MRI. In a simple fear conditioning experiment (odor with foot shock)a small increase in manganese influx from olfactory cortex to orbital frontal cortex was the only significant change detected. Analysing this change at higher spatial resolution indicated that tracing of manganese was increased by 50% into layer 1 of orbital frontal cortex. This predicts a strengthening of this synapse. There were increases into sub-regions of amygdala as well. In the next year we will verify the synaptic strength increase using optogenetics with MRI and work towards performing slice physiology experiments to verify the manganese work.
Aim 3 : Over the past few years we established a rodent model that uses peripheral denervation to study brain plasticity in response to the injury. Over the past couple of years we have shown that denervation of the infraorbital nerve leads to large increases in barrel cortex responses along the spared whisker pathway as well as large ipslateral cortical activity consistent with our previouus work in the forepaw and hindpaw. fMRI and manganese enhanced MRI predicted a strengthening of thalamo-cortical input along the spared pathway which was verified in slice electrophysiology studies in collaboration with John Isaac's laboratory. Prior to this it was widely believed that the thalamo-cortical input was not capable of strengthening after the critical period. The mechanisms of this strengthening are under study. Interestingly, the denervation is causing a re-activation of the ability of this synapse to demonstrate long term potentiation which is usually lost after the critical period during the first week of life. This LTP is NMDA dependent but not dependent on the NR2B subunit of the NMDA receptor. Experiments to test whether activation of silent synpases explains the LTP detected and the relation between the potentiaition and LTP are being explored. Over the past year we have implemented optogenetics techniques into our fMRI studies to asses plasticity on the ipsilateral side. Furthermore evidence from manganese based track tracing shows a strengthening of input into layer 2/3 and 5. Over the next year we will use slice electrophysiology to verify this strengthening in analogy to the work done on the thalamo-cortical synapse.
Aim 4 : We have begun to explore the use of advanced MRI tools for studying simple learning paradigms in the rodent. In order to accomplish this we have been developing techniques that will enable routine fMRI in awake rodents. While fMRI is widely performed in humans and awake primates there have only been a few scattered studies on awake rodents. Training regimens and techniques to hold the head have been developed over the past two years. Interestingly, we have large differences in brain fMRI activation due to somatosensory stimulation or visual stimulation in the awake animal vs anesthetized animal that are stimulation dependent. Somatosensory stimulation gives a strong fMRI response in anesthetized but not awake rodents and visual stimulation give a strong response in awake but not anesthetized animals. Electrophysiology from these areas verifires this result. This is important to lay the ground work for the best types of stimuli that give good fMRI responses in the awake rodent to better design behavioral pardigms that are consistent with fMRI. In the coming year we will begin to see if fMRI can detect changes in circuit level activity during fear conditioning.
Showing the most recent 10 out of 43 publications
