What information does laminar fMRI provide about cortical sensorimotor circuits? Validation with electrophysiology and targeted nerve injuries

Understanding how the brain processes information and how this is impaired in disease is one of the great challenges in science. The way information flows in cortical circuits consisting of interconnected brain areas, and the processing of information by these circuits is fundamental to brain function. However, we lack effective ways to noninvasively study such 'neural circuits' in animals or humans, because current methods are either too coarse, or they are fine-grained and invasive but cover only a small part of the brain. Functional magnetic resonance imaging (fMRI) is widely used to investigate brain function, and it provides information about which brain areas are active when a person performs a task, but it does not show how information flows in the brain. Although standard fMRI is too coarse to show cortical circuits, they can potentially be studied using high-resolution fMRI at 7T. It has sufficiently high resolution to visualize the layers (laminae) of the cortex, which is essential to understand cortical activity and connectivity, because neurons in different cortical layers perform different functions in the circuit. For instance, information arrives in the middle layers, while deep layers send output to other brain areas. High resolution fMRI is starting to allow the visualization of cortical layers. However, fMRI measures blood flow and metabolism, and does not directly measure neural activity, and the understanding and validation of how these laminar fMRI signals relate to laminar neural activity measured with electrophysiology, is lagging behind. This information can only be gained from studying laminar circuits in animals using both methods. In this study, we compare laminar fMRI, resting state fMRI (rs-fMRI) and laminar electrophysiological recordings in rats to determine what information laminar fMRI provides about the neural computations in the layers. Rs-fMRI is based on the principle that if two areas are connected, their activity fluctuates together. Current rs-fMRI methods are too coarse to resolve how information flows, but laminar rs-fMRI can potentially solve this problem, because knowledge of which cortical layers are connected makes it possible to establish those sending and receiving information.

We will first optimize the data acquisition methods and characterize the laminar signals in the somatosensory cortex of healthy animals. Subsequently we will investigate how cortical circuits respond to the withdrawal of sensory input, done by making injuries to the spinal cord and peripheral nerve. We will make lesions that selectively interrupt one of the main spinal cord pathways (dorsal columns) carrying sensory information from the body to the brain, and compare the response to this partial injury with the response to complete loss of sensory input caused by cutting the peripheral nerve. To understand how reorganization of brain circuits in response to sensory deprivation changes over time, we will carry out laminar fMRI and electrophysiology at different times after injury, under circumstances where there is no repair (dorsal column injuries do not spontaneously repair, nerve regeneration is prevented by ligation). We will then investigate how brain sensorimotor circuits further change under circumstances where there is successful regeneration of peripheral nerve fibres using a nerve crush model. The timing of changes in cortical circuits will be correlated with behavioral measures of returning function.

We expect the study to lead to improved interpretation of (laminar) fMRI, and provide new information about the functional activity and connectivity of sensorimotor circuits. In addition, it will illuminate plastic mechanisms in the brain highly relevant to understanding adaptive functional recovery, rehabilitation strategies utilizing activity-dependent plasticity and the limitations that brain plasticity imposes on functional recovery after peripheral nerve repair.

We propose to study information processing in the laminar circuits in the rat somatosensory and motor cortex and determine how this changes after plastic reorganization following peripheral nerve injury. We will use laminar fMRI, resting-state fMRI (rs-fMRI) and laminar electrophysiology. This work will address the question what information laminar fMRI provides about laminar neural activity, how cortical circuits change after injury and plasticity, and how well these changes are reflected in laminar fMRI signals. Although laminar fMRI is becoming popular in humans, it remains unknown what laminar fMRI signals reflect at a neural level. By using laminar fMRI, rs-fMRI, and electrophysiology in a system in which we can make highly specific changes to the circuits, we aim to answer these mechanistic questions, and learn more about laminar fMRI's utility for inferring neural circuit changes. We will evaluate the correspondence between layer-dependent fMRI and electrophysiology in the sensorimotor system of healthy rats and after targeted injuries. The injuries specifically interrupt input and output of the cortical circuits, and we will monitor information-flow in the circuit before and after injury, and after subsequent nerve regeneration. We will use dorsal column injuries, which interrupt the sensory input to the cortex, and peripheral nerve injuries, which cause large-scale changes to the cortical circuit. We will follow the reorganization of the laminar cortical circuits after two different nerve injuries: cases where regeneration of the nerve is facilitated are contrasted with cases where regeneration is prevented. Since the neural basis of laminar fMRI signals is not understood, this project will improve our ability to interpret these fMRI signals and improve our understanding of laminar neurovascular coupling. This will provide an important tool for obtaining mechanistic insight into brain function and plastic reorganization of cortical circuits.

Understanding neural circuits, such as laminar circuits in the cortex, is critical to understanding brain function. However, suitable tools to study laminar cortical circuits in vivo are lacking because current methods either have insufficient resolution or lack large-scale coverage. Laminar fMRI and resting state fMRI (rs-fMRI) can provide a tool for studying functional activity and connectivity, allowing separation of feedforward-feedback or input and output. Cortical circuits underlie a broad range of brain functions and thus this technique could have significant impact on neuroscience research in many areas. Because fMRI is based on blood flow and metabolism, a better understanding of how fMRI signals relate to neural activity will improve the interpretation of fMRI data. This project is expected to provide important information about what neural processes (laminar) fMRI and rs-fMRI represent. Improving the tools to study brain circuits has implications for research into a wide range of neurological conditions. This can improve inferences that can be made about cognitive processing or disease processes, which in turn increases the probability of finding new treatments for a variety of neurological disorders. It may also lead to wider use of fMRI in clinical research and diagnosis.

This project can provide important insights into the function of sensorimotor cortical circuits, and injury- and plasticity-induced changes to these circuits. Spinal cord injury results in devastating loss of function. The damaged CNS does not regenerate but modest spontaneous functional recovery is attributed to plasticity in the brain and spinal cord. Development of rehabilitative strategies aimed at maximizing return of function by driving activity-dependent plasticity is a rapidly growing research area important in spinal cord injury, brain injury and stroke. Following peripheral nerve injury, successful repair and regeneration can occur but functional recovery is slow (1-5 years) and incomplete. Cortical plasticity is thought to be one of the main factors limiting return of function. In addition, neuropathic pain is common following spinal cord and peripheral nerve injury. This pain is difficult to treat, and although its mechanisms remain poorly understood, it is attributed partly to maladaptive plasticity. Little is known about how brain circuits adapt after spinal cord or nerve injuries, but study of these circuits in injury models is likely to lead to a better understanding of its mechanisms, and has potential to advance treatments. For instance, better understanding of the alterations to cortical circuits after injury could lead to improved methods of monitoring regeneration and evaluating the efficacy of rehabilitation. Plasticity is also implicated in phantom sensations and neglect disorders, and understanding somatosensory and motor circuits can potentially benefit clinicians and scientists studying a range of conditions. If improved understanding feeds through to improved treatment options, then the ultimate beneficiaries would be patients.

The project can potentially drive further development of high-resolution fMRI-technology in humans. Developing methodology and understanding mechanisms can facilitate translation to humans. Laminar fMRI in humans is a key reason for many institutions to acquire human 7T scanners, and translating the findings of this project to humans can potentially impact MR-physics research, and further technological development of ultra-high field human scanners. Demonstrating the importance of laminar fMRI in a well-defined animal model of disease can provide an incentive for researchers and manufacturers to increase the spatial resolution of scanners by improving the scanners' hardware and software. Ability to use laminar fMRI to probe cortical circuits would be a major step in advancing the utility of fMRI, and one large enough to warrant technological development.