Final Report Summary - SEROTONIN IN ACTION (Serotonergic modulation of cortical activity in vivo: from neurons to networks and implications for autism spectrum disorders.)
Project Objectives Summary
Anti-depressant drugs such as fluoxetine (Prozac) increase levels of serotonin (5HT) in the brain, but little is known about how they affect the function of the nervous system. 5HT is implicated in a wide spectrum of brain functions and disorders. However, its precise role remains controversial and enigmatic. This is partly due to the previous inability to adequately isolate 5HT action in vivo. Recent developments in genetic tools and imaging techniques now offer an unprecedented opportunity to investigate such mechanisms in the living brain. This project investigated how 5HT transmission modulates cortical activity in the awake-behaving mouse, under physiologically relevant conditions. Our objectives were to determine the action of 5HT in awake-behaving animals in three main ways: (1) on specific neuronal populations, at the network and single-cell level, by investigating the state-dependent activity in three main classes of inhibitory neurons both in physiological conditions and after the local application of 5HT modulators; (2) when endogenously released, by monitoring the activity of 5HT neurons at the level of their axon terminals in the cortex; and (3) on pathologic cortical network activity in an animal model of autism spectrum disorders. We used the primary visual cortex as a model cortical area, investigating the role of 5HT in visually-evoked and activity-dependent cortical activation under normal conditions and during network dysfunction in autism spectrum disorders. In this project we used state of the art genetic tools and in vivo two-photon calcium imaging, allowing us to investigate neuronal populations activity with single-cell resolution.
Summary of work performed and main results achieved
• The Fellow established the technique of two-photon calcium imaging in awake-behaving mice. The project was expected to begin with experiments with anaesthetized animals as awake-behaving imaging is technically challenging. However, the Fellow surpassed the expectations and timeline of the project on this front and is now an expert in imaging techniques with awake-behaving animals. This allows investigating neuronal activity in physiological (non-anesthetized) conditions.
• The fellow was involved in the development of the two-photon imaging set-up and the improvement of surgical procedures in order to be able to image neuronal activity deep within the cortical layers. This allowed the breadth of the project to be expanded to examine functional differences both within as well as between cortical layers.
• The first part of this research project has been to establish the activity pattern of excitatory and inhibitory neurons in the mouse primary visual cortex, during different behavioural states (resting state and locomotion, darkness and visual stimulation). This resulted in a rich data set including awake-behaving imaging of 4 different cell-type specific neuronal populations, and over 6000 neurons. The main results of these experiments have been presented in several national and international conferences as well as during invited lectures. These results are presented in a manuscript that is currently in revision. Manuscript reference: Janelle M.P. Pakan, Scott C. Lowe, Evelyn Dylda, Sander W. Keemink, Stephen P. Currie, Christopher A. Coutts and Nathalie L. Rochefort. Behavioural-state modulation of inhibition is context-dependent and cell type specific in mouse visual cortex. 2016. Abstract of the article: Cortical responses to sensory stimuli are modulated by behavioral state. In the primary visual cortex (V1), visual responses of pyramidal neurons increase during locomotion. This response gain was suggested to be mediated through inhibitory neurons, resulting in the disinhibition of pyramidal neurons. Using in vivo two-photon calcium imaging in layers 2/3 and 4 in mouse V1, we reveal that locomotion increases the activity of vasoactive intestinal peptide (VIP), somatostatin (SST) and parvalbumin (PV)-positive interneurons during visual stimulation, challenging the disinhibition model. In darkness, while most VIP and PV neurons remained locomotion responsive, SST and excitatory neurons were largely non-responsive. Context-dependent locomotion responses were found in each cell type, with the highest proportion among SST neurons. These findings establish that modulation of neuronal activity by locomotion is context-dependent and contest the generality of a disinhibitory circuit for gain control of sensory responses by behavioural state.
• Additionally, the Fellow performed a set of experiments (over 500 neurons) that showed a clear effect of 5HT as well as a 5HT1A receptor agonist (5HT1A) and antagonist (WAY-100135) on the pattern of neuronal activity in V1. This was first shown in wild-type animals and then additionally investigated in transgenic mouse lines where a subpopulation of interneurons could be isolated (Parvalbumin expressing interneurons). Data collection for this part of the project is complete and late phase analysis is currently ongoing.
• The Fellow has also optimized technical procedures for the labelling of axons projecting from the raphe nucleus to the visual cortex. Additionally, protocols and imaging equipment have been established to image axons in vivo, which is a technically challenging task. Finally, the Fellow has completing a number of experiments imaging axons (9 mice) and is currently analysing results and continuing to collect data towards a publication.
Conclusions and Potential Impact
The increased gain of visual responses during locomotion provides a model to elucidate the circuit mechanisms underlying behavioural-state dependent changes of sensory responses. In this project, we found that the modulation of neuronal activity by locomotion is context-dependent and cell type specific, in layer 2/3 and layer 4 of mouse primary visual cortex. This work was essential to establish the baseline, state-dependent responses of different neuronal populations for two main reasons: 1) to resolve substantial controversy in the literature regarding the state-dependent responses of inhibitory neurons in primary visual cortex and 2) to investigate the modulation of state-dependent responses by 5HT in physiological and pathological conditions. Towards the second point, ongoing analysis of experimental data collected will reveal how 5HT transmission modulates cortical activity of excitatory and inhibitory neurons in the visual cortex of awake-behaving mice. The results will apply beyond the visual system to more broadly inform on cellular mechanisms of 5HT action in the central nervous system and its dysfunction in neurological disorders, leading to future advancements for therapeutic targeting.
Anti-depressant drugs such as fluoxetine (Prozac) increase levels of serotonin (5HT) in the brain, but little is known about how they affect the function of the nervous system. 5HT is implicated in a wide spectrum of brain functions and disorders. However, its precise role remains controversial and enigmatic. This is partly due to the previous inability to adequately isolate 5HT action in vivo. Recent developments in genetic tools and imaging techniques now offer an unprecedented opportunity to investigate such mechanisms in the living brain. This project investigated how 5HT transmission modulates cortical activity in the awake-behaving mouse, under physiologically relevant conditions. Our objectives were to determine the action of 5HT in awake-behaving animals in three main ways: (1) on specific neuronal populations, at the network and single-cell level, by investigating the state-dependent activity in three main classes of inhibitory neurons both in physiological conditions and after the local application of 5HT modulators; (2) when endogenously released, by monitoring the activity of 5HT neurons at the level of their axon terminals in the cortex; and (3) on pathologic cortical network activity in an animal model of autism spectrum disorders. We used the primary visual cortex as a model cortical area, investigating the role of 5HT in visually-evoked and activity-dependent cortical activation under normal conditions and during network dysfunction in autism spectrum disorders. In this project we used state of the art genetic tools and in vivo two-photon calcium imaging, allowing us to investigate neuronal populations activity with single-cell resolution.
Summary of work performed and main results achieved
• The Fellow established the technique of two-photon calcium imaging in awake-behaving mice. The project was expected to begin with experiments with anaesthetized animals as awake-behaving imaging is technically challenging. However, the Fellow surpassed the expectations and timeline of the project on this front and is now an expert in imaging techniques with awake-behaving animals. This allows investigating neuronal activity in physiological (non-anesthetized) conditions.
• The fellow was involved in the development of the two-photon imaging set-up and the improvement of surgical procedures in order to be able to image neuronal activity deep within the cortical layers. This allowed the breadth of the project to be expanded to examine functional differences both within as well as between cortical layers.
• The first part of this research project has been to establish the activity pattern of excitatory and inhibitory neurons in the mouse primary visual cortex, during different behavioural states (resting state and locomotion, darkness and visual stimulation). This resulted in a rich data set including awake-behaving imaging of 4 different cell-type specific neuronal populations, and over 6000 neurons. The main results of these experiments have been presented in several national and international conferences as well as during invited lectures. These results are presented in a manuscript that is currently in revision. Manuscript reference: Janelle M.P. Pakan, Scott C. Lowe, Evelyn Dylda, Sander W. Keemink, Stephen P. Currie, Christopher A. Coutts and Nathalie L. Rochefort. Behavioural-state modulation of inhibition is context-dependent and cell type specific in mouse visual cortex. 2016. Abstract of the article: Cortical responses to sensory stimuli are modulated by behavioral state. In the primary visual cortex (V1), visual responses of pyramidal neurons increase during locomotion. This response gain was suggested to be mediated through inhibitory neurons, resulting in the disinhibition of pyramidal neurons. Using in vivo two-photon calcium imaging in layers 2/3 and 4 in mouse V1, we reveal that locomotion increases the activity of vasoactive intestinal peptide (VIP), somatostatin (SST) and parvalbumin (PV)-positive interneurons during visual stimulation, challenging the disinhibition model. In darkness, while most VIP and PV neurons remained locomotion responsive, SST and excitatory neurons were largely non-responsive. Context-dependent locomotion responses were found in each cell type, with the highest proportion among SST neurons. These findings establish that modulation of neuronal activity by locomotion is context-dependent and contest the generality of a disinhibitory circuit for gain control of sensory responses by behavioural state.
• Additionally, the Fellow performed a set of experiments (over 500 neurons) that showed a clear effect of 5HT as well as a 5HT1A receptor agonist (5HT1A) and antagonist (WAY-100135) on the pattern of neuronal activity in V1. This was first shown in wild-type animals and then additionally investigated in transgenic mouse lines where a subpopulation of interneurons could be isolated (Parvalbumin expressing interneurons). Data collection for this part of the project is complete and late phase analysis is currently ongoing.
• The Fellow has also optimized technical procedures for the labelling of axons projecting from the raphe nucleus to the visual cortex. Additionally, protocols and imaging equipment have been established to image axons in vivo, which is a technically challenging task. Finally, the Fellow has completing a number of experiments imaging axons (9 mice) and is currently analysing results and continuing to collect data towards a publication.
Conclusions and Potential Impact
The increased gain of visual responses during locomotion provides a model to elucidate the circuit mechanisms underlying behavioural-state dependent changes of sensory responses. In this project, we found that the modulation of neuronal activity by locomotion is context-dependent and cell type specific, in layer 2/3 and layer 4 of mouse primary visual cortex. This work was essential to establish the baseline, state-dependent responses of different neuronal populations for two main reasons: 1) to resolve substantial controversy in the literature regarding the state-dependent responses of inhibitory neurons in primary visual cortex and 2) to investigate the modulation of state-dependent responses by 5HT in physiological and pathological conditions. Towards the second point, ongoing analysis of experimental data collected will reveal how 5HT transmission modulates cortical activity of excitatory and inhibitory neurons in the visual cortex of awake-behaving mice. The results will apply beyond the visual system to more broadly inform on cellular mechanisms of 5HT action in the central nervous system and its dysfunction in neurological disorders, leading to future advancements for therapeutic targeting.