Plasticity of inhibitory synaptic transmission in the hippocampus

The building blocks of the brain are nerves cells, also called neurons, connected to each other by synapses. Neurons form two categories: excitatory neurons promote activity in their connected neurons, whereas inhibitory neurons depress activity. Even though inhibitory neurons are less numerous than excitatory neurons (20%), they play a central role in brain function and computation. Disruption of inhibitory neurons leads to an imbalance in excitation and inhibition that can, in turn, lead to diseases such as epilepsy, schizophrenia and autism spectrum disorders. Thus, understanding what regulates the strength of excitatory and inhibitory synapses is an important research goal. Interestingly, although there has been much focus on plasticity of excitatory synapses, plasticity of inhibitory synapses has been largely neglected. In this BBSRC project, we aim to investigate the phenomenology and function of inhibitory synaptic plasticity.
The hippocampus is a brain area that plays an important role in episodic memory and spatial navigation. Neuronal network activity in the hippocampus exhibits multiple discrete states that are identified by rhythmic oscillations and perform distinct computational functions. Memories are initially encoded during exploration when theta (5-12Hz) and gamma (30-120Hz) frequency oscillations are prevalent whereas consolidation of memories occurs offline in periods of transient sharp wave ripple (~200Hz) activity during rest or sleep. Specific subtypes of inhibitory neurons have been found to be active in these distinct behavioural states and, intriguingly, inhibitory plasticity has been shown to depend on the precise co-ordination of inhibitory and excitatory neuron firing and on the state of the neural network. This indicates that different behavioural states are likely to trigger distinct forms of inhibitory plasticity at specific inhibitory synapses. However, there is very limited evidence for the role of inhibitory plasticity in the hippocampus.
Our hypotheses are that plasticity of inhibitory synapses: a) can regulate network computations, b) is specific to the inhibitory neurons subtypes and c) determines input selectivity for excitatory neurons in a behavioural state dependent manner. We propose a joint effort from an experimental team, which has expertise in synaptic plasticity in the hippocampus, with a computational team that recently proposed a theoretical model of inhibitory plasticity. By a tight interaction allowing the computational model to be informed by and to guide experiments and experiments to refine the model, we aim to investigate how neural activity engages inhibitory plasticity depending on the inhibitory neuron subtype, and how inhibitory plasticity shapes network computations in the hippocampus. This work will provide important information on the mechanisms and function of inhibitory plasticity that will ultimately improve our understanding of how inhibition may be modified in disease states potentially leading to new therapies.

Inhibitory interneurons play a central role in brain function. They maintain excitatory-inhibitory balance, and generate a range of oscillatory network dynamics. Disruption to these processes by inappropriate inhibition can lead to diseases such as epilepsy and autism spectrum disorders. In the hippocampus there are multiple different interneuron subtypes that target different regions of the network. We have chosen to focus on two distinct interneuron subtypes in the CA1 region: basket cells (BCs) and oriens lacunosum moleculare (OLM) cells that inhibit the soma/proximal and distal dendrites of CA1 pyramidal cells respectively and are active within different phases of theta, gamma and sharp wave ripple oscillations.
Network computations can be learned and dynamically regulated through the process of synaptic plasticity. It is increasingly clear that the diverse array of interneuron subtypes within the hippocampus perform specific functions on network processing. Furthermore, recent evidence shows that inhibitory synaptic transmission is not static but can be dynamically regulated in an activity-dependent manner with far-reaching but currently unexplored implications for the hippocampus. Thus, we hypothesise that inhibitory plasticity will be engaged by BCs and OLM cells during different behaviours and that this will in turn dynamically regulate the preferred inputs for CA1 pyramidal cells.
Here, we will address these hypotheses using a combination of computational modelling and electrophysiological recording in hippocampal slices coupled with optogenetic techniques to stimulate and record from select subtypes of interneurons. The experimental and theoretical aspects will be interdigitated to enable theoretical predictions to be informed by experimental work and vice versa. The goal is to understand when inhibitory plasticity is induced at specific interneuron-pyramidal cell synapses and how it regulates hippocampal function.

Who will benefit from the research?

As well as specific academic beneficiaries, the public (particularly school pupils and teachers) and wider academic community will benefit from the increase in knowledge about the role of inhibitory synaptic plasticity. In addition, sectors of the pharmaceutical industry working to develop effective drug therapies for neurological diseases will also benefit from the proposed work. Indirectly, and in the long term, people suffering from such diseases may also benefit. Therefore, there is the potential for beneficial impact on both the health and wealth of the UK.

How will they benefit from this research?

Enabling us to adapt to our environment is one of the most fundamental functions of the brain, a process that is believed to be underpinned by synaptic plasticity. Since memory is so integral to all our lives, gaining knowledge about the mechanisms of synaptic plasticity is of interest not only to the academic community but also to the wider public.

Pharmaceutical industry: Research into numerous neurological diseases such as schizophrenia, autism and Alzheimer's disease has found deficits in synaptic plasticity that could contribute to disease symptoms. Our close working relationship with specific pharmaceutical companies means our work is likely to enhance their understanding of the fundamental science of learning and memory, pharmacological approaches to manipulating it and putative novel drugs and targets. JRM has ongoing collaborations with Eli Lilly & co (through the Centre for Cognitive Neuroscience) to study the effects of receptor agonists on synaptic transmission in the hippocampus. JRM co-supervises CASE award studentships at Bristol University since October 2011. Through the research described in this proposal we can offer these companies academic expertise to further this goal. This is particularly important since Eli Lilly & co are developing receptor selective agonists for use in the treatment of cognitive disorders.

The social impact and economic costs of the diseases mentioned above are enormous. Therefore our work will benefit society from the advances we make in investigating mechanisms that may underlie such diseases, and will benefit the economy both in terms of costs saved in care for patients suffering from these conditions, and also in profits from pharmaceuticals developed and sold by UK-based companies. We acknowledge that these indirect benefits may take several years before they are realised.

High-tech industry: This work will benefit current and future developers of smart technologies, since the inhibitory learning rule developed for this project is likely to inspire new machine learning algorithms, new implementations in neuromorphic engineering, and new learning rules for intelligent robotics. We will make use of CC's contact, Tom Schaul, at Google Deepmind to encourage the use of our model of inhibitory plasticity in novel machine learning techniques.

Educational impact: We will train a new generation of scientists by training the staff in our laboratory and by teaching at summer schools. By teaching at Bristol and Imperial, CC and JRM will train a new generation of non-academic workers in the UK, teaching a solid skillset for working in pharmaceutical, biotechnological, or engineering companies such as high-tech companies using machine learning or robotics, but also in banks and insurances that use artificial neural network techniques.