Cellular determinants of odor discrimination behaviour in rodents

Our senses are our brains' window to the world - they are also the scientists' window to understanding the brain. By investigating how information about the external world is processed by networks of neurones, we aim to elucidate how the brain performs its variety of tasks. Many lines of research require invasive methods to observe or disturb the system and asses subsequent alterations in function. For obvious reasons, such lines of research have to use animal models, most notably the mouse as a popular mammalian model system allowing genetic modifications. In order to investigate sensory processing in mice, it is beneficial to focus on senses of prime importance to the animal of choice. Mice are nocturnal animals, thus my model system of choice is the olfactory system - not only because mice form direct associations between smells and food rewards easily and naturally, but also because in contrast to other senses the olfactory system is organized in a rather simple way: Receptor cells in the nose project to a brain area called olfactory bulb where they connect mainly with one group of neurones that in turn directly project to various areas of the brain involved in topics so diverse as memory formation, fear processing, decision making etc. Thus, the olfactory bulb seems to be the prime area for the processing of smell-specific information. In the olfactory bulb, activated neurones tend to inhibit other neurones mediated by a large number of small cells called 'granule cells' (GCs). Many people have speculated that these connections are important for odour processing, e.g. by enhancing the differences between similar odours: Strongly activated neurones inhibit weakly activated ones, thus further reducing their activity and increasing the initial difference. Though this particular model is hotly disputed by many, most people agree that such inhibitory connections are of crucial importance in shaping the way odours are represented in the OB. In my project I will directly test this hypothesis by altering these inhibitory connections, measuring the alterations and assessing their behavioural consequence. A prime challenge in this project is to restrict the modification to only these connections. Most classical genetic modifications eliminate or modify one gene in the entire body. Our approach, however, is to inject a virus directly into the olfactory bulb. This virus will be rendered harmless (that is, it can infect cells but not grow in them) but produce a 'molecular pair of scissors'. It will cut out one specific gene in all cells the virus infects. This is made possible by the use of a specific strain of mice, genetically modified to contain 'cutting sites' on both sides of our target gene that are specifically recognised by the molecular scissors. Altering genes important in communication in the olfactory bulb we can test the hypothesis that the inhibitory circuitry of the early olfactory system is crucial for odour processing. In a first step, we have to demonstrate that our specific modifications have indeed changed communication. To achieve this we will record the activity of single neurones in the olfactory bulb and measure the amount of inhibition they receive. In order to investigate whether our modifications also influence behaviour we will accurately (with millisecond precision) measure the time it takes a mouse to discriminate two odours. This time is highly reproducible between animals allowing us to measure the small changes associated with modifications restricted to only GCs in the olfactory bulb. By analyzing the effects of altered inhibition in the circuitry of the olfactory bulb on behaviour we will make one step towards understanding how odours are represented and processed in the olfactory system. Thus, our research will give us the chance to glance into the brains' workings through the rodents' sense of smell.

The aim of this project is to investigate how the synaptic circuitry of the olfactory bulb (OB) governs odor discrimination in mice. We have shown that deletion of the AMPA receptor subunit GluR-B in the forebrain, rendering AMPA receptors Ca permeable, results in improved odor learning / discrimination. Here, I will restrict such modification to granule cells (GCs) in the olfactory bulb by injecting Cre-expressing virus into the OB of mice transgenic for loxP flanked GluR-B. The dendro-dendritic synapses between mitral cells (MCs) and GCs faces an unusual arrangement in that Ca influx through glutamate receptors on GCs can directly result in release of the inhibitory transmitter GABA back onto MCs. As GluR-B normally blocks Ca influx through AMPA receptors, its deletion is hypothesized to result in increased inhibition. In both spatial and temporal models of odor representation inhibition plays a pivotal role; thus, GluR-B deletion could improve odor discrimination. Under normal conditions, synaptic Ca influx is mediated predominantly by NMDA receptors. Ablating the subunit NR1 by injecting Cre expressing virus into the OB of NR1-2lox animals is thus hypothesized to result in reduced inhibition and possibly impaired odor discrimination. I will test these hypotheses on three levels:1) Discrimination times (DT) are a highly reproducible tool for analyzing olfactory behavior. I will thus measure DTs in mice with GC-specific deletion of GluR-B and NR1. 2) In vivo whole cell recordings from MCs will be used to assess the amount of inhibition impinging on MCs from the entire intact GC network as a result of MC activation. 3) The subcellular aspects of GluR-B and NR1 deletion will be analysed by measuring synaptic IVs in vitro as well as Ca influx into GC dendrites. Thus, combining behavioral and physiological analysis in mice selectively modified at the MC-GC synapse will allow us to molecularly dissect the mechanisms of odor processing in the olfactory bulb.