G·A·B·B·A

The current mechanistic view on how the brain is able to store memories over long periods of time is based on two key concepts. The first is that memories are stored in the configuration of the connectivity of neurons in an assembly and in the set of synaptic weights of those connections; the second being that experience can mold and rewire the network connectivity and its synaptic weights. It becomes clear that the understanding of cortical function will always require the unraveling of synaptic connectivity in cortical circuits, that is, establishing the wiring diagrams between individual neurons.

In the present work, a first effort was made in order to investigate the excitatory synaptic local circuitry in the adult mouse auditory cortex, a brain area critically involved in sound encoding required for proper associative motivational leaning.

For this purpose, coronal whole-brain slices from adult (8-14 weeks old) C57Bl6/6J mice were used. Several simultaneous quadruple whole-cell recordings from layer 2/3 and layer 5 pyramidal neurons were made, a method that allows for quantitative functional measures of synaptic connectivity at the level of individually indentified neurons. It was observed that local circuitry is characterized by low connection probabilities between pairs of neurons, and that bidirectional connections are more common than expected in a random network. The distribution of synaptic connections strengths (defined as the peak of excitatory postsynaptic potential (EPSP) amplitude), has a heavier tail and implies that synaptic weight is concentrated among few synaptic connections. In both layers it was found the existence of rare but reliable large-amplitude synaptic connections, which are likely to contribute strongly to reliable information processing. Moreover, another central finding is that the EPSP amplitude variability can be ascribed to changes in the number of release presynaptic sites, or due to the probability of neurotransmission release, implying that modulations in synaptic transmission can be described by changes in both parameters independently.

In the second part, the relative contribution, with precise temporal resolution, of excitatory and inhibitory drives that impinge onto layer 2/3 pyramidal neurons was investigated. The strict balance of these two synaptic conductances plays a critical role in cortical function and in the shaping of the tuning properties of cortical neurons. It is of utmost importance to describe how this balance is achieved and maintain. By means of intracortical extracellular stimulation of two independent but convergent input pathways into layer 2/3 neurons, synaptic conductances could be recorded and decomposed into their excitatory and inhibitory components. It was observed that excitatory/inhibitory balance is of equal magnitude in both stimulated pathways, and that on average a time difference less than 2 ms between the arrival of inhibition compared with the excitation favors for a monosynaptic nature of the stimulated intracortical projections that synapses onto the recorded layer 2/3 pyramidal neurons. On the other hand, it was observed that on almost half of the recorded neurons, the excitation conductance was flanked by two inhibitory barrages, a phenomenon never described so far. A possible feedback or feedforward inhibitory circuitry made by local interneurons could explain this observation.

In the third part, one final question was posed: are the features that describe local synaptic circuitry changed upon optogenetic manipulation in a behavioural task? By means of combining expression of channelrhodopsin in auditory cortex pyramidal neurons, with their direct photostimulation in the context of a behaviour task, it was possible to assess the role of a subset of neurons in driving behaviour. Possible changes in their intrinsic interconnectivity were also studied upon learning. Though extremely labour intense, it was concluded that ChR2-based optical microstimulation can be used to dissect the impact of precisely timed action potentials in a subset of neurons in driving behaviour. Whole-cell recordings from layer 2/3 neurons from the subset of mice that reached correct performance levels were performed as before. It was observed that ChR2-expressing neurons in trained mice had similar intrinsic excitability features when compared with non-trained mice. The recorded EPSP amplitudes from pairs of connected neurons had similar rages among both groups of mice, indicating that periodic depolarizations of ChR2-positive neurons does not induce any synaptic scaling effect in these neurons.