The balance between excitation and inhibition in the cortex is crucial in determining sensory processing. Because the amount of excitation varies, maintaining this balance is a dynamic process; yet the underlying mechanisms are poorly understood. We show here that the activity of even a single layer 2/3 pyramidal cell in the somatosensory cortex of the rat generates widespread inhibition that increases disproportionately with the number of active pyramidal neurons. This supralinear increase of inhibition results from the incremental recruitment of somatostatin-expressing inhibitory interneurons located in layers 2/3 and 5. The recruitment of these interneurons increases tenfold when they are excited by two pyramidal cells. A simple model demonstrates that the distribution of excitatory input amplitudes onto inhibitory neurons influences the sensitivity and dynamic range of the recurrent circuit. These data show that through a highly sensitive recurrent inhibitory circuit, cortical excitability can be modulated by one pyramidal cell.

No abstract

The neurones of the medial superior olive (MSO) are the most temporally sensitive neurones in the brain. They respond to the arrival time difference of sound at the two ears with a microsecond resolution; these interaural time differences are used to localize low-frequency sounds. In addition to the excitatory inputs from each ear, the MSO neurones also receive binaural glycinergic projections, which have a critical role in sound localization processing. Recently, it was shown that the glycinergic input to the MSO undergoes an experience-dependent structural reorganization after hearing onset. To explore the maturation of inhibition during the development of sound localization on a cellular level, glycinergic currents and potentials were measured in gerbil MSO principal cells from postnatal (P) day P12-P25 by whole-cell patch-clamp recordings. The synaptic glycinergic currents accelerated to rapid decay kinetics (∼2 ms) and rise times (∼0.4 ms) after hearing onset, reaching maturity around P17. Since the kinetics of miniature glycinergic currents did not change with age, it is likely that a higher degree of transmitter release synchrony is the underlying mechanism influencing the acceleration of the kinetics. During the same period, the synaptic glycinergic potentials accelerated four-fold, largely as a result of a prominent decrease in input resistance. In accordance with a reorganization of the glycinergic inputs, the evoked peak conductances decreased more than two-fold, together with a three-fold reduction in the frequency of miniature events after hearing onset. These age-dependent changes were absent in animals that had been reared in omni-directional noise, indicating that an experience-dependent pruning of synaptic inputs is important for the maturation of functional inhibition in the MSO. Taken together, these striking developmental adjustments of the glycinergic inhibition in the MSO most probably reflect an adaptation to improve the encoding of auditory cues with great temporal precision and fidelity during the maturation of sound localization behaviour.

The ascending auditory pathway is characterized by parallel processing. At the brain stem level, several structures are involved that are known to serve different well-defined functions. However, the function of one prominent brain stem nucleus, the rodent superior paraolivary nucleus (SPN) and its putative homologue in other mammals, the dorsomedial periolivary nucleus, is unknown. Based on extracellular recordings from anesthetized gerbils, we tested the role of the SPN in sound localization and temporal processing. First, the existence of binaural inputs indicates that the SPN might be involved in sound localization. Although almost half of the neurons exhibited binaural interactions (most of them excited from both sides), effects of interaural time and intensity differences (ITD; IID) were weak and ambiguous. Thus a straightforward function of SPN in sound localization appears to be implausible. Second, inputs from octopus and multipolar/stellate cells of the cochlear nucleus and from principal cells of the medial nucleus of the trapezoid body could relate to precise temporal processing in the SPN. Based on discharge types, two subpopulations of SPN cells were observed: about 60% of the neurons responded to pure tones with sustained discharges, with irregular spike patterns and no phase-locking. Only four neurons showed a regular spike pattern ("chopping"). About 40% of the neurons responded with phasic ON or OFF discharges. Average first spike latency observed in neurons with sustained discharges was significantly shorter than that of ON responders, but had a considerably higher trial-to-trial variation ("jitter"). A subpopulation of ON responders showed a jitter of less than +/-0.1 ms. Most neurons (66%) responded to sinusoidally amplitude-modulated sounds (SAM) with an ongoing response, phase-locked to the stimulus envelope. Again, ON responders showed a significantly higher temporal precision in the phase-locked discharge compared with the sustained responders. High variability was observed among spike-rate-based modulation transfer functions. Histologically, a massive concentration of cytochemical markers for glycinergic input to SPN cells was demonstrated. Application of glycine or its blockade revealed profound effects of glycinergic inhibition on the auditory responses of SPN neurons. The existence of at least two subpopulations of neurons is in line with different subsets of SPN cells that can be distinguished morphologically. One temporally less precise population might modulate the processing of its target structures by providing a rather diffuse inhibition. In contrast, precise ON responders might provide a short, initial inhibitory pulse to its targets.