This paper by Kruglikov and Rudy examines in detail the profile of neuromodulatory influences on GABA release from fast-spiking (FS), parvalbumin-containing interneurons in sensory neocortex. The work elegantly demonstrates that this interneuron subtype is exquisitely sensitive to a diverse range of neuromodulatory chemicals including those acting on muscarinic, purinergic, serotonergic, and GABA_B receptors. Agonists at each of these receptors produced a strong inhibition of GABA release from electrically stimulated FS synaptic terminals and, as a result, reduced inhibitory influence both locally in cortex and on ascending thalamocortical projections. These interneurons are of particular interest currently in schizophrenia research as functional markers for them are found to be robustly reduced in postmortem brain samples from schizophrenic patients. They are also one of the key interneuron subtypes involved in the generation of certain EEG rhythms—in particular, those in the gamma (30-80 Hz) band—involved in primary sensory processing, short-term memory, and cortico-cortical communication. The pattern of generation of gamma rhythms is disrupted in people with schizophrenia.

However, it is hard to see direct implications for our understanding of processes underlying cortical dysfunction in schizophrenia from this work. This is mainly due to the fact that only a subset of parvalbumin-containing FS cells appear to be affected in schizophrenia and related animal models. It is not yet known what makes this subset of interneurons so labile in psychiatric illness, but there is much exciting work ongoing which is highly suggestive of a role for NMDA receptor-mediated excitation and changes in redox state. The paper does not subclassify the FS interneurons studied directly, but there are a couple of issues which may be of interest to the field:

1. Firstly, the importance of neuromodulation for generating gamma rhythms can clearly be seen in Figure 6. Forty Hz artificial stimulation—matching the modal frequency of activation of these neurons during network gamma rhythms—terminates pyramidal cell action potential generation in the absence of neuromodulatory influences. This suggests that inhibition-based gamma oscillations in cortex would be useless as an information coding strategy. However, the reduced inhibitory post-synaptic potential size (and thus effective duration) under muscarinic neuromodulation permits sparse, but precisely timed pyramidal cell action potentials—exactly the signature for principal cell spiking during gamma rhythms in vivo. Loss of parvalbumin from FS cells causes a large increase in GABA release, and thus increased inhibitory post-synaptic potential size and duration. It is therefore interesting to consider what the reduced parvalbumin levels in schizophrenic cortex can tell us: is it a primary cause of the observed decrease in gamma rhythm generation, owing to its enhancement of GABA release and thus perhaps termination of pyramidal cell spiking? Or is it a compensatory mechanism, as previously proposed, for an underlying deficit in FS cell excitation?

2. Secondly, it is interesting to note in this paper that cannabinoid receptor activation did not change evoked GABA release from FS cells at all. There is a growing corpus of work linking cannabis use to increase risk of psychotic episodes and exacerbate symptoms of schizophrenia. Given the proposed critical role of FS cells in cortical dysfunction in schizophrenia, it is perhaps surprising that no effect was seen. However, the study used artificial, electrical stimulation to evoke GABA release from these cells. In active networks, cannabinoid agonists can act on CCK-containing interneurons, which may indirectly change excitability of parvalbumin-containing FS interneurons, changing GABA release patterns via altered rates of action potential generation. In addition, cannabinoid receptors have been shown to directly reduce excitatory post-synaptic potential size, something that might be expected to directly reduce FS cell recruitment during network activity.

View all comments by Miles Whittington