Summary

Although gamma-aminobutyric acid (GABA) is thought of as the primary inhibitory neurotransmitter in vertebrate nervous systems, it has been suggested that it has an excitatory function in early developmental stages of the brain. Further nuance has been added to the picture by evidence that in some brain regions GABAergic innervation is inhibitory even in early development. Here, Murata and Colonnese demonstrate, using Designer Receptors Exclusively Activated by Designer Drugs (DREADDs), that GABAergic interneurons cause local excitation in hippocampal CA1 at postnatal day 3 (P3), which switches to inhibition at postnatal day 7 (P7), while cortical GABAergic interneurons cause inhibition at both ages.

Findings

Murata and Colonnese expressed either the inhibitory kappa-opioid receptor DREADD or the excitatory DREAD hM3Dq in hippocampus or visual cortex of GAD2-cre mice via an adeno-associated virus. The majority of the recordings were made in unanaesthetised, head-fixed mice using multi-electrode arrays, although some current-clamp recordings were made in brain slices. Expression of the DREADDs were limited to GAD2+ neurons (GAD2 is an enzyme in the production of GABA, and is used as a marker for GABAergic neurons). Only animals with viral expression of the DREADDs surrounding the recording site and limited to the region of injection were analysed.

First, current clamp recordings are used to establish that activating the inhibitory and excitatory DREADDs have a hyperpolarising and depolarising effect on GABAergic neurons respectively at both P3 and P11, demonstrating that changes in the effect of DREADD agonist administration are due to changes in the effect of GABAergic neurons, and not changes in the effect of the DREADDs.

In line with the paradoxical excitatory early-development effect of GABA, neural activity in the hippocampus at P3 was suppressed by subcutaneous injection of SalB (the agonist for the inhibitory DREADD). That is, GABA activity was suppressed by the DREADD, and activity in the region that receives GABAergic projections was also suppressed, suggesting that an excitatory effect had been reduced. Specifically, multiunit activity in the pyramidal layer decreased, the amplitude and occurence but not duration of early Sharp Waves (eSPWs) and the LFP in a broad frequency range were decreased. In contrast, injections of CNO, the agonist of the excitatory DREADD, increased pyramidal MUA but did not alter eSPW amplitude nor LFP power.

At P7 the effects were in line with the traditional inhibitory effect of GABA: inhibiting GABAergic neuron activity via KORD-DREADD agonist increased MUA in the GABAergic-receiving pyramidal layer of the hippocampus. LFP power in 6-14 Hz frequencies also increased; however there was no alteration of eSPW occurrence, duration, or amplitude of eSPWs. Enhancing GABAergic neuron activity via the hM3Dq-DREADD agonist decreased LFP power in a broad range and decreased pyramidal MUA, but did not significantly affect eSPW. At P11 modulating GABAergic neuronal activity had similar effects on MUA activity and LFP power.

In contrast to hippocampus, administration of inhibitory DREADD agonists at P3 increased MUA, while administration of the excitatory DREADD agonist decreased MUA, suggesting that there was a net inhibitory affect of enhancing GABAergic activity in the cortex. LFP power or eSPW activity was not significantly affected by either modulation. Modulation at P7 and P11 had similar effects.

Finally, they investigate the effect of modulating both cortical and hippocampal GABAergic neurons simultaneously on hippocampal CA1 neurons. This is examined because there are substantial cortical projections to CA1 at this age. Murata and Colonnese found that cortical inhibitory GABAergic input overwhelms the local excitatory GABAergic input in hippocampus – administration of inhibitory DREADD agonist has a net excitatory effect, while administration of the excitatory DREADD agonist has a net inhibitory effect.

Why is it important?

The results demonstrate that the picture of developing inhibitory and excitatory circuitry is more complex than first thought. Understanding changes in the developing neural circuitry is key to understanding how the circuit is shaped by input and intrinsic activity. Demonstrating heterogeneity in developmental changes in the effects of major neurotransmitter groups demonstrates how important it is to examine a wide range of model systems, and not use one circuit (e.g. synapses onto hippocampal CA1 neurons) as a generally representative model. Considering developmental changes in circuitry is particularly important in brain slice electrophysiology, where tissue from younger animals is often used to make it easier to patch.

Questions arising

An interesting point arising from this paper, which seems to be unaddressed, is that the Murata and Colonnese found opposing effects in hippocampal neurons from activating different populations of GABAergic inputs (i.e. different presynaptic neurons), but the change in the effect of GABA from excitatory to inhibitory has been linked with changes in the postsynaptic neuron: expression of chloride transporters such as KCC2 (potassium chloride transporter member 5). KCC2 is responsible for reducing chloride concentrations and so as its expression increases through development there is greater inward GABA-mediated chloride influx. If KCC2 is connected to the regional differences in the change of the effect of GABA from excitatory to inhibitory, then we might expect the effects of GABA to be consistent within a given region regardless of the source of input.

Tags: development, dreadd, gaba, in vivo, mouse

doi: https://doi.org/10.1242/prelights.12141

Read preprint (No Ratings Yet)