The cornerstones of research on postsynaptic inhibition were laid by Charles Sherrington, who deduced its hyperpolarizing nature and, notably, emphasized “Inhibition as a coordinative factor” in the CNS (Nobel lecture,1932). The prescient ideas of Sherrington have not only materialized in work on neuronal information processing in the mature CNS: it has become obvious that major milestones of brain development are associated with remarkable qualitative changes in the responses generated by postsynaptic GABA_A receptors. These, in turn, are attributable to coordinated alterations in the spatiotemporal expression patterns of plasmalemmal chloride transporters – such as the Na-K-2Cl cotransporter NKCC1 and the neuron-specific K-Cl cotransporter, KCC2 – and other ion-regulatory proteins, including carbonic anhydrases (CAs), whereof isoform CA7 is neuron-specific.             

The default state of mammalian cells, including immature CNS neurons, is a high NKCC1-dependent intracellular Cl^– concentration ([Cl^–]_i) which leads to a depolarizing action of GABA. Thus, the adult neurons with their low [Cl^–]_i and hyperpolarizing GABA responses are an “aberrant” type of cells. As shown in our early work, upregulation of KCC2 expression accounts for the developmental hyperpolarizing shift in GABA action. This shift follows the distinct time courses of maturation in various neuronal populations and animal species, with striking differences in GABA actions during the perinatal period between altricial and precocial mammals. Current data indicate that in the full-term healthy human baby, GABA is hyperpolarizing in the cortex and, by implication, elsewhere in the brain.  

A subsequent developmental shift or “switch” is caused by the abrupt emergence of the neuron-specific CA isoform 7 (CA7) at a developmental stage by which KCC2 has reached a near-maximum functionality. GABA_ARs have a significant permeability to HCO₃^–, and because of its rather positive equilibrium (at around -10 mV), the HCO₃^– current component is depolarizing. In the cortex, intense activity of GABAergic interneurons results in fast collapse of the chloride gradient in principal neurons, giving rise to depolarizing and excitatory GABA actions. Indeed, selective interneuronal stimulation in the healthy brain can promote seizure activity, accentuated by an increase in extracellular [K⁺].

Soon after identifying the causal role of KCC2 in the ontogeny of hyperpolarizing postsynaptic GABA_AR responses, we found that kindling-induced seizures in adult mice led, within a few hours, to down-regulation of KCC2. This observation has since been replicated in virtually all models of cortico-hippocampal neuronal trauma (e.g. stroke, mechanical damage, and neuroinflammation). KCC2 down-regulation, paralleled by neuronal NKCC1 upregulation, may well be one aspect of neuronal dedifferentiation required for re-wiring of functional circuits in CNS disorders. Maintaining a low [Cl^–]_i imposes a high burden on neuronal energy metabolism, especially during an energy crisis. Therefore, it is not immediately clear whether the depolarizing GABA responses have a disease-promoting (maladaptive) or an adaptive/compensatory role. NKCC1-dependent depolarizing GABA signaling is likely to promote interictal activity – but not seizures – in the epileptic brain. The presence of NKCC1 in virtually all kinds of cells within and outside the brain has led to lots of confusion regarding the pharmacological actions of bumetanide and other NKCC1 blockers, especially in vivo.

In addition to their roles in modulating the efficacy of GABAergic inhibition, some ion-regulatory proteins act as morphogenic factors. For instance, the large C-terminus of KCC2 interacts with cytoskeletal elements influencing dendritic spinogenesis in cortical neurons, suggesting a role for KCC2 in coordinating the development of GABAergic and glutamatergic synapses. Such cytoskeletal effects are also likely to affect the neurological phenotype of disease mutations of KCC2, which is known to have a very high genic intolerance.