GABA and glycine sometimes are interchangeable, by Jaime de Juan Sanz

The sensory environment contains multiple streams of information that must be efficiently processed and categorized by the CNS. Sound, for example, is transformed into neural signals in the cochlea, which are then transmitted and processed in the auditory brainstem, an ensemble of nuclei belonging to the efferent and afferent auditory systems. One particular region of the auditory brainstem, the central nucleus of the inferior colliculus (ICC), is known to integrate most ascending auditory information from lower brainstem regions, receiving inputs from several nuclei, including a prominent long-range inhibitory input from the ventral nucleus of the lateral lemniscus (VNLL). The ICC acts as an integrative station that is involved in the routing of multi-modal sensory perception, controlling for example the startle response and the vestibulo-ocular reflex. The startle response is finely modulated by inhibitory inputs from the VNLL, which are classically considered to be glycinergic (Ono and Ito, 2015), and it is known that global lack of glycinergic neurotransmission in humans causes Startle Disease or Hyperekplexia (OMIM 149400), a rare neurological disorder characterized by an exaggerated startle reflex in response to acoustic or tactile stimuli, among other symptoms. Despite the role of glycinergic release from VNLL neurons is known, histological evidence suggests that these neurons may corelease glycine and GABA in the ICC, which could be relevant for startle disease. However, functional evidence for glycine/GABA corelease has not been shown experimentally in this pathway, in part because ascending fibers from lower brainstem regions contain multiple sources of pure GABAergic inputs and electrical stimulation in these regions would activate these inputs, making it unfeasible to measure corelease from VNLL glycinergic fibers in isolation.

In a recent study published in The Journal of Neuroscience, Moore and Trussell (2017) circumvented this limitation using optogenetics, which allowed them to control and selectively manipulate glycinergic neurons without co-activating surrounding ascending GABAergic axons. Glycinergic neurons uniquely express a presynaptic glycine transporter, GlyT2, which recycles glycine to refill presynaptic vesicles (de Juan-Sanz et al., 2013). Moore and Trussell used a GlyT2-Cre mouse line to express Channelrhodopsin-2 (ChR2) in glycinergic neurons in a floxed/reversed (Cre-dependent) manner, either globally or specifically within the VNLL. Using these two approaches, they photostimulated ascending glycinergic fibers and recorded light-evoked IPSCs from single neurons in the ICC using electrophysiology. These experiments easily identified inhibitory currents, which were further dissected using pharmacology against glycine or GABA receptors to understand whether GABA is coreleased with glycine, as suggested by histological data. These experiments unequivocally demonstrated that GABA is coreleased with glycine in VNLL inputs and revealed an average GABAergic component of ~40% of the light-evoked ICC currents.

Why GABA/glycine corelease? Do they have different functions?

Optogenetics allowed the authors to explore the characteristics of glycinergic and GABAergic components separately to probe for possible functional differences in the signaling of these coreleased neurotransmitters. These experiments are particularly interesting for the field, as there is currently limited understanding of the physiological roles of corelease in the same synapse, especially in the auditory midbrain. In the CNS GABA is coreleased with a variety of other neurotransmitters (Tritsch et al., 2016), but GABA/glycine corelease is one of the most paradoxical examples as both are fast inhibitory neurotransmitters. Many functions of GABA/glycine corelease have been proposed, including subserving tonic versus phasic inhibition via different time courses of glycinergic and GABAergic synaptic currents, extra-fast inhibition, and compensation by one neurotransmitter during periods of sustained activity. To better understand the unique functions of each neurotransmitter in the VNLL-ICC synapse, the authors explored the responses in the ICC to glycine or GABA released from VNLL neurons, specifically measuring differences in kinetics, contacts with cell types in the ICC, short-term depression dynamics and effects of GABA on presynaptic glycine release.

Kinetics of glycine and GABA are typically different, as it is known that GABAergic currents normally decay more slowly than glycinergic currents (Russier et al., 2002). Thus, glycine and GABA could serve different functions through differing postsynaptic kinetics. Surprisingly, the authors found that in the ICC, GABAergic currents were nearly as fast as glycinergic currents (3.7 vs 3.0 ms respectively), indicating that in this synapse slow depolarizing GABAergic IPSCs do not occur as observed in other situations. Cellular subtypes in the ICC express mainly the α1-GABA receptor subunit that endows GABA receptors with fast kinetics, which likely accounts for the observed phenotype. The authors then explored whether GABA and glycine present unique roles in short-term synaptic depression, as this high frequency stimulation could result in the selective depletion of one transmitter over the other. Similarly, no difference was found, as the relative contribution of glycine and GABA to the postsynaptic current was not dependent on the length and frequency of stimulation. Taken together, these results demonstrated that coreleased glycinergic and GABAergic currents in the ICC had strikingly similar features, not supporting popular hypotheses for the different functions of GABA and glycine during corelease and suggesting that the two neurotransmitters may function interchangeably.

What is the relevance of these discoveries?

These results are particularly relevant to our molecular understanding of startle disease and how the medical community is treating this dysfunction. It is known that the startle response is finely tuned by inhibitory inputs from the VNLL, classically considered glycinergic. However, results from Moore and Trussell clearly demonstrate that a GABAergic component contributes to VNLL inhibition in the ICC, and their detailed studies of glycine and GABA responses in this pathway suggest that the two neurotransmitters may be functionally interchangeable. This suggests symptoms of the exaggerated startle response in hyperekplexia might be reduced by compensating for the lack of glycinergic neurotransmission with increased GABAergic neurotransmission. In fact, currently the most effective treatment for startle disease is the administration of GABA receptor agonists, such as Clonazepam, which significantly reduces startle activity in hyperekplexia patients (Zhou et al., 2002). Moore and Trussell’s results offer new details that should help to understand the molecular basis of this pharmacological compensation, helping the medical community to better exploit this pathway for future benefit. In addition, this work reveals a new role for glycine/GABA corelease in the CNS, where the action of both neurotransmitters appears interchangeable. Independent receptor modulation coupled with the highly similar features of coreleased GABA and glycine observed in the study by Moore and Trussell may further permit the neurotransmitters to compensate for one another under conditions where one system is compromised, maintaining homeostasis of inhibitory networks integrated in the ICC.

References:

1. de Juan-Sanz J, Núñez E, López-Corcuera B, Aragon C (2013) Regulation of the glycinergic neurotransmission during inflammatory pain: A new pathway in the action of Prostaglandin E2 in the spinal cord. An la Real Acad Nac Farm 79.
2. Moore LA, Trussell LO (2017) Corelease of Inhibitory Neurotransmitters in the Mouse Auditory Midbrain. J Neurosci 37:9453–9464.
3. Ono M, Ito T (2015) Functional organization of the mammalian auditory midbrain. J Physiol Sci 65:499–506.
4. Russier M, Kopysova IL, Ankri N, Ferrand N, Debanne D (2002) GABA and glycine co-release optimizes functional inhibition in rat brainstem motoneurons in vitro. J Physiol 541:123–137.
5. Tritsch NX, Granger AJ, Sabatini BL (2016) Mechanisms and functions of GABA co-release. Nat Rev Neurosci 17:139–145.
6. Zhou L, Chillag KL, Nigro MA (2002) Hyperekplexia: a treatable neurogenetic disease. Brain Dev 24:669–674.

Image credit: Flickr

Any views expressed are those of the author, and do not necessarily reflect those of PLOS. 

Jaime de Juan-Sanz holds an M.Sc. in Molecular Biomedicine from Universidad Autónoma de Madrid and a Ph.D. in Molecular Neurobiology obtained at Centro de Biología Molecular “Severo Ochoa” (CBMSO). He is currently a postdoctoral associate at Weill Cornell Medical College (Cornell University) in New York City where he is investigating the molecular basis of synaptic transmission, studying biophysical aspects of synaptic biology. He is also interested in encouraging open science in our society promoting open access research and science outreach. You can follow him on twitter @JaimedeJuan