GABA and glutamate: the Yin and Yang of fragile X

Comment on: Braat S, et al. The GABAA Receptor is an FMRP Target with Therapeutic Potential in Fragile X Syndrome. Cell Cycle 2015; PMID: 25790165

One of the more popular concepts guiding autism research in recent years is the notion of a crucial balance between excitatory and inhibitory neurotransmission, often called the “E/I balance” or “E/I ratio,” first proposed in 2003.¹ Essentially, this model proposes that a defining feature of many autism spectrum disorders is hyperexcitability in critical brain systems, and that this could come about in many different ways. Conditions causing excessive excitatory neurotransmission could yield clinical phenotypes similar to other disorders causing decreased inhibitory neurotransmission—balance is the key. Thus, a large number of genetically and etiologically heterogeneous disorders could share a kind of final common pathway, and they could potentially respond to similar treatment strategies.

This concept certainly resonates in the study of fragile X syndrome, a monogenic disorder with a striking hyperexcitability phenotype. Fragile X patients display hyperactivity, tactile defensiveness, multi-sensory hypersensitivity, a wide range of anxiety disorders, mood lability, aggression, and seizures in 20–30% of males.² Animal models of fragile X recapitulate many of the human phenotypes, and electrophysiologic studies confirm hyperexcitability at the synaptic and circuit level.³ Much of the focus to date has been on glutamatergic dysfunction in fragile X, and how this might render the fragile X brain hyperexcitable. However, clinical trials of glutamatergic agents in human subjects with fragile X have been disappointing, leaving an unmet need for fragile X therapeutics.

In this issue, Braat et al. contribute valuable new insights into the other side of this dynamic balance.⁴ Here they confirm previous findings of decreased expression of GABA_A receptor subunit mRNAs, and they further demonstrate interactions between these messages and the Fragile X Mental Retardation Protein (FMRP). Their data strongly suggest that FMRP normally acts as a chaperone for GABA_A mRNAs, protecting them from degradation. While there is precedence for this, as cited in the case of PSD-95, most previous studies of FMRP's RNA binding activity have emphasized its role in the regulation of activity-dependent protein synthesis. Proteomic studies in fragile X have consistently shown that many protein species are produced excessively compared to WT controls, but there are some potentially important species which are underexpressed.⁵ This has often been interpreted as an artifact of competition for the translational machinery, but in light of these data, it appears that this may be a distinct disease mechanism in fragile X, and this may be a direct cause of GABAergic dysfunction. Importantly, significant differences were found between brain regions.

These findings lend themselves to rapid translation into improved therapeutics for fragile X and other autism spectrum disorders, but GABA_A agonists (or PAMs) have a checkered history in clinical use for developmental disorders. Chronic treatment with benzodiazepines, for example, has generally been avoided because of the risk of cognitive impairment and paradoxical behavioral effects. Newer classes of GABA_A agonists, including neurosteroids and subtype-specific PAMs, offer the promise of “re-balancing” neurotransmission in autism spectrum disorders without these liabilities. Braat et al. demonstrate significant rescue of relevant fragile X mouse phenotypes with the neurosteroid ganaxolone; equally importantly, they also show a limited risk of sedation and motor impairment. These findings bode well for clinical trials of ganaxolone currently under way in fragile X subjects, though they are certainly no guarantee of clinical success.

Other recent studies have suggested other distinct pathologies in fragile X inhibitory systems. Gibson et al. have demonstrated network hyperexcitability in sensory cortex which is attributable to decreased excitatory drive onto inhibitory interneurons.⁶ In this brain region, their electrophysiologic studies showed no evidence of abnormalities at GABAergic synapses per se, or of increased excitatory transmission, but rather deficient feedback inhibition in specific cell types. He et al. have found that the normal developmental switch from excitatory to inhibitory neurotransmission at GABAergic synapses is delayed and abnormally regulated in fragile X mice.⁷ How persistent these changes are later life in humans with fragile X remains unknown, but this type of polarity reversal could wreak havoc with the formation of inhibitory networks, and could complicate pharmacotherapy with GABAergic agents. Nonetheless, interest is clearly shifting to GABA_A as a treatment target for fragile X.