One of the key problems in mice that model Down’s syndrome is excessive inhibition of neuronal circuits due to the overabundance of GABAergic interneurons in the forebrain. However, a paper in the March 16 Nature Medicine offers a counter view—that one type of GABA receptor, GABAA, becomes excitatory. Mice that model the disease overproduce a chloride transporter, called NKCC1, according to scientists led by Laura Cancedda and Andrea Contestabile at the Istituto Italiano di Tecnologia, in Genova, Italy. NKCC1 raises the concentration of chloride ions in the neuron so that they flow out, rather than in, when GABAA channels open. This depolarizes the cell, making an action potential more likely. The researchers found that an NKCC1 inhibitor called bumetanide restored normal chloride concentrations, inhibitory signaling, and cognitive function in the mice. Since the drug is FDA-approved as a diuretic, it could be rapidly tested as a treatment approach for clinical trials in Down’s, wrote the authors.

“This study describes several potentially interesting observations and, if confirmed, will have great impact on the field of neurobiology of Down’s syndrome,” wrote William Mobley and Alexander Kleschevnikov from the University of California, San Diego.

Previous studies have looked in the brain of the Ts65Dn mouse to uncover the underlying reason for cognitive deficits in Down’s syndrome (DS). Because the animals have more GABAergic interneurons and GABA is typically inhibitory, scientists presumed that this meant more inhibition for the DS brain (Hernández-González et al., 2015Chakrabarti et al., 2010). In keeping with that idea, GABAA antagonists and inverse agonists rescue long-term potentiation and cognitive impairments in Ts65Dn mice. Roche currently has a GABAA inverse agonist, RG1662, in a Phase 2 trial in adolescents with DS. However, no one had previously tested whether GABAergic inputs led to inhibition in these mice, said Cancedda.

Co-first authors Gabriele Deidda, Martina Parrini, and Shovan Naskar compared patch-clamp recordings from CA1 pyramidal neurons in hippocampal slices of wild-type and Ts65Dn animals. When the researchers bathed wild-type slices in 100 to 200 μM GABA, spike frequency fell, as expected for an inhibitory neurotransmitter. Cancedda said GABA can reach 200 μM in synapses. However, in hippocampal slices from Ts65Dn mice, GABA brought on more action potentials. The same held true in neurons from the mouse neocortex. Endogenous GABA seemed to have a similar effect. Compared with cells from wild-type mice, CA1 neurons from Ts65Dn animals were twice as active, and a GABA receptor antagonist called bicuculline toned them down. Could GABA be excitatory in those cells? Patch-clamp recordings confirmed that GABA hyperpolarized wild-type, but depolarized Ts65Dn neurons. This suggested that chloride ions flow out of the latter upon GABA stimulation. A chloride-sensitive dye confirmed that Ts65Dn neurons had a higher resting chloride concentration than wild-type cells.

What caused this change? The propensity for inward or outward flow of Cl- upon GABAA activation depends in part on the transporters NKCC1 and KCC2, which pump chloride ions in or out, respectively. The researchers found more NKCC1 than usual in the hippocampi of Ts65Dn mice and in adults with DS. By inhibiting this transporter with bumetanide, the scientists returned the intracellular concentration of chloride ions to normal, lowered the frequency of spontaneous spikes, and halted GABA-induced excitation in Ts65Dn hippocampal neurons. Bumetanide also rescued long-term potentiation—essential for learning and memory—in hippocampal slices. It had no effect on neurons or slices from wild-type mice.

To see if bumetanide could stave off memory deficits, Deidda and colleagues treated Ts65Dn mice with the drug for either one or four weeks. Both treatments restored normal performance in a contextual fear-conditioning test, an object-location test, and a novel-object recognition task. The benefits disappeared when the drug washed out. A week after treatment stopped, Ts65Dn mice performed as poorly on all of the tests as their untreated littermates. Hippocampal slices from these mice also reverted to their lower levels of LTP.

“This is an elegant series of experiments,” wrote Michael Rafii, University of California, San Diego (see full comment below). “They further support the idea that the GABAergic system may be readily amenable to pharmacological intervention to improve cognition.”

This is not the first time that signals from GABAA receptors have been found to be excitatory. Mouse models of autism and fragile X syndrome also display the phenotype (Tyzio et al., 2014; He et al., 2014). A clinical trial found that bumetanide lessened autism severity in children and improved their ability to recognize facial expressions (Lemmonier et al., 2012). Some trials are testing bumetanide for neonatal seizures. Cancedda and Contestabile are collaborating with researchers at the Bambino Gesù Children's Hospital in Rome on a pilot trial in people with DS. They plan to test whether three months of bumetanide improves cognitive functions, long-term memory, or psychopathology in 60 young volunteers aged 10 to 24. “We might have found a drug that can reverse at least partially some of the cognitive deficits in Down’s syndrome,” Contestabile said. The researchers noted that their work does not close the door on GABA antagonists or reverse agonists, but suggests an alternative explanation for their efficacy—they would block GABAergic excitatory signals.

“A new door has opened in the treatment of Down’s syndrome,” said Ahmad Salehi, Stanford University School of Medicine, Palo Alto, California, who was not involved in the work. Repurposing bumetanide for DS would cut down on the cost and time needed to find a successful treatment, he said, adding that the behavioral data in this paper are compelling. If successful, a bumetanide trial could have implications for Alzheimer's, he told Alzforum, because drugs that work for DS have a good chance of working for AD. The mechanism of excitatory GABAA receptor signaling is surprising but intriguing, he said, and will need to be replicated by independent lab groups. The next step would be to identify the gene responsible for the changes in NKCC1, in hopes of finding a drug that reduces the expression of that gene, he said. An extra copy of chromosome 21 causes DS, but NKCC1 is located on a different chromosome. Salehi and Cancedda agreed that one of the triplicated genes responsible for the disorder might affect NKCC1 expression.

Marie-Claude Potier, Salpêtrière Hospital, Paris, said this original finding was important. “It provides a new view on what happens in the balance between excitation and inhibition in the brain in Down’s syndrome,” she told Alzforum. She suggested that it would be interesting to determine whether the uptick in NKCC1 occurs from birth. For the treatment of young children, that would be important to know, she said.—Gwyneth Dickey Zakaib


  1. This is an elegant series of experiments, whereby Deidda et al. have identified the cation chloride co-transporter as a potential therapeutic target in Down’s syndrome (DS). Intriguingly, they have discovered that GABA receptor signaling is excitatory rather that inhibitory in the Ts65Dn mouse, and that modulation with an FDA-approved compound restores normal synaptic function and improves cognition. Knowing that dysfunction in the GABAergic system underlies certain aspects of intellectual disability in DS, this further supports the idea that the GABAergic system may be readily amenable to pharmacological intervention to improve cognition in DS.

    View all comments by Michael Rafii

Make a Comment

To make a comment you must login or register.


Research Models Citations

  1. Ts65Dn

Paper Citations

  1. . Altered distribution of hippocampal interneurons in the murine Down Syndrome model Ts65Dn. Neurochem Res. 2015 Jan;40(1):151-64. Epub 2014 Nov 16 PubMed.
  2. . Olig1 and Olig2 triplication causes developmental brain defects in Down syndrome. Nat Neurosci. 2010 Aug;13(8):927-34. PubMed.
  3. . Oxytocin-mediated GABA inhibition during delivery attenuates autism pathogenesis in rodent offspring. Science. 2014 Feb 7;343(6171):675-9. PubMed.
  4. . The developmental switch in GABA polarity is delayed in fragile X mice. J Neurosci. 2014 Jan 8;34(2):446-50. PubMed.
  5. . A randomised controlled trial of bumetanide in the treatment of autism in children. Transl Psychiatry. 2012 Dec 11;2:e202. PubMed.

External Citations

  1. Phase 2
  2. clinical trial
  3. trials 

Further Reading


  1. . Treating enhanced GABAergic inhibition in Down syndrome: use of GABA α5-selective inverse agonists. Neurosci Biobehav Rev. 2014 Oct;46 Pt 2:218-27. Epub 2014 Jan 9 PubMed.
  2. . GABAA receptor subtypes: Therapeutic potential in Down syndrome, affective disorders, schizophrenia, and autism. Annu Rev Pharmacol Toxicol. 2014;54:483-507. Epub 2013 Oct 23 PubMed.
  3. . Reducing GABAergic inhibition restores cognitive functions in a mouse model of Down syndrome. CNS Neurol Disord Drug Targets. 2014 Feb;13(1):8-15. PubMed.
  4. . Cognitive enhancement by pharmacological and behavioral interventions: the murine Down syndrome model. Biochem Pharmacol. 2012 Oct 15;84(8):994-9. Epub 2012 Aug 8 PubMed.

Primary Papers

  1. . Reversing excitatory GABAAR signaling restores synaptic plasticity and memory in a mouse model of Down syndrome. Nat Med. 2015 Mar 16; PubMed.