Mom Genes: A Role for Loss of Maternal Ube3a in GABAergic Neurons in Angelman Syndrome

GABAergic Neuron-Specific Loss of Ube3a Causes Angelman Syndrome-Like EEG Abnormalities and Enhances Seizure Susceptibility

Judson MC, Wallace ML, Sidorov MS, Burette AC, Gu B, van Woerden GM, King IF, Han JE, Zylka MJ, Elgersma Y, Weinberg RJ, Philpot BD. Neuron 2016;90:56–69.

Loss of maternal UBE3A causes Angelman syndrome (AS), a neurodevelopmental disorder associated with severe epilepsy. We previously implicated GABAergic deficits onto layer (L) 2/3 pyramidal neurons in the pathogenesis of neocortical hyperexcitability, and perhaps epilepsy, in AS model mice. Here we investigate consequences of selective Ube3a loss from either GABAergic or glutamatergic neurons, focusing on the development of hyperexcitability within L2/3 neocortex and in broader circuit and behavioral contexts. We find that GABAergic Ube3a loss causes AS-like increases in neocortical EEG delta power, enhances seizure susceptibility, and leads to presynaptic accumulation of clathrin-coated vesicles (CCVs)-all without decreasing GABAergic inhibition onto L2/3 pyramidal neurons. Conversely, glutamatergic Ube3a loss fails to yield EEG abnormalities, seizures, or associated CCV phenotypes, despite impairing tonic inhibition onto L2/3 pyramidal neurons. These results substantiate GABAergic Ube3a loss as the principal cause of circuit hyperexcitability in AS mice, lending insight into ictogenic mechanisms in AS.

Commentary

Loss of UBE3A is a major genetic cause of Angelman Syndrome (AS), a neurodevelopmental disorder characterized by severe seizures, movement disorders, and cognitive delays. Because the paternal UBE3A allele is silenced in neurons, the maternal allele is the primary source of UBE3A expression in the brain. Consequently, silencing the maternal allele drives the development of AS in humans (1) and produces AS-like features in animal models. In mice with loss of the maternal Ube3a allele, inhibitory GABAergic input to excitatory pyramidal neurons in layer 2/3 of primary visual cortex (V1) was found to be substantially reduced (2). This same mouse line had previously been observed to demonstrate strong EEG abnormalities and heightened susceptibility to audiogenic seizures (3). Therefore, a reasonable conclusion was that loss of Ube3a in GABAergic neurons, and downstream deficits in GABA release, is a likely factor driving AS-associated epilepsy. This study, however, demonstrates that the story is not quite so clear-cut.

Cre-lox recombination is a powerful genetic tool that enables selective deletion of genes of interest in certain cell types. To delete Ube3a from GABAergic neurons, the authors crossed a Ube3a “floxed” mouse, in which the Ube3a gene is “flanked by two LoxP sites,” with a second line expressing Cre recombinase in GABAergic neurons. The pups of this cross thus would have Ube3a deleted only from GABAergic neurons. These mice showed increased susceptibility to audiogenic and flurothyl-induced seizures, suggesting that impaired GABAergic neuron function may underlie these phenotypes. Surprisingly, however, inhibitory transmission to layer 2/3 pyramidal neurons in V1 was unchanged, in contrast to the previous findings in mice with complete loss of the maternal Ube3a allele.

The authors postulated that the ability of excitatory layer 2/3 pyramidal neurons to receive inhibitory GABAergic input was perhaps impaired. To test this possibility, they crossed the Ube3a floxed mouse with a line expressing Cre in the excitatory pyramidal cells. In these mice, GABAergic currents recorded in the presence of tetrodotoxin, which blocks action potential firing, were not affected. This finding suggests that GABA_A receptors located within the synaptic cleft are not affected by the loss of Ube3a in the postsynaptic pyramidal cell. However, the amplitude of GABAergic currents evoked by electrical stimulation was reduced. Because electrical stimulation drives stronger release of GABA, which can spill over and bind to receptors located outside of the synaptic cleft, the authors next tested whether tonic inhibition, which primarily results from binding of GABA to extrasynaptic GABA_A receptors, was affected. Indeed, the size of tonic GABA currents was reduced. Because alterations in tonic inhibition have been associated with changing seizure susceptibility in various epilepsy models (4, 5), the next test was to determine whether these mice had altered seizure susceptibility. Loss of Ube3a only in pyramidal cells, however, did not alter susceptibility to flurothyl induction of seizures.

To determine whether a more widespread loss of Ube3a in excitatory cells (i.e., those in ventral brain areas and hippocampal dentate gyrus) was necessary to drive altered seizure susceptibility, mice expressing Cre in GABAergic neurons were then crossed to mice in which Cre would turn on expression of Ube3a in Cre-expressing cells. In this way, Ube3a expression could be restored in GABAergic neurons, but not glutamatergic or other neuron populations. These mice showed a complete lack of audiogenic seizure induction, indicating that expression of Ube3a in GABAergic neurons is sufficient to restore seizure resistance. Similarly, mice with loss of Ube3a in GABAergic neurons showed increased activity of local field potentials in V1, particularly in the delta band (δ, 2–4 Hz), and this effect could be reversed by genetic restoration of Ube3a in GABAergic neurons. Increased EEG activity at this frequency has been observed in AS patients (6). Therefore, selective loss of Ube3a in GABAergic neurons in mice recapitulates some of the EEG abnormalities of AS.

Lastly, the authors examined the degree of accumulation of clathrin-coated vesicles (CCVs) in presynaptic GABAergic terminals in V1. CCV accumulation may reflect impaired vesicle recycling, particularly at times of high rates of neurotransmitter release. Somewhat unexpectedly, given that inhibitory transmission appeared intact in mice without Ube3a in GABAergic neurons, CCV accumulation was noted in these mice, but not in mice with loss of Ube3a in other neuronal populations. Even more remarkably, CCV accumulation was also noted in glutamatergic terminals of mice with GABAergic neuron loss of Ube3a. Although basal and electrically evoked synaptic inhibition in V1 appears largely unaffected by loss of Ube3a in GABAergic neurons, the CCV accumulation might signal summative effects of elevated circuit hyperexcitability, as demonstrated in the higher delta EEG power. It should be noted that the frequency of electrical stimulation used to “deplete” GABAergic transmission (30 Hz) in these experiments was significantly lower than the rates of firing achieved by fast spiking interneurons (>200 Hz), the major type of cortical GABAergic interneuron. Therefore, it would be interesting to reexamine whether GABAergic transmission in V1 is truly intact in these mice, using stimulation rates that better approximate the fast patterns of activity of these interneurons, and examining inhibition in other cortical layers besides 2/3.

A key consideration is that these transgenic mouse models do not confine deletion/restoration of Ube3a in GABAergic neurons to V1. Therefore, another question is whether GABAergic transmission in other brain areas besides V1 is affected, and whether this is linked to the increased seizure susceptibility and EEG abnormalities observed. In this regard, the authors highlight the thalamic reticular nucleus (TRN), a GABAergic structure that plays a major role in thalamocortical synchrony and activity in both normal and pathological states, including seizures (7). Of interest, mice that lack the β3 subunit of GABA_A receptors harbor impaired GABAergic transmission in the TRN (8), and show seizure phenotypes and other clinical features reminiscent of AS (9, 10). Without clarifying effects of GABAergic neuron-specific loss of Ube3a in other brain areas besides V1, it remains premature to conclude that increased seizures and EEG abnormalities are not linked with impaired GABAergic synaptic transmission somewhere else in the brain.

It is important to remember, however, that pan-neuronal deletion of Ube3a produced deficits in GABAergic synaptic transmission in V1 (2). Therefore, it would be interesting to use viral vectors to selectively delete/restore Ube3a expression in specific brain areas, and to identify regions in which changes in the level of Ube3a in GABAergic neurons are linked to alterations in seizure susceptibility and EEG activity. Nevertheless, together these studies indicate that the level of Ube3a expression in GABAergic neurons is a critical factor in the development of AS-like seizure phenotypes in mice, and thus provide a strong rationale for focusing on GABAergic neurons in the future development of AS therapies.