Abstract
LGI1 Acts Presynaptically to Regulate Excitatory Synaptic Transmission During Early Postnatal Development.
Boillot M, Lee CY, Allene C, Leguern E, Baulac S, Rouach N. Sci Rep 2016;6:21769.
The secreted leucine-rich glioma inactivated 1 (LGI1) protein is an important actor for human seizures of both genetic and autoimmune etiology: mutations in LGI1 cause inherited temporal lobe epilepsy, while LGI1 is involved in antibody-mediated encephalitis. Remarkably, Lgi1-deficient (Lgi1−/−) mice recapitulate the epileptic disorder and display early-onset spontaneous seizures. To understand how Lgi1-deficiency leads to seizures during postnatal development, we here investigated the early functional and structural defects occurring before seizure onset in Lgi1−/− mice. We found an increased excitatory synaptic transmission in hippocampal slices from Lgi1−/− mice. No structural alteration in the morphology of pyramidal cell dendrites and synapses was observed at this stage, indicating that Lgi1- deficiency is unlikely to trigger early developmental abnormalities. Consistent with the presynaptic subcellular localization of the protein, Lgi1-deficiency caused presynaptic defects, with no alteration in postsynaptic AMPA receptor activity in Lgi1−/− pyramidal cells before seizure onset. Presynaptic dysfunction led to increased synaptic glutamate levels, which were associated with hyperexcitable neuronal networks. Altogether, these data show that Lgi1 acts presynaptically as a negative modulator of excitatory synaptic transmission during early postnatal development. We therefore here reveal that increased presynaptic glutamate release is a key early event resulting from Lgi1-deficiency, which likely contributes to epileptogenesis.
Commentary
In 2002, when two groups first reported that variants in the secreted leucine-rich glioma inactivated 1 (LGI1) gene cause autosomal-dominant partial epilepsy with auditory features (ADPEAF), both heralded the discovery as the first example of a gene causing idiopathic epilepsy that was not an ion channel–encoding gene (1, 2). In contrast to voltage- and ligand-gated ion channels, LGI1 is a secreted protein, presumably without a direct role in controlling either neuronal membrane excitability or synaptic transmission, and the discovery of its link to ADPEAF led to speculation about novel pathogenic mechanisms of epileptogenesis.
Soon, however, evidence began to accumulate that the mechanisms through which LGI1 dysfunction leads to epilepsy may not be so far removed from ion channels after all. This evidence came from two studies published in 2006; one showed that LGI1 interacts with a transsynaptic protein complex that regulates ligand-gated AMPA-type glutamate receptor trafficking, and another showed that it interacts with presynaptic potassium channels and regulates their inactivation kinetics, which could potentially affect neurotransmitter release (3, 4). Since then, LGI1 knockout mouse have been created and homozygous deletion shown to cause lethal seizures beginning around postnatal day 10 (P10) to P14 (5–7). Attempts to characterize changes in neurotransmission in these mouse models have proven confounding, as some studies found increased excitatory synaptic strength and others decreased excitatory synaptic strength.
In the present study, Boillot et al. wanted to further investigate potential synaptic changes in these mice that could lead to epilepsy, and speculated that at least one of the reasons for the conflicting data from previous studies was that the electrophysiological characterizations were done on brain slices taken from mice that were already having seizures, making it impossible to determine whether the observed changes were due to loss of LGI1 and contribute to the development of epilepsy, or whether they resulted from the seizures. To do this they performed their experiments on hippocampal slices from P8 to P9 mice, an age before seizure onset. They started by recording miniature (m)EPSCs in CA1 pyramidal cells, the size of which reflect the effect that release of a single synaptic vesicle filled with glutamate has on the postsynaptic membrane. In a finding that differed from what had previously been reported, they found that the peak amplitude and charge of these mEPSCs were increased in slices from Lgi1−/− mice, but not the frequency. In agreement with the unchanged frequency, morphological analysis of excitatory synapses in the CA1 region showed that the synapse density and number of docked synaptic vesicles per synapse were also unchanged.
Usually, increases in mEPSC size are interpreted to indicate an increase in the number or type of AMPA receptors in the postsynaptic membrane. To test this the authors performed an electrophysiological assay comparing the ratio of the AMPA receptor-mediated current with the NMDA receptor–mediated current, but did not find the increase that would be expected if this were the case. They also performed Western blot analysis of AMPA receptor subunit protein levels and again found no difference, suggesting that the increase in the mEPSC size was not due to changes in the way glutamate is sensed by the postsynaptic compartment.
Although less common, presynaptic mechanisms can also account for changes in mEPSC size, for example, if each synaptic vesicle contains more glutamate, or if the site of vesicle release is closer to the postsynaptic receptors, both of which will increase the concentration of glutamate that the AMPA receptors encounter. While very difficult to directly measure, the relative concentration of glutamate that the AMPA receptors encounter can be determined by measuring the extent to which a low-affinity competitive AMPA receptor antagonist inhibits evoked EPSCs. The authors found that this antagonist was less effective at inhibiting evoked EPSCs in the Lgi1−/− slices, suggesting increased levels of glutamate in the synaptic cleft. However, these increased synaptic glutamate levels could also be due to increased number of synaptic vesicles released per stimulus, as an analysis of the variance of the evoked responses suggested. Strangely, the authors did not report whether the peak amplitude of these evoked EPSCs was increased in the Lgi1−/− slices, even though this would be expected if there were an increase in both the quantal size and quantal content.
Finally, the authors tested whether these slices from P9 pre-epileptic Lgi1−/− mice that showed changes in excitatory synaptic transmission also showed signs of broader network hyperexcitability. Using simultaneous whole-cell and field potential recordings, they found that most slices from mutant mice showed bursts of activity resembling interictal spikes, while no slices from Lgi1+/+ mice did. These events were present in artificial (a)CSF with standard ionic concentrations and without pharmacological treatment, and were blocked by coapplication of AMPA and NMDA receptor blockers, showing that the generation of these events is dependent on glutamatergic transmission, and probably linked to the increased presynaptic glutamate release.
Historically, changes in presynaptic function that may lead to the development of epilepsy have received less attention than postsynaptic and membrane excitability changes. It is possible that presynaptic changes are not as important in the development of epilepsy, but this lack of attention is also partially due to the fact that presynaptic mechanisms are generally more difficult to assay. However, there are now several known human epilepsy genes that have demonstrated roles in presynaptic function in animals (e.g., STXBP1, DNM1 and STX1B), a few described effects of other epilepsy genes specifically on presynaptic function, and several antiepileptic drugs with known presynaptic mechanisms of action. This paper by Boillot and colleagues provides more evidence that presynaptic mechanisms should be considered when designing studies to investigate mechanisms of epileptogenesis, such as numbers of synaptic vesicles in the releasable pool, vesicular release probability, and short-term plasticity. Even when there are other known potential mechanisms at play, be they postsynaptic or nonsynaptic, many gene variants are likely to exert their epileptogenic effects through multiple converging mechanisms. As our knowledge of the molecular and physiological underpinnings of presynaptic functions grows, along with the knowledge of how these functions are impacted in epilepsy, opportunities to modulate epilepsy via presynaptically targeted therapies may emerge.