Elfn1 Helps Neurons Make the Right Connections

What guides neurons to establish their highly specific connections? In this post, Kevin Mitchell explores some of the research his lab and others have done to identify Elfn1 as a key player in synaptic labeling. — E.R.

By Kevin Mitchell

The diversity of cell types in the nervous system is mirrored by a similar, perhaps even greater, diversity of synapse types. Each synapse is characterised by a particular biochemical profile of pre- and post-synaptic proteins that determine its neurotransmitter responsiveness, electrophysiological properties and also how the synapse changes in response to activity. Each neuron makes different types of synapses onto diverse target cell types, in an organised and target cell-appropriate manner. This must entail some kind of recognition system that lets the presynaptic neuron know which cell type it has encountered.

In this post I review a series of studies that have identified the first protein that acts as a label to specify the type of synapse made onto target cells. This research has begun to elucidate both the biochemical mechanisms underlying its function and the phenotypes associated with the absence of this important signalling protein. We can presume that many more such labels exist, to account for the enormous diversity of synapse types in the mammalian nervous system. These studies illustrate an important general point – that a highly selective and fairly subtle change to the connections made onto a very specific cell type can lead to wholesale disruption of neural network dynamics and the emergence of neurological phenotypes.

Searching for connectivity labels

Elfn1 is a member of the extracellular leucine-rich repeat (LRR) protein superfamily, known to include many proteins involved in innate immunity (recognising diverse antigens) or in nervous system development (recognising diverse cell types). Several years ago, my lab surveyed the entire set of extracellular LRR proteins in worms, flies, mice and humans. Multiple members of this family had already been identified as playing important roles in mediating neuronal guidance and synaptic connectivity, but it was unknown how large the family was. Using a bioinformatics pipeline and clustering method, Jackie Dolan, Karen Walshe and others identified over 130 such proteins in mammals, 66 in flies and 29 in worms. The functions of most of these were unknown at the time, and many remain so.

An expression screen identified a number of novel LRR genes that were expressed in highly selective fashion in the developing nervous system of mice or flies. One subfamily of two genes, which we named the Elfns, showed particularly intriguing expression. In the cortex and hippocampus Elfn2 mRNA is broadly expressed in pyramidal neurons, while Elfn1 is expressed in a subset of inhibitory interneurons. These genes also show differential expression in various subcortical structures, with Elfn2 strong in striatum and Elfn1 expression notable in the globus pallidus and habenula. The known functions of other LRR proteins and the selective expression pattern of Elfn1 suggested it might play a role in specifying neuronal connectivity in the developing brain. This hypothesis turned out to be correct, but in a more subtle way than we had anticipated.

Elfn1 specifies synapse type

Emily Sylwestrak and Anirvan Ghosh also noticed Elfn1 in a screen of the Allen Brain Atlas for genes expressed in subsets of hippocampal interneurons. Elfn1 is expressed in a distinct set of interneurons, the majority of which are also positive for the marker somatostatin, particularly those found in the layer of neuropil called the stratum oriens-lacunosum moleculare (OLM cells). Pyramidal neurons in area CA1 make synapses onto these OLM cells and also onto a distinct set of interneurons that express parvalbumin but not Elfn1 (PV cells).

These two types of interneurons have different computational functions within hippocampal microcircuits and the synapses that CA1 cells make onto them have different electrophysiological properties. In particular, the synapses onto OLM cells have initially low probability of release of synaptic vesicles in response to an action potential, but they are facilitating – i.e., they respond to repeated stimulation by short-term strengthening of the synaptic connection. This means they respond best to high-frequency stimulation and thus act as high-pass filters. By contrast, the synapses onto PV cells start out with high probability of release, but are depressing – the synapses undergo short-term weakening in response to activation; these synapses act as low-pass filters. These different types of plasticity are determined by the differential deployment of various proteins on the presynaptic side (including metabotropic glutamate receptors and kainate receptors, among many others) and are an important part of the functional architecture of hippocampal circuits.

Using knockdown and ectopic expression Sylwestrak and Ghosh were able to show that Elfn1 was required in OLM cells, and sufficient, if expressed in PV cells, to instruct formation of a facilitating synapse. This was the first molecule shown to act as a postsynaptic label to specify what type of synapse the presynaptic cell should make at that position.

Mutation of Elfn1 disrupts neural function

To address how important this role was to the functioning of the nervous system, Jackie Dolan in my lab analyzed Elfn1 knockout mice. At the anatomical level, these animals did not present any obvious differences from wild-type mice. The number and position of Elfn1-expressing cells was normal and the axonal projections to and from Elfn1-positive regions was also unaffected. However, behavioural analyses revealed important consequences of mutation of Elfn1 on the integrity of neuronal networks.

The most obvious effect is that the mutant animals have seizures, which begin after three or four months of age. These seizures are partial, with a variable presentation that can include loss of postural control, rapid running and jumping or tonic convulsions.

Subsequently published work from the lab of Jun Aruga on a separate line of Elfn1 knockout mice confirmed the epilepsy phenotype and showed that these seizures can be triggered by sensory stimulation, particularly handling after a period of isolation. Using these animals, Aruga and colleagues have also gone on to elucidate more mechanistic details of Elfn1’s function in specification of the electrophysiological properties of synapses.

In a beautiful series of analyses, they show that Elfn1 binds trans-synaptically to the metabotropic glutamate receptor mGluR7 and recruits it to synapses. In the absence of Elfn1, the levels of mGluR7 protein in the fibres of CA1 cells in the hippocampus are dramatically reduced. Other synaptic markers were unaffected, indicating that absence of Elfn1 did not affect the number of synapses but rather their biochemical profiles. They also found, in line with the results of Sylwestrak and Ghosh, that in the absence of Elfn1, synapses made from CA1 cells onto OLM interneurons showed short-term depression, rather than facilitation, in response to brief activation. Together with the observation that mGluR7 mutant mice show a very similar seizure phenotype to Elfn1 mutants, these data strongly suggest that the changes in nature of synapses onto OLM interneurons are the root cause of epilepsy in these animals.

Additional effects of Elfn1 mutation in mice and possibly humans

In addition to seizures, we and the Aruga lab both also observed ADHD-like phenotypes in Elfn1 mutants. We showed that hyperactivity in these animals is paradoxically reversed by amphetamine (as in ADHD patients). These defects may relate to additional functions of Elfn1 in other brain regions, possibly the habenula (as lesions of the medial habenula cause similar phenotypes).

Finally, Aruga and colleagues identified rare mutations in ELFN1 in a small number of cases of epilepsy, autism and ADHD. Recapitulation of these mutations in mice should be able to address whether they are really pathogenic. If so, they would add to the growing number of synaptic genes implicated as individually rare, but collectively common, causes of neuropsychiatric disorders.

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

References

Blackman AV, Abrahamsson T, Costa RP, Lalanne T, Sjostrom PJ (2013) Target-cell-specific short-term plasticity in local circuits. Frontiers in synaptic neuroscience 5: 11. doi: 10.3389/fnsyn.2013.00011.

Dolan J, Mitchell KJ (2013) Mutation of Elfn1 in mice causes seizures and hyperactivity. PLoS ONE 8: e80491. doi: 10.1371/journal.pone.0080491

Dolan J, Walshe K, Alsbury S, Hokamp K, O’Keeffe S, et al. (2007) The extracellular leucine-rich repeat superfamily; a comparative survey and analysis of evolutionary relationships and expression patterns. BMC Genomics 8: 320. doi:10.1186/1471-2164-8-320.

Kullmann DM. (2011) Interneuron networks in the hippocampus. Curr Opin Neurobiol. 21(5):709-16. doi: 10.1016/j.conb.2011.05.006.

Lee YA, Goto Y. (2011) Neurodevelopmental disruption of cortico-striatal function caused by degeneration of habenula neurons. PLoS One. Apr 29;6(4):e19450. doi: 10.1371/journal.pone.0019450.

Mitchell KJ. (2011) The genetics of neurodevelopmental disease. Curr Opin Neurobiol. 21(1):197-203. doi: 10.1016/j.conb.2010.08.009.

O’Rourke NA, Weiler NC, Micheva KD, Smith SJ (2012) Deep molecular diversity of mammalian synapses: why it matters and how to measure it. Nature Reviews Neuroscience 13: 365-379. doi: 10.1038/nrn3170.

Sansig G, Bushell TJ, Clarke VR, Rozov A, Burnashev N, et al. (2001) Increased seizure susceptibility in mice lacking metabotropic glutamate receptor-7. J Neurosci. 21(22):8734-45.  doi:0270-6474/01/218734-12.

Sylwestrak EL, Ghosh A (2012) Elfn1 regulates target-specific release probability at CA1-interneuron synapses. Science 338: 536-540. doi: 10.1126/science.1222482.

Tomioka NH, Yasuda H, Miyamoto H, Hatayama M, Morimura N, et al. (2014) Elfn1 recruits presynaptic mGluR7 in trans and its loss results in seizures. Nature Communications 5: 4501. doi: 10.1038/ncomms5501.

Kevin Mitchell is a neurogeneticist interested in the genetics of brain wiring and of neurodevelopmental disorders. You can find him on twitter at @WiringtheBrain