For many years now, neuroscientists have observed a curious phenomenon: large numbers of mRNAs clustered at the dendritic end of neurons, where incoming signals are received. These localized mRNAs are believed to underlie learning, memory, and synaptic plasticity, but precisely which of these molecules are involved and how they are regulated has remained elusive.
“Studying local RNA expression during dendritic signaling is a ‘holy grail’ for understanding the molecular mechanisms of memory formation,” says Robert Darnell, a Howard Hughes Medical Institute Investigator at The Rockefeller University.
That quest, however, has always been limited by one factor: the lack of molecular tools powerful enough to observe dendritic mRNA in the infinitesimally small synapse spaces they inhabit.
But in a paper publishedexternal link, opens in a new tab in May, Darnell and his team describe a new technique that offers a breakthrough solution to the longstanding problem. The method allows mRNAs and their associated proteins to be tagged within 10-to-50-nanometers from the end of dendrites – a 10,000-fold improvement over existing techniques.
“The proximity labeling methods they’ve developed are a clever way to get around one of the main bottlenecks in this field,” says neuroscientist Elly Nediviexternal link, opens in a new tab at the Massachusetts Institute of Technology, who was not involved in the work.
“It’s like coming up with a new telescope and all of a sudden finding things in the galaxy you couldn’t see before,” says Darnell. “It has revealed some astounding and previously unanticipated discoveries.”
Neurons' nifty trick
To create the new tag, the team engineered a bacterial enzyme that attached biotin molecules specifically to dendritic proteins – a “bait of sorts to see what is around this region,” explains the paper’s first author Ezgi Hacisuleymanexternal link, opens in a new tab, then a postdoc in Darnell’s lab, who pioneered the technique.
Because the researchers were keen to study what happens on a molecular level when dendrites receive an incoming signal – a process known as neuronal depolarization – they performed the tagging procedure both in resting neurons and 20 to 30 minutes after the cells had been stimulated. The latter state was obtained by introducing substances that mimic the influx of calcium ions into the neuron during depolarization.
What they observed was illuminating: upon neuronal depolarization, ribosomes on some RNAs would ‘jump’ onto other, previously latent, RNA molecules. “That was a completely unexpected discovery,” says Darnell.
The mechanism is a really smart way for neurons to use the limited resources present in the synapse, adds Hacisuleyman, now an assistant professor at The Herbert Wertheim University of Florida Scripps Institute for Biomedical Innovation & Technology. “There’s only so many ribosomes you can fit in there to be able to synthesize all these different proteins.”
Importantly, the jumping ribosomes enable neurons to respond rapidly to incoming stimuli. “For us to be able to learn and remember things, we need to make new proteins in a timely manner,” says Hacisuleyman. But because neurons can stretch more than a meter long, it’s critical for them to have dendrite-localized RNA and protein-making machinery “ready to go in order for them to respond on the milliseconds scale when a stimulus arrives.”
The fact that mRNAs were present in dendrites but translationally silenced “was very confusing and not understood in the field,” she adds. “Why would all these RNAs be distributed in these tiny spaces? Why would the neuron spend so much energy to be able to put these RNAs out there but not make any proteins?”
But as their experiments demonstrated, the molecules were just waiting for the right signal to spring into action.
Kickstarting protein production
The next thing the researchers observed was where these ‘switched’ ribosomes were jumping to. “It’s like they’re landing on a new landscape,” says Darnell. “So all kinds of new information comes out.”
For example, they learned that the ribosomes would bind to a stretch of mRNAs outside the main coding sequence – a region called upstream open reading frames, or uORFs – that’s otherwise dormant in resting dendrites. This binding would, in turn, recruit and phosphorylate eukaryotic translation initiation factor (eIF4G2), a protein complex which is involved in kickstarting the rapid production of dendritic proteins needed for synaptic plasticity and memory formation.
It was surprising to discover that uORFs and eIF4G2 are involved, says Hacisuleyman. “eIF4G2 is not very well-studied. But we found that it actually functions in the synapse in a way that wasn’t known before.”
Additionally, the uORFs “were not previously suspected to have biologic function,” says Darnell. “We found that instead of coding for traditional proteins, they code for up to 1,000 different micropeptides, ranging from five to 25 amino acids long, that seem to be linked in some way to the up- or down-regulation of protein production in that same mRNA.”
The team is now trying to figure out exactly what these micropeptides do. Additionally, they’re probing further into the proteins translated by the switched ribosomes, many of which are involved in mitochondrial biology. It’s an interesting line of inquiry, says Darnell, as mitochondria are “intimately tied” to many neurodegenerative diseases such as Parkinson’s.
“The importance of this [new] study is in showing that activity not only induces expression of specific gene sets, but also influences which of these RNAs are translated, and potentially where,” says Nedivi.
“It’s a really important and experimentally challenging problem,” she adds. “I wish more people were attacking it at this molecular level of analysis.”
###
Citations
Hacisuleyman, E., Hale, C.R, Noble, N. et al. "Neuronal activity rapidly reprograms dendritic translation via eIF4G2:uORF bindingexternal link, opens in a new tab." PMID: 8589584