It is generally thought that a single mRNA encodes one protein. However, in recent years, with the rise of translatomics, researchers have discovered that many mRNAs have multiple translation initiation sites (TIS), which can produce several versions of a protein, called proteoforms. Despite these findings, the physiological significance of this mechanism remains largely unclear. In a new study published in Molecular Cell on September 23, the lab of Junjie Guo, assistant professor of neuroscience at Yale School of Medicine, investigated alternative translation initiation of mRNAs encoding synaptic organizers—molecules that regulate the organization of neuronal synapses— and its potential physiological function.
The work, led by Paul Jongseo Lee, graduate student in Yale’s Interdepartmental Neuroscience Program (INP), focused on the synaptic organizer neuronal pentraxin receptor (NPR), a membrane protein that has been shown to regulate AMPA-type glutamate receptors at excitatory synapses. The study showed that NPR mRNA has two distinct TISs: a conventional start codon (AUG), and an unconventional start codon (CUG). These two TISs give rise to two proteoforms: the CUG TIS produces the known membrane-bound long proteoform, whereas the AUG TIS produces a novel short proteoform. And this pair of NPR proteoforms can be found across all vertebrate species.
Our study was driven by the insight that alternative translation initiation sites showing clear evolutionary conservation are likely to have functional significance.
Junjie Guo, PhD
Do these proteoforms have, indeed, different neuronal functions?
While the long NPR proteoform is known to be a membrane protein with a single N-terminal transmembrane domain, the team discovered that the novel short NPR proteoform had a truncated N-terminal signal sequence and was converted to a secreted factor. Therefore, the two proteoforms have different localizations: one remains on the cell surface, while the other is secreted.
Next, the team investigated what governs the choice between the production of long or short NPR proteoforms. They found that a specific RNA structure was critical in driving CUG initiation and the production of long NPR. They also showed that increased neuronal activity leads to greater production of short NPR proteoform. That is the case, for example, for neurons in the hippocampus of mice when they explore a novel and enriched environment.
Finally, Lee showed that the short NPR proteoform promotes the clustering of GluR4-containing AMPA receptors, mostly found in parvalbumin-positive (PV) interneurons in the hippocampus, which have important roles in learning and memory. By introducing a single-nucleotide mutation in the CUG TIS, the team generated mice that predominantly expressed the long NPR proteoform, lacking the short form. These mice exhibited impaired cognitive functions, indicating that the proper ratio between long and short NPR proteoforms is essential for PV neuron function and related behaviors.
The discovery of this alternative translation mechanism and its activity-dependent modulation raises the question of its generality. The researchers demonstrated that, in another synaptic organizer family, the conversion was happening the other way around: their alternative TIS converted secreted factors into membrane-bound proteoforms. Based on translatomics data, they predict that similar alternative TIS-mediated conversions occur in many other mRNAs in human and mouse cells.
What surprised us the most was that the function of NPR alternative translation, namely, changing the nature of the N-terminal signal sequence that determines protein localization, also applied to a variety of other mRNAs with alternative translation sites.
Junjie Guo, PhD
These findings demonstrate a previously underappreciated gene regulatory mechanism that diversifies protein functions. Since translation dysregulation has been widely associated with neurological disorders, changes in alternative TIS usage may play crucial roles in these pathologies.
The findings were published on September 23 in the journal Molecular Cell. The first author is Paul Jongseo Lee, graduate student in the Interdepartmental Neuroscience Program (INP), and the senior author is Junjie U. Guo, assistant professor of neuroscience. Behavioral experiments were conducted in collaboration with the lab of Marina Picciotto, Charles B. G. Murphy Professor of Psychiatry and professor in the Child Study Center, of neuroscience and of pharmacology.