New Paper: Computational Study of a Brain Protein

A new paper from our lab has just been published! We looked at how the brain works at atomic resolution – well, not the whole brain yet, unfortunately, but one important tiny part of it. We studied an NMDA receptor, which is a key actor in how our memory forms and how we arrive at abstract concepts. NMDA receptors are also involved in numerous neurological disorders, including schizophrenia, epilepsy, Alzheimer’s, Parkinson’s and many other diseases. This protein is really large, and it would be difficult to simulate it all. Fortunately, it consists of several relatively autonomous parts. We took two of those parts, simulated and analyzed them.

The main conclusion of the paper is: sugars affect the receptor’s behavior. Wait a minute, what are sugars doing here? Many proteins in our body have sugars (also called glycans, to make it sound more scientific) attached to these proteins. NMDA receptors are no exception: each has at least 30 glycan tails attached. However, researchers often don’t think about glycans. When they want to experimentally study some parts of NMDA receptors, they express them in bacteria, but bacteria normally don’t put glycans on proteins. When people run simulations, they don’t add glycans to the simulated system. How bad is ignoring glycans? Does it change anything? Our work shows that glycans are definitely important.

The parts of the receptor that we studied are like pacmen or like clamshells: they can open and close. This opening and closing is important for NMDA receptors to work properly. It turns out that one of the glycans is attached to one side of the clamshell (or a jaw of a pacman), and it can also interact with the other side when the clamshell is closed. This glycan serves as a latch, making it easier for the clamshell to close. Our experimental collaborators proved it in an experiment: when this glycan is removed, it is more difficult to open the ion channel (closing the clamshell corresponds to opening of the ion channel, and vice versa).

This observation suggests a new mechanism of how some neurological disorders could develop in some cases. If this glycan is not put to where it is supposed to be, then NMDA receptors will be insufficiently active. As is well-known, their insufficient activity may lead to schizophrenia. We haven’t found any patients with disrupted glycosylation of NMDA receptors. However, a hundred of diseases caused by disruptions in glycosylation of other proteins are known, so this scenario is quite plausible.

Finally, it’s really amazing how many details in a studied protein our computer simulations can reveal. Even up-to-date experimental methods would not give us such a highly resolved picture of how atoms move in the NMDA receptor in such short periods of time as nanoseconds or microseconds (that is, billionth or millionth parts of a second). And thanks a lot again to all of you who donate computer time to Folding@home!

You will find the abstract here: http://folding.stanford.edu/2017/04/05/computationally-…n-nmda-receptors/

And the article is found here: https://www.ncbi.nlm.nih.gov/pubmed/28378791