In the image of the ‘Demogorgon’ spike on
this page, each of the three proteins that form
the spike has been given a different colour.
These three proteins all need to spread apart
to open up access to the ACE2 binding site, so
that it can interact with the surface of human
cells and initiate infection.
After the opening motion was visualised,
the Folding@home Twitter account explained
that Folding@home’s next step was to
‘help figure out where the protein spends
the majority of its time using thousands of
simulations run by our donors. This kind
of information will help to prioritise drug
development efforts’.
Developing therapies
Helping with prioritisation is one way that
Folding@home’s research helps with the
development of therapeutics. At the moment,
Bowman says Folding@home is collaborating
with Diamond (diamond.ac.uk) in the UK,
an X-ray beamline company that’s currently
using macromolecular crystallography (MX) to
study COVID-19.
Exploring beamline technology in depth
is beyond the scope of this feature, but in
basic terms, Diamond’s techniques can
enable researchers to observe the shape
of biological molecules at atomic resolution
experimentally, rather than virtually. If you
head to diamond.ac.uk/covid-19.html,
you can see a video of how the COVID-19
crystallography setup works.
It’s amazing stuff. Hundreds of samples are
stored under liquid nitrogen, and each sample
in turn is put under a cool gas nozzle rotated
in an x-ray beam, while the machine captures
3,600 diffraction images. That data is then
processed, forming all the diffraction patterns
into a three-dimensional image of the
structure. Diamond is observing how different
chemical compounds bind to the SARS-
CoV-2 Mpro protein, and Folding@home
can help by narrowing down which chemical
compounds to try.‘Diamond has been solving experimental
structures of the COVID-19 proteins and
screening for small molecule drugs,’ says
Bowman. ‘What we’re trying to do with these
free energy calculations I mentioned earlier,
is to take large libraries of chemicals that you
can either buy or synthesise, and see which
ones might be the most useful, based on our
simulations. We then help to prioritise them for
subsequent experiments by Diamond MX.
‘This is really important, because they
could buy a bunch of random chemicals and
synthesise a bunch of stuff, but the likelihood
that any of it works is low with the finite
resources they have. What we can do is help
prioritise what to buy or make in order to
maximise the chances that they’re useful.’
Finding the binding sites in the first
place is also a key part of Folding@home’s
research, thanks to the project’s ability to
model how proteins move at the atomic
level. The Demogorgon opening motion is
one example, but there have been recent
important developments in other areas too.
‘What we’re able to do with the simulations is
watch all these moving parts,’ says Bowman,
‘and often these motions create novel
binding sites for small molecule drugs, which
you’d never guess were there based on the
experimental structure.’A visualisation of the SARS-CoV-2 Mpro protein
from Diamond. The company is observing how
different chemical compounds bind to it
A visualisation of Diamond’s crystallography
technique in action