New Software Uses X-Ray Diffraction to Generate 3-D Images of Molecular Machines

Scientists are making it simpler for pharmaceutical companies and researchers to visualize the precise inner processes of molecular machines.

“Inside each cell in our bodies and inside every bacterium and virus are tiny but complex protein molecules that synthesize chemicals, replicate genetic material, turn each other on and off, and transport chemicals across cell membranes,” said Tom Terwilliger, a US Department of Energy (DOE) Los Alamos National Laboratory (Los Alamos, NM, USA) scientist. “Understanding how all these machines work is the key to developing new therapeutics, for treating genetic disorders, and for developing new ways to make useful materials.”

These molecular machines are very minute: a million of them positioned side-by-side would take up less than an inch of space. Researchers can see them however, using X-rays, crystals, and computers. Researchers produce billions of copies of a protein machine, dissolve them in water, and grow crystals of the protein, similar to growing sugar crystals except that the machines are larger than a sugar molecule.

After that, the scientists shined a beam of X-rays at a crystal and measure the brightness of each of the thousands of diffracted X-ray spots that are produced. Then researchers use the powerful Phenix software, developed by scientists at Los Alamos, Lawrence Berkeley National Laboratory (LBNL; Berkeley, CA, USA), Duke (Durham, NC, USA), and Cambridge Universities (UK), to analyze the diffraction spots and produce a three-dimensional (3-D) image of a single protein machine. This image shows the researchers precisely how the protein machines are constructed.

Recently, Los Alamos scientists worked with their colleagues at LBNL and Cambridge University to make it even easier to visualize a molecular machine. In a report published online December 22, 2014, in the journal Nature Methods, Los Alamos scientists and their colleagues demonstrated that they captured 3D images of molecular machines using X-ray diffraction spots that could not previously be analyzed.

Some molecular machines contain a few metal atoms or other atoms that diffract X-rays differently than the carbon, oxygen, nitrogen, and hydrogen atoms that comprise most of the atoms in a protein. The Phenix software locates those metal atoms first, and then uses their locations to find all the other atoms. For most molecular machines, however, metal atoms have to be incorporated into the machine artificially to make this all work.

The key new advance to which Los Alamos scientists have contributed was revealing that useful statistical techniques could be applied to find metal atoms even if they do not scatter X-rays very differently than all the other atoms. Even metal atoms such as sulfur that are naturally part of nearly all proteins can be found and used to generate a 3-D picture of a protein. Now that it will often be possible to see a three-dimensional picture of a protein without artificially integrating metal atoms into them, many more molecular machines can be examined.

Molecular machines that have recently been seen in 3-D detail include a massive molecular machine called Cascade that will be reported in the journal Science later in 2015. The Cascade machine is present in bacteria and can recognize DNA that comes from viruses that infect the bacteria. The Cascade machine is comprised of 11 proteins and an RNA molecule and looks similar to a seahorse, with the RNA molecule snaking through the whole ‘body’ of the seahorse. If a foreign piece of DNA in the bacterial cell is complementary to part of the RNA molecule then another specialized machine can come by and cut up the foreign DNA, saving the bacterium from infection.

Los Alamos and Cambridge University scientists who were developing the Phenix software were part of the team that visualized this protein machine for the first time. The Phenix software has been used to determine the 3-D shapes of over 15,000 different protein machines and has been cited by over 5,000 scientific publications.

Related Links:
Los Alamos National Laboratory