A protein’s shape plays a fundamental role in its function. Structural biology strives to construct models, ultimately at atomic resolution that represent snapshots of biological macromolecules and to describe the ways in which these molecules move.
The current dearth of protein structural information reflects the complexity of this challenge. Of the approximately 15,000 protein families, there are still about 5,200 with unknown structure outside the range of comparative modeling. Moreover, the behavior of the vast variety of proteins and their rapidly changing conformations depends on the experimental conditions, making it difficult to study them with a single technique. Over the last few decades, biologists analyzed protein structures using X-ray crystallography, nuclear magnetic resonance (NMR), or electron microscopy (cryo-EM) on samples at cryogenic temperatures. These are vital techniques, because the resolution achieved can be down to the nanometer, angstra, or atomic level. They provide essential information, but they capture the structure in a frozen state. To unravel protein function, scientists must explore protein dynamics, and that can be done with mass spectrometry (MS).
MS captures a sample’s mass-to-charge ratio, which can be used to identify and quantify proteins. By integrating results from different types of MS, scientists can determine protein structures and the mechanisms behind specific functions. This process often requires computational tools. The combination of data and models from different experiments reveals how a protein or protein complex works, including the role of binding factors, post-translational modifications, and interactions with other molecules such as drugs.
Such integrative approaches unveil the basic biology of proteins, and how they can be used. By combining MS with the right set of more conventional techniques, such as EM, researchers can make the most of a method’s strong points and offset its weaknesses. Despite advances in using and combining these techniques, scientists and engineers keep searching for improvements.
Even though MS can be combined with traditional techniques used in structural biology, one kind of MS is often not enough. Unfortunately, no MS technique does everything the best. For example, a protein or complex of proteins can be kept in the native state i.e. its typical shape under ordinary biological environmental conditions and analyzed with MS. The intact weighing of the mass of the protein complex lets us to find out which proteins and cofactors are part of it. This method keeps proteins in natural assemblies when delivering them to the detector. Another kind of MS technique, crosslinking MS (XL-MS), can be used to determine which parts of a protein or complex are in contact. A chemical glue is used to connect two lysine groups in close proximity. They might be in a single protein or proteins close to each other. Applying this technique to many lysine groups reveals structural constraints because we can see which parts of a protein or which proteins in a group are in proximity.
XL-MS can also be combined with cryo-EM. The combination of cryo-EM and crosslinking was used to study a molecular complex involved in transcribing DNA to RNA. The cryo-EM and XL-MS was also combined to explore the structures involved in splicing RNA.
Scientists can also study the structure of macromolecules with hydrogen-deuterium exchange MS (HDX-MS). Here, the sample is dissolved in heavy water, D2O. All the amide hydrogen on the protein’s surface starts to get exchanged for deuterium. Hydrogens that are less accessible, buried somewhere inside the protein structure are exchanged substantially slower, and this can tell which parts of the protein are outside, and which are inside.
Although scientists developed HDX-MS several decades ago, it could only be used on one small protein at a time. Now, scientists can apply HDX-MS to whole viruses, because of several advances in MS and data processing.
Ups and downs of MS
Although today’s scientists can select from a range of MS techniques, that doesn’t make structural analysis easy. For one thing, exploring protein structure with MS requires upstream processing, including sample preparation and some form of separation, like liquid chromatography (LC) or capillary electrophoresis. The MS platform also needs to provide high sensitivity. In some samples, scientists search for extremely rare components, such as crosslinked peptides. There we need Nano-LC to separate all the peptides, followed by fast and sensitive MS.
Despite some of the challenges of applying MS to protein structure determination, this technology comes with many strengths, such as identifying small binding proteins and protein post-translational modifications; quantifying the heterogeneity of a sample; determining the ratio of the subunits in a protein complex and how the ratio changes over time or under different conditions; and tracking changes in protein conformations.
Advances in MS technology, both in hardware and software have turned it into a tool for probing structural biology. Today’s mass spectrometry is so much faster and more sensitive and the software to analyze the data is faster, more flexible, and provides smarter algorithms for looking at different sets of large data.
The conventional techniques used to analyze the structure of biological molecules, like X-ray crystallography, can reveal the locations of components down to the atom. To use this technique, however, the recombinant protein must be crystallized, which is extremely challenging with some proteins, particularly if they are membrane-bound. In those cases, researchers can use cryo-EM to prepare very high-resolution images. But cryo-EM gives us one image of one specific moment in time.
To study the dynamics of protein structures, today’s scientists turn to MS. Although the resolution is lower with MS, the ability to examine temporal changes increases substantially with this technology and combining the various forms of MS can tease out different aspects of a molecular structure. So, X-ray crystallography, NMR, or cryo-EM can be combined with one or more forms of MS such that each collects information on some aspect of a protein’s structure.
Further on down the road
Currently, scientists must cobble together various methods and techniques, often manually integrating the results to generate the best data. As those steps turn into a more cohesive workflow, integrative structural biology will be applied to an even wider range of questions, including novel functions of structures, protein–protein interactions, therapeutic targets, and more. Along the way, this field will uncover new knowledge about how biological systems work, and how they fail. The latter will help clinical researchers understand, diagnose, and treat diseases. However, doing that depends on combining areas of expertise from protein biophysics to drug discovery and beyond with the right collection of tools for probing and analyzing complicated biological structures, all on a very fine scale. Only then will we have a complete understanding of the very specific ways that a protein’s shape determines its function.