Combining different experiment and instrument approaches to reconstruct how a protein moves and function

Structural biology is a field in biology research that seeks to understand the role of structure in potentiating function of an enzyme, receptor, or intracellular effector. To do that, it needs to first obtain the structure of the biomolecule of interest through either the gold standard X-ray diffraction technique or the more recent method of cryo electron microscopy (cryo-EM) and X-ray free electron laser spectroscopy (XFEL).


But, the biggest hindrance in structural biology research lies in the tremendous difficulty in obtaining high quality crystals (also known as diffraction quality crystals) suitable for structure elucidation by X-ray crystallography. Hence, multiple approaches have been developed to obtain good quality crystals that yield diffraction images which help reconstruct a particular segment of a protein. Although not of atomic resolution, these X-ray diffraction data nonetheless informs researchers of the areas on the biomolecule requiring more information for detailed structure reconstruction in the next iterative cycle. On the other hand, researchers are also using other molecular techniques to reduce the rate of movement of the protein in an attempt to obtain atomic resolution (~2 to 3 Angstrom) structural snapshots.


A different methodology for tackling the lack of diffraction quality crystals uses alternative structure elucidation instruments such as cryo electron microscopy to obtain good structural depictions of the target protein even though those, at present, are of coarser resolution relative to X-ray diffraction.


But there is more to structural biology research than mere crystallization of proteins followed by X-ray diffraction studies or cryo electron microscopy. Specifically, while obtaining atomic resolution structure of a protein in binding to its target ligand is a significant achievement, it could not help us answer questions on the type of structural changes involved in moving specific protein domains into positions suitable for ligand binding or catalysis. Such information requires other molecular approaches to help peel away the mystery. Reporting the structure of human sterol transporter ABCG5/ABCG8, Rosenbaum and coworkers1 (“Crystal structure of the human sterol transporter ABCG5/ABCG8”, Link) used a variety of molecular biology tools together with molecular dynamics simulation to tease away different facets of the structural story of the transmembrane protein whose structure was elucidated via X-ray diffraction.


After the structure of the transporter was obtained by mapping electron density calculations onto X-ray diffraction data, the authors proceeded with in vivo functional reconstitution of the transmembrane protein in mice for probing the functional consequences of a critical heterodimer interface, where different domains were hypothesized to interact in different ways to offer access to the transporter’s channel. Specifically, recombinant G5 and G8 domains of the transporter were expressed in knockout mice without the sterol transporter, and cholesterol levels in bile measured to assess the effect of specific mutations in either domains on sterol export function of the transporter. Thus, the experiment system offered readout of the phenotypic effect of specific amino acid mutations in either of the two main domains of the sterol transporter.


For a multidomain macromolecular complex such as the sterol transporter investigated, there are many facets and details oblivious to the researchers without the help of tools capable of perturbing (i.e., adjusting) the molecular structure of the complex. To this end, modern molecular dynamics simulations offer fine grained relatively long duration simulation of protein segments or an entire macromolecule. Different from atomic simulations or electronic level calculations, molecular dynamics simulations use different levels of abstraction for different parts of a molecule. For example, in the critical ligand binding pocket, individual atoms may be represented in detail, while other parts of the protein mapped onto a coarse electron density field with characteristics representative of the protein in solution. Using a combination of molecular dynamics simulation and vibrational mode analysis, the authors obtained answers to the question of the likely consequences of ATP binding and hydrolysis on transmembrane domain conformation and sterol binding.


More importantly, the authors also use coevolution analysis to explore the sequence space in which transmembrane domain (TMD) of the transporter would form close contacts at the transmembrane interface during transport of cholesterol, a property important to understanding the diversity of possible molecular conformations useful for cholesterol export. What is important to understand is that such analysis, in addition to yielding information on the evolutionary selection pressure that potentiated the development or selection of particular amino acid residues, also cast a forward glance at the molecular possibilities for designing the inner core channel of a sterol transporter.


Finally, homology modelling was used in understanding the extent of structural conservation between the sterol transporter and a white/brown heterodimer that confers brown eye colour to Drosophila, which illuminates how similar protein structure mediates divergent functions in different organisms. While the conventional notion is that structure defines function, conservation of structure in proteins with different amino acid sequences such as in the binding pocket of interest teaches us how slight variation in structure allow diversity of functions to be mediated by a conserved backbone structure.


Collectively, structural biology has come a long way since the days when the structure of DNA was elucidated via a combination of theoretical modelling and X-ray crystallography. With a greater plethora of molecular biology, bioinformatics, computational biology and cell biology tools, structural biologists are seeking both dynamic (usually computational) and static views of protein in action such as binding a ligand in a key domain. While X-ray diffraction remains the gold standard, cryo electron microscopy has played an increasingly important role in obtaining hard to obtain protein structures, albeit at a lower resolution. Proving the functional consequence of specific protein folds or domain structure is easier now with an expanded genetic engineering toolbox capable of expressing heterologous sequence in knockout animal models useful for checking the physiological effect of specific proteins at the organismal scale. But, difficulty of obtaining crystals of proteins in multiple stages of conformational change needed for specific binding activity meant that molecular dynamics simulation is increasingly playing the de facto role for understanding how structure change in anticipation of action. Finally, bioinformatics sequence analysis continues to inform the amino acid repertoire suitable for specific functions, which could either be used for understanding the sequence space that nature has explored in deriving a protein function, or possible residues for site specific tinkering in adjusting a protein’s affinity to a particular ligand.


Readers may download a pdf copy (with an abstract attached) of this commentary piece at


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s