Q: For any two of the following, describe how our knowledge of high-resolution three-dimensional structure has contributed to our understanding of:
i) The mitochondrial ATP synthase
ii) The proteasome
iii) The SecYEG translocon
Q: a) ‘Protein crystallography has made a major contribution to our understanding of macromolecular machines.’ Discuss the validity of this statement, using examples of at least two macromolecular machines. (80% of marks)
b) What other methods have been usefully applied to study how macromolecular machines work at the molecular level? (20% of marks)
Crystallography holds as shorthand in protein science for crystallisation and subsequent structural elucidation through interpretation of X-ray diffraction patterns. This structural technique is complementary to other methods, but to first examine its contribution to our understanding of macromolecular machines I will describe its application to the mitochondrial ATP synthase and SecYEG translocon.
The mitochondrial ATP synthase (an ATPase working ‘in reverse’) is a wonder of biological engineering, featuring a 6-fold pseudosymmetric stator which undergoes conformational change between 3 states (L, loosely bound ATP; O, a conformation with low affinity for ligand; T, tightly binding ATP) to catalyse ATP condensation through some 150 rotations per second.
Despite common misconception, ATP does not contain a unique ‘high energy’ phosphate bond. In the living cell, the ratio of ATP to ADP is maintained at ~5 (an order of magnitude out of equilibrium, i.e. against hydrolysis), and work must be put in to move against this equilibrium in forming ATP.
The surface of the F1 domain of mitochondrial ATP synthase provides shelter — a platform away from solution equilibrium. Consequently the difference in free energy from ADP (and Pi) to ATP is all but negligible. ATP spontaneously forms on the enzyme surface.
The input of energy to drive the reaction, rather peculiarly, is used to untether the ATP (i.e. move it against the concentration gradient, into solution). It is for this task that the ATPase harnesses the proton-motive force through the mitochondrial inner membrane.
The FO region is a proton pore embedded in the membrane with three main subunits A, B and C. The A protein rests against the ring of C transmembrane helices which form the rotor. The B protein runs from the membranous A protein down to the ATP-catalysing ‘head’ of the F1 subunit, while the C ring is associated with the F1 ε subunit, and the γ ‘stem’ bridges head and rotor.
The catalytic ‘head’ of the FO domain mentioned above could be seen to bear three distinct conformational states, which were directly responsible for deducing (or at first proposing) a mechanism.
Through obtaining structures of the two domains in the complex by X-ray crystallography, a detailed mosaic structure may be constructed within the context of a low-resolution electron microscopy structure of the entire assembly, averaging the images of single enzyme particles. Sub-structures solved by NMR are also invaluable to the finer details of the problem.
Currently solved substructures comprise the F1 domain, the F1 domain with bound inhibitor peptide IF1, the F1 domain with attached c-ring, and a sub-complex representing most of the peripheral (B) stalk.
2014 stood as the UN-designated International Year of Crystallography, in honour of the centennial of the X-ray crystallography method. Coincidentally, the celebration marked the closing of the third and final phase of the US Protein Structural Initiative (PSI:Biology), in which biologically relevant targets including membrane targets such as ATP synthase were subjected to concerted attempts at crystallisation.
The PSI offers a fairly standardised framework to evaluate the method. Since the millennium, ~90% of its gains were achieved from crystallography, the rest largely NMR (over 6,000 structures to date).
Beyond direct observation of structures, these efforts have significantly advanced the state of the art in solving unique protein folds, crucial to unearthing associated functions and in the related project of homology modelling. These no doubt augment our understanding of macromolecular machines.
Hybrid methods, as noted above, are also vital. Mass spectrometry and computational modelling are also integrated into structural characterisation, yet these approaches are often beyond the scope of individual labs in implementation let alone development.
The bacterial SecYEG translocon is another macromolecular machine, exemplary of translational rather than rotary motion.
The transmembrane complex is again powered by an ATPase*, SecA, which transduces energy from ATP hydrolysis to push secretory proteins across the cytoplasmic membrane. To accomodate the translocating polypeptide chain, and release transmembrane segments of membrane proteins into the lipid bilayer, SecYEG opens its central channel and lateral gate. Crystal structures have been essential to visualising these steps.
* other nucleoside triphosphates are available
Unlike ATP synthase, SecYEG interacts with other macromolecular machines. Ribosomes bearing nascent polypeptide ripe for secretion approach the translocon, and a fleet of further modifying proteins fine-tune its positioning, and the enzyme’s membrane insertion.
SecA is a soluble protein, binding the transmembrane SecYEG complex, turning on its motor. Crystallisation depicted two states: open and closed. ATP binds between two nucleotide-binding domains (NBD1 and NBD2), preprotein cross-linking domain (PPXD) and helical scaffold domain cross-link the nascent polypeptide chain (‘preprotein’) to SecA during its translocation.
X-ray images show the channel’s funnel-like entrance and hour-glass body (formed from the tilt of the SecY transmembrane segments), constricted halfway in a hydrophobic plug of isoleucines that seal off the cytoplasm and periplasm when in the closed state.
In 2008, Zimmer and colleagues pieced together these components to propose a mechanism whereby ATP hydrolysis powered translocation through a two helix finger protruberance on SecA, which pokes down into the transmembrane channel, i.e. upon binding the SecA PPXD breaks the small helical ‘plug’ and the ‘finger’ threads the preprotein through. What’s more, SecA’s binding opens a ‘window’ at the lateral gate of the SecY channel as it displaces the plug domain, preparing the channel for signal sequence binding and subsequent channel opening as the ribosome docks.
SecA adopts a clamp-like structure, the widening and tightening of which is linked to ATP hydrolysis.
Recent work through FRET and fluorescence spectroscopy with single-molecule sensitivity has allowed for a more careful dissection of SecY complex dynamics and assembly (notably dismissing the necessity for SecYEG complex dimerisation, though it has been suggested from experimental covalent linkage and may indeed occur).
To distinguish the exact mechanism of SecYEG’s translocation (e.g. Brownian ratchet, power stroke and peristalsis models), crystallography alone is insufficient. Kusters and Driessen recommend fresh biophysical techne, with both single-molecule and time-resolution.
⇢ More questions answered
