Limits...
Structure and Function in Homodimeric Enzymes: Simulations of Cooperative and Independent Functional Motions.

Wells SA, van der Kamp MW, McGeagh JD, Mulholland AJ - PLoS ONE (2015)

Bottom Line: The results indicate that the opening/closure motions are intrinsic features of both unliganded enzymes.In DcpS, conformational change is dominated by an anti-symmetric cooperative motion, causing one active site to close as the other opens; however a symmetric motion is also significant.In CS, we identify that both symmetric (suggested by crystallography) and asymmetric motions are features of the protein structure, and as a result the behaviour in solution is largely non-cooperative.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemistry, University of Bath, Bath, United Kingdom.

ABSTRACT
Large-scale conformational change is a common feature in the catalytic cycles of enzymes. Many enzymes function as homodimers with active sites that contain elements from both chains. Symmetric and anti-symmetric cooperative motions in homodimers can potentially lead to correlated active site opening and/or closure, likely to be important for ligand binding and release. Here, we examine such motions in two different domain-swapped homodimeric enzymes: the DcpS scavenger decapping enzyme and citrate synthase. We use and compare two types of all-atom simulations: conventional molecular dynamics simulations to identify physically meaningful conformational ensembles, and rapid geometric simulations of flexible motion, biased along normal mode directions, to identify relevant motions encoded in the protein structure. The results indicate that the opening/closure motions are intrinsic features of both unliganded enzymes. In DcpS, conformational change is dominated by an anti-symmetric cooperative motion, causing one active site to close as the other opens; however a symmetric motion is also significant. In CS, we identify that both symmetric (suggested by crystallography) and asymmetric motions are features of the protein structure, and as a result the behaviour in solution is largely non-cooperative. The agreement between two modelling approaches using very different levels of theory indicates that the behaviours are indeed intrinsic to the protein structures. Geometric simulations correctly identify and explore large amplitudes of motion, while molecular dynamics simulations indicate the ranges of motion that are energetically feasible. Together, the simulation approaches are able to reveal unexpected functionally relevant motions, and highlight differences between enzymes.

No MeSH data available.


DcpS crystal structures with the cleft-distance and the ‘hinge-angle’ measurements labelled.Chain A is depicted in dark purple, chain B in dark green, with C-terminal domains transparent. A) symmetric structure, PDB ID: 1XML.[1] B) asymmetric structure, PDB ID: 1XMM (complexed with m7GDP; not shown here).[1]
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4524684&req=5

pone.0133372.g001: DcpS crystal structures with the cleft-distance and the ‘hinge-angle’ measurements labelled.Chain A is depicted in dark purple, chain B in dark green, with C-terminal domains transparent. A) symmetric structure, PDB ID: 1XML.[1] B) asymmetric structure, PDB ID: 1XMM (complexed with m7GDP; not shown here).[1]

Mentions: DcpS is an enzyme that catalyses the hydrolysis of the 5’ cap structure in eukaryotic mRNA, a central step in mRNA turnover and regulation of gene expression. Crystal structures of this domain-swapped homodimeric enzyme indicate a symmetric apo-form with two open active sites (PDB ID: 1XML) and an asymmetric ligand-bound state (PDB IDs: 1XMM, 1ST0, 1ST4) in which one active site is closed with a ligand bound, and the other is open with no ligand bound [1, 2]. The active sites are located between the N- and C-terminal domains (with a domain-swapped N-terminal domain in the dimer, see Fig 1). Molecular dynamics (MD) simulations[3] revealed that in solution, the apo-form undergoes a continual process of conformational change, where one active site opens (by the N- and C-terminal domains moving away from each other) and the other closes (though not to the full extent of the bound crystal structure), in a strikingly anti-symmetric cooperative fashion: as one closes, the other opens. Flexible hinge regions connecting the N- and C-terminal domains permit this ‘rocking’ motion.


Structure and Function in Homodimeric Enzymes: Simulations of Cooperative and Independent Functional Motions.

Wells SA, van der Kamp MW, McGeagh JD, Mulholland AJ - PLoS ONE (2015)

DcpS crystal structures with the cleft-distance and the ‘hinge-angle’ measurements labelled.Chain A is depicted in dark purple, chain B in dark green, with C-terminal domains transparent. A) symmetric structure, PDB ID: 1XML.[1] B) asymmetric structure, PDB ID: 1XMM (complexed with m7GDP; not shown here).[1]
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4524684&req=5

pone.0133372.g001: DcpS crystal structures with the cleft-distance and the ‘hinge-angle’ measurements labelled.Chain A is depicted in dark purple, chain B in dark green, with C-terminal domains transparent. A) symmetric structure, PDB ID: 1XML.[1] B) asymmetric structure, PDB ID: 1XMM (complexed with m7GDP; not shown here).[1]
Mentions: DcpS is an enzyme that catalyses the hydrolysis of the 5’ cap structure in eukaryotic mRNA, a central step in mRNA turnover and regulation of gene expression. Crystal structures of this domain-swapped homodimeric enzyme indicate a symmetric apo-form with two open active sites (PDB ID: 1XML) and an asymmetric ligand-bound state (PDB IDs: 1XMM, 1ST0, 1ST4) in which one active site is closed with a ligand bound, and the other is open with no ligand bound [1, 2]. The active sites are located between the N- and C-terminal domains (with a domain-swapped N-terminal domain in the dimer, see Fig 1). Molecular dynamics (MD) simulations[3] revealed that in solution, the apo-form undergoes a continual process of conformational change, where one active site opens (by the N- and C-terminal domains moving away from each other) and the other closes (though not to the full extent of the bound crystal structure), in a strikingly anti-symmetric cooperative fashion: as one closes, the other opens. Flexible hinge regions connecting the N- and C-terminal domains permit this ‘rocking’ motion.

Bottom Line: The results indicate that the opening/closure motions are intrinsic features of both unliganded enzymes.In DcpS, conformational change is dominated by an anti-symmetric cooperative motion, causing one active site to close as the other opens; however a symmetric motion is also significant.In CS, we identify that both symmetric (suggested by crystallography) and asymmetric motions are features of the protein structure, and as a result the behaviour in solution is largely non-cooperative.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemistry, University of Bath, Bath, United Kingdom.

ABSTRACT
Large-scale conformational change is a common feature in the catalytic cycles of enzymes. Many enzymes function as homodimers with active sites that contain elements from both chains. Symmetric and anti-symmetric cooperative motions in homodimers can potentially lead to correlated active site opening and/or closure, likely to be important for ligand binding and release. Here, we examine such motions in two different domain-swapped homodimeric enzymes: the DcpS scavenger decapping enzyme and citrate synthase. We use and compare two types of all-atom simulations: conventional molecular dynamics simulations to identify physically meaningful conformational ensembles, and rapid geometric simulations of flexible motion, biased along normal mode directions, to identify relevant motions encoded in the protein structure. The results indicate that the opening/closure motions are intrinsic features of both unliganded enzymes. In DcpS, conformational change is dominated by an anti-symmetric cooperative motion, causing one active site to close as the other opens; however a symmetric motion is also significant. In CS, we identify that both symmetric (suggested by crystallography) and asymmetric motions are features of the protein structure, and as a result the behaviour in solution is largely non-cooperative. The agreement between two modelling approaches using very different levels of theory indicates that the behaviours are indeed intrinsic to the protein structures. Geometric simulations correctly identify and explore large amplitudes of motion, while molecular dynamics simulations indicate the ranges of motion that are energetically feasible. Together, the simulation approaches are able to reveal unexpected functionally relevant motions, and highlight differences between enzymes.

No MeSH data available.