Limits...
Structural studies of P-type ATPase-ligand complexes using an X-ray free-electron laser.

Bublitz M, Nass K, Drachmann ND, Markvardsen AJ, Gutmann MJ, Barends TR, Mattle D, Shoeman RL, Doak RB, Boutet S, Messerschmidt M, Seibert MM, Williams GJ, Foucar L, Reinhard L, Sitsel O, Gregersen JL, Clausen JD, Boesen T, Gotfryd K, Wang KT, Olesen C, Møller JV, Nissen P, Schlichting I - IUCrJ (2015)

Bottom Line: The data reveal the binding sites of a variety of ligands, including lipids and inhibitors such as the hallmark P-type ATPase inhibitor orthovanadate.Even at relatively low resolution and multiplicity, the identification of ligands can be demonstrated.This makes SFX a useful tool for ligand screening and thus for unravelling the molecular mechanisms of biologically active proteins.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Molecular Biology and Genetics, Centre for Membrane Pumps in Cells and Disease - PUMPkin, Danish National Research Foundation, Aarhus University , Gustav Wieds Vej 10c, 8000 Aarhus C, Denmark.

ABSTRACT
Membrane proteins are key players in biological systems, mediating signalling events and the specific transport of e.g. ions and metabolites. Consequently, membrane proteins are targeted by a large number of currently approved drugs. Understanding their functions and molecular mechanisms is greatly dependent on structural information, not least on complexes with functionally or medically important ligands. Structure determination, however, is hampered by the difficulty of obtaining well diffracting, macroscopic crystals. Here, the feasibility of X-ray free-electron-laser-based serial femtosecond crystallography (SFX) for the structure determination of membrane protein-ligand complexes using microcrystals of various native-source and recombinant P-type ATPase complexes is demonstrated. The data reveal the binding sites of a variety of ligands, including lipids and inhibitors such as the hallmark P-type ATPase inhibitor orthovanadate. By analyzing the resolution dependence of ligand densities and overall model qualities, SFX data quality metrics as well as suitable refinement procedures are discussed. Even at relatively low resolution and multiplicity, the identification of ligands can be demonstrated. This makes SFX a useful tool for ligand screening and thus for unravelling the molecular mechanisms of biologically active proteins.

No MeSH data available.


Related in: MedlinePlus

Plot of free R factor against resolution after refinement (all atom coordinates, individual B factors) to 2.8 Å resolution against the original data set (dashed black lines) and hybrid data sets (red lines) with scrambled intensities. The arrows indicate the beginning of the scrambled data at (a) 5.82 Å, (b) 4.62 Å, (c) 4.04 Å, (d) 3.67 Å, (e) 3.41 Å, (f) 3.21 Å, (g) 3.04 Å and (h) 2.91 Å. These cutoffs were chosen in accordance with the resolution shells used by phenix.refine to calculate the Rfree values of the model.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4491313&req=5

fig3: Plot of free R factor against resolution after refinement (all atom coordinates, individual B factors) to 2.8 Å resolution against the original data set (dashed black lines) and hybrid data sets (red lines) with scrambled intensities. The arrows indicate the beginning of the scrambled data at (a) 5.82 Å, (b) 4.62 Å, (c) 4.04 Å, (d) 3.67 Å, (e) 3.41 Å, (f) 3.21 Å, (g) 3.04 Å and (h) 2.91 Å. These cutoffs were chosen in accordance with the resolution shells used by phenix.refine to calculate the Rfree values of the model.

Mentions: As a control for the significance of the information in the low-multiplicity 2.8 Å resolution data, we generated data sets with randomly scrambled reflection intensities ranging from a wide shell of scrambled intensities (5.8–2.8 Å) to a narrow shell (2.9–2.8 Å) at high resolution only. We compared the statistics of refinement against these data with those derived from refinement against the unperturbed data. As seen in Fig. 3 ▸, the introduction of the scrambled data was clearly detectable as a deviation point in Rfree values, even if only the highest resolution shell between 2.9 and 2.8 Å had been scrambled. This shows that the SFX intensities contain consistent signal. As a further analysis of the quality of the higher resolution data, we performed refinement of scrambled models to challenge the benefit and phase bias of a pre-refined model. Here, however, we found that the model refinement could not converge towards an improved model. Thus, the higher resolution data only improve map features when high-quality prior phases are available.


Structural studies of P-type ATPase-ligand complexes using an X-ray free-electron laser.

Bublitz M, Nass K, Drachmann ND, Markvardsen AJ, Gutmann MJ, Barends TR, Mattle D, Shoeman RL, Doak RB, Boutet S, Messerschmidt M, Seibert MM, Williams GJ, Foucar L, Reinhard L, Sitsel O, Gregersen JL, Clausen JD, Boesen T, Gotfryd K, Wang KT, Olesen C, Møller JV, Nissen P, Schlichting I - IUCrJ (2015)

Plot of free R factor against resolution after refinement (all atom coordinates, individual B factors) to 2.8 Å resolution against the original data set (dashed black lines) and hybrid data sets (red lines) with scrambled intensities. The arrows indicate the beginning of the scrambled data at (a) 5.82 Å, (b) 4.62 Å, (c) 4.04 Å, (d) 3.67 Å, (e) 3.41 Å, (f) 3.21 Å, (g) 3.04 Å and (h) 2.91 Å. These cutoffs were chosen in accordance with the resolution shells used by phenix.refine to calculate the Rfree values of the model.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4491313&req=5

fig3: Plot of free R factor against resolution after refinement (all atom coordinates, individual B factors) to 2.8 Å resolution against the original data set (dashed black lines) and hybrid data sets (red lines) with scrambled intensities. The arrows indicate the beginning of the scrambled data at (a) 5.82 Å, (b) 4.62 Å, (c) 4.04 Å, (d) 3.67 Å, (e) 3.41 Å, (f) 3.21 Å, (g) 3.04 Å and (h) 2.91 Å. These cutoffs were chosen in accordance with the resolution shells used by phenix.refine to calculate the Rfree values of the model.
Mentions: As a control for the significance of the information in the low-multiplicity 2.8 Å resolution data, we generated data sets with randomly scrambled reflection intensities ranging from a wide shell of scrambled intensities (5.8–2.8 Å) to a narrow shell (2.9–2.8 Å) at high resolution only. We compared the statistics of refinement against these data with those derived from refinement against the unperturbed data. As seen in Fig. 3 ▸, the introduction of the scrambled data was clearly detectable as a deviation point in Rfree values, even if only the highest resolution shell between 2.9 and 2.8 Å had been scrambled. This shows that the SFX intensities contain consistent signal. As a further analysis of the quality of the higher resolution data, we performed refinement of scrambled models to challenge the benefit and phase bias of a pre-refined model. Here, however, we found that the model refinement could not converge towards an improved model. Thus, the higher resolution data only improve map features when high-quality prior phases are available.

Bottom Line: The data reveal the binding sites of a variety of ligands, including lipids and inhibitors such as the hallmark P-type ATPase inhibitor orthovanadate.Even at relatively low resolution and multiplicity, the identification of ligands can be demonstrated.This makes SFX a useful tool for ligand screening and thus for unravelling the molecular mechanisms of biologically active proteins.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Molecular Biology and Genetics, Centre for Membrane Pumps in Cells and Disease - PUMPkin, Danish National Research Foundation, Aarhus University , Gustav Wieds Vej 10c, 8000 Aarhus C, Denmark.

ABSTRACT
Membrane proteins are key players in biological systems, mediating signalling events and the specific transport of e.g. ions and metabolites. Consequently, membrane proteins are targeted by a large number of currently approved drugs. Understanding their functions and molecular mechanisms is greatly dependent on structural information, not least on complexes with functionally or medically important ligands. Structure determination, however, is hampered by the difficulty of obtaining well diffracting, macroscopic crystals. Here, the feasibility of X-ray free-electron-laser-based serial femtosecond crystallography (SFX) for the structure determination of membrane protein-ligand complexes using microcrystals of various native-source and recombinant P-type ATPase complexes is demonstrated. The data reveal the binding sites of a variety of ligands, including lipids and inhibitors such as the hallmark P-type ATPase inhibitor orthovanadate. By analyzing the resolution dependence of ligand densities and overall model qualities, SFX data quality metrics as well as suitable refinement procedures are discussed. Even at relatively low resolution and multiplicity, the identification of ligands can be demonstrated. This makes SFX a useful tool for ligand screening and thus for unravelling the molecular mechanisms of biologically active proteins.

No MeSH data available.


Related in: MedlinePlus