Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jul 26;7(1):6636.
doi: 10.1038/s41598-017-06290-0.

Conformational Preludes to the Latency Transition in PAI-1 as Determined by Atomistic Computer Simulations and Hydrogen/Deuterium-Exchange Mass Spectrometry

Affiliations
Free PMC article

Conformational Preludes to the Latency Transition in PAI-1 as Determined by Atomistic Computer Simulations and Hydrogen/Deuterium-Exchange Mass Spectrometry

Michael Petersen et al. Sci Rep. .
Free PMC article

Abstract

Both function and dysfunction of serine protease inhibitors (serpins) involve massive conformational change in their tertiary structure but the dynamics facilitating these events remain poorly understood. We have studied the dynamic preludes to conformational change in the serpin plasminogen activator inhibitor 1 (PAI-1). We report the first multi-microsecond atomistic molecular dynamics simulations of PAI-1 and compare the data with experimental hydrogen/deuterium-exchange data (HDXMS). The simulations reveal notable conformational flexibility of helices D, E and F and major fluctuations are observed in the W86-loop which occasionally leads to progressive detachment of β-strand 2 A from β-strand 3 A. An interesting correlation between Cα-RMSD values from simulations and experimental HDXMS data is observed. Helices D, E and F are known to be important for the overall stability of active PAI-1 as ligand binding in this region can accelerate or decelerate the conformational inactivation. Plasticity in this region may thus be mechanistically linked to the conformational change, possibly through facilitation of further unfolding of the hydrophobic core, as previously reported. This study provides a promising example of how computer simulations can help tether out mechanisms of serpin function and dysfunction at a spatial and temporal resolution that is far beyond the reach of any experiment.

Conflict of interest statement

The authors declare that they have no competing interests.

Figures

Figure 1
Figure 1
Structure of active and latent PAI-1. (a) crystal structure of the W175F mutant of PAI-1 in the active conformation (PDB 3Q02). β-sheet A, B and C are shown in red, blue and green, respectively. RCL residues missing in the PDB 3Q02 are sketched as a purple broken line. Other secondary structures and named regions are indicated in the figure. (b) Structure of wild-type PAI-1 in the latent conformation (PDB 1DVN). Color coding is the same as in (a). In this conformation the RCL (purple) is inserted as an extra β-strand in the middle of β-sheet A.
Figure 2
Figure 2
Localized dynamics in active and latent PAI-1. (a) The average Cα-RMSF values from the four replicate one-microsecond simulations on active PAI-1 and the two one-microsecond simulations on latent PAI-1 are indicated as bright red and blue lines, respectively. The average Cα-RMSF values +/− one standard deviation are shown for active and latent PAI-1 as pale red and pale blue lines, respectively. Secondary structural elements are depicted above the plot for reference. The RCL is colored orange, α-helices are colored grey, β-strands from β-sheet A, B and C are colored red, blue and green, respectively. The average Cα-RMSF values from the simulation on active PAI-1 and the simulation on latent PAI-1 are indicated on the structure of (b) active and (c) latent PAI-1 according to the color scale shown in (d). The range of this color scale is indicated by the grey area in (a).
Figure 3
Figure 3
Hydrogen/deuterium-exchange mass spectrometry on active and latent PAI-1 at 5 °C. 40 pmol of active PAI-1 was diluted 100-fold into deuterated PBS pD 7.4 and incubated for 0.5, 1, 2, 5, 10, 15, 20, 30, 40 and 80 minutes at 5 °C prior to proteolytic digestion and analysis by LCMS. Latent PAI-1 was analyzed in a similar manner although only after 1, 5 and 20 minutes incubation. All 1, 5 and 20 minute time points were conducted in triplicate. (a) Heat map representing the deuterium content of individual peptides relative to a full deuteration control (see methods) after 1, 5 and 20 minutes incubation. (b) The HDXMS heat map of active PAI-1 from (a) shown on the active PAI-1 structure (PDB 3Q02) with the W175F substitution and RCL modelled in using Modloop, . Grey areas represent sequences not covered by peptide HDX data. Similar representation of the latent PAI-1 HDXMS data is shown in Figure S5. (c) Heat map from (a) shown on residues 66-161 only. The upper right depiction is a color map to locate peptides for which deuterium uptake plots are shown in (d). (d) deuterium uptake plots of the indicated peptides. Active and latent PAI-1 are indicated by the blue and red line, respectively, and the full deuteration level by the black line. Error bars represent the standard deviation from triplicate measurements.
Figure 4
Figure 4
Comparison of simulation Cα-RMSF values with hydrogen/deuterium-exchange data. Amino acid specific Cα-RMSF values (from Fig. 2a) averaged for the residues in each of the peptides analyzed in the HDX experiment as well as relative deuterium uptake after 1 minute exposure to D2O of active and latent PAI-1 are plotted for active and latent PAI-1 in Figure S6. In the present figure the correlation between Cα-RMSF values and relative deuterium uptake for active (red) and latent (blue) peptides (taken from Figure S6) is investigated by plotting the two data sets on each axis. Trendlines and Pearson correlation coefficients ρ are shown.
Figure 5
Figure 5
Low frequency – large amplitude motions in active PAI-1. The per-residue, per-frame Cα-RMSD values relative to the average structure was calculated once every 200 ps for the four simulations on active PAI-1 (total of 20000 frames). (a) For each residue, the frames used for calculations were grouped in Cα-RMSD value bins of 0.1 Å size and the %-fraction of frames in each bin was plotted as histograms according to the indicated color scale. (b) Data for S37, L69 and M83 in active PAI-1 are shown as examples. The histograms were then collapsed to a 1-dimensional representation and rotated 90° counter clockwise to facilitate the construction of a density plot of the binned Cα-RMSD data as a function of PAI-1 residue number in (b). The per-residue average Cα-RMSD values are indicated by the black line.
Figure 6
Figure 6
W86-loop dynamics leads to progressive detachment of β2A from β3A. Example structures of W86-loop-β2A-β3 A configurations are shown from (a) the PDB:3Q02 crystal used as basis for simulations and (be) from the indenticated frames in the simulations on active PAI-1. The W86 sidechain and β2A and β3 A backbones are shown as sticks and remaining residues as cartoon. Residue G230 is colored yellow and the distance between Cα atoms of G230 and W86 is indicated by the black broken line. Hydrogen-bonds between β2 A and β3 A are indicated by dashed colored lines as follows: 172 O:90 N (black), 172 N:90 O (red), 170 O:92 N (blue), 170 N:92 O (green) and 168 O:94 N (yellow). (f) β2A-β3A hydrogen-bond lengths, according to the color coding in (a), are plotted as a function of simulation time consecutively for all simulations on active PAI-1, as indicated. (e) the distances between Cα atoms of G230 and W86 are plotted as a function of simulation time consecutively for all simulations on active PAI-1, as indicated.
Figure 7
Figure 7
Helix D and F dynamics. Example structures of helix D and F configurations are shown in panels (a–e and h–k), respectively, from the indicated crystal structure and frames in the simulations on active PAI-1. Structures are generally shown as white cartoon representation with the exception of the loop between helix F and β-strand 3 A which is displayed in red. Residues V157 and A318 are colored yellow and the distance between the Cα atoms of these residues is illustrated by a black broken line and plotted as a function of simulation time in Figure S9. The average hydrogen bond lengths of helix D (all hydrogen bonds between 70 O:74 N and 79 O:83 N) and helix F (all hydrogen bonds between 128 O:131 N and 141 O:145 N) are plotted as a function of simulation time consecutively for all simulations on active PAI-1 in panel (f and g), respectively.

Similar articles

See all similar articles

Cited by 1 article

References

    1. Gettins PG. Serpin structure, mechanism, and function. Chem Rev. 2002;102:4751–4804. doi: 10.1021/cr010170+. - DOI - PubMed
    1. Gooptu B, Lomas DA. Conformational pathology of the serpins: themes, variations, and therapeutic strategies. Annu Rev Biochem. 2009;78:147–176. doi: 10.1146/annurev.biochem.78.082107.133320. - DOI - PubMed
    1. Beauchamp NJ, et al. Antithrombins Wibble and Wobble (T85M/K): archetypal conformational diseases with in vivo latent-transition, thrombosis, and heparin activation. Blood. 1998;92:2696–2706. - PubMed
    1. Bruce D, Perry DJ, Borg JY, Carrell RW, Wardell MR. Thromboembolic disease due to thermolabile conformational changes of antithrombin Rouen-VI (187 Asn–>Asp) The Journal of clinical investigation. 1994;94:2265–2274. doi: 10.1172/JCI117589. - DOI - PMC - PubMed
    1. Hekman CM, Loskutoff DJ. Endothelial cells produce a latent inhibitor of plasminogen activators that can be activated by denaturants. J Biol Chem. 1985;260:11581–11587. - PubMed

Publication types

Substances

Feedback