Effect of variations in the conserved residues E371 and S359 on the structural dynamics of protein Z dependent protease inhibitor (ZPI): a molecular dynamic simulation study

Abstract

Protein Z (PZ) dependent protease inhibitor (ZPI) is a natural anticoagulant inhibiting blood coagulation proteases fXa and fXIa. Despite being a member of the serpin superfamily, it possesses unique structural features such as activation by PZ, regulating its inhibitory function. In order to understand the Reactive Centre Loop (RCL) dynamics of ZPI, which is absolutely critical for its activity, we performed Molecular Dynamics (MD) simulation on ZPI and its E371 and S359 variants located at important conserved functional sites. Unexpectedly, the RCL of E371 variants, (E371K, E371R, and E371Q), were shown to be very stable due to compensatory interactions at the proximal end of RCL. Interestingly, RCL flexibility was shown to be enhanced in the double mutant K318E-E371K due to the repulsive effect of increased negative charge on top of the breach region. Principal component analysis (PCA) coupled with residue wise interaction network analysis(RIN) revealed correlated motion between the RCL and the PZ binding regions in the WT. However, a loss of regulation in correlated motion between RCL and PZ binding hotspot Tyr240 in the double mutant was also observed. Additionally, the S359F and S359I mutations resulted in increased RCL flexibility owing to the disruption of stabilizing hydrogen bonding interaction at the distal end of strand S5A. Thus, the current study proposes that the overall stabilizing interactions of S5A is a major regulator of proper loop movement of ZPI for its activity. The results would be beneficial to engineer activity compromised ZPI as a prophylactic agent for the treatment of hemophilia.

Communicated by Ramaswamy H. Sarma

Keywords:

• Serpins
• anticoagulant
• protein Z-dependent protease inhibitor
• protein Z
• MD-simulation
• fXa

Acknowledgements

Authors wish to thank Dr Rajeswari Appadurai, IISC Bangalore, for her helpful comments and suggestions. The authors also acknowledge the computing facility of the P. G. Senapathy Centre, IIT Madras.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the funding from DST-SERB, Govt. of India, (Project no. ECR/2017/000447) and SSN trust (Project no. SSN/IFFP/January 2019/1-18/17)