Computational insights into the selectivity mechanism of APP-IPover matrix metalloproteinases
Lingling Geng,† Jian Gao,† Wei Cui,† Yancheng Tang,‡ Mingjuan Ji,*,† and Bozhen Chen,*,†
†School of Chemistry and Chemical Engineering, Graduate University of Chinese Academy of Sciences, Beijing 100049, P. R. China
‡School of Basic Medical Sciences,Peking University Health Science Center, Beijing 100191, P. R. China
Corresponding authors.
Bozhen Chen: E-mail: . Telephone: 8610 88256129. Fax: 8610 88256093.
Mingjuan Ji: E-mail: . Telephone: 8610 88256326. Fax: 8610 88256093.
MMP-7Y214L/APP-IP system
The simulation of MMP-7Y214L/APP-IP system was studied using the same methods as mentioned in the ‘MATERIALS AND METHODS’ of the manuscript. The same MM/GBSA method was applied to calculate the binding free energy and free energy decomposition analysis.
The initial structureof MMP-7Y214L/APP-IP complex was constructed based on the wide type structure of MMP-7/APP-IP by mutating Y214L in MMP-7. Fig. S1 shows the RMSD values of theMMP-7Y214L/APP-IP complex and APP-IP during the 12 ns simulation. It is clear that the system is stable and the sampling from 8 to 12 ns is reasonable and reliable. The result of the binding free energyis shown in Table S1. Compared with the WT, the mutant gives a more favorable binding affinity (-56.45 kcal/mol), which implies that Y214L in MMP-7 improves the binding affinity and residue Tyr214 indeed affects the binding between MMP-7 and APP-IP. In detail, the van der Waals contribution is more crucial for distinguishing the binding affinities between the WT and MMP-7Y214L/APP-IP complexes. We also compared the structures of the WT and mutant complexes. After the mutation at Y214L, Tyr3APP-IPenters into the binding pocket and the binding mode of N-terminal of APP-IP has changed (see Fig. S2). The result of decomposition analysis is shown in Fig. S3. It is clear that residues Leu181, Leu214, Ala215, His218, Thr239 and Tyr240 in MMP-7Y214L are more favorable for APP-IP binding, while residues Phe103, Tyr172 and His228 in MMP-7Y214L are less favorable for the interaction. These results also indicate that our previous analysis regarding the difference between MMP-7/APP-IP and MMP-2/APP-IP complexes is reasonable.
(a)
(b)
Fig. S1 The root-mean-square deviation (rmsd) of the backbone atoms (CA, N, C) of MMP-7Y214L/APP-IP complex (a) and APP-IP (b) relative to the starting structure.
Table S1 Binding Free Energies and Individual Energy Terms of MMPs/APP-IP Calculated by MM/GBSA (kcal/mol)a
MMPs/APP-IP complexes / ΔEele / ΔEvdw / ΔGGB / ΔGSA / -TΔS / ΔGpredMMP-7/APP-IP / -71.22±3.24 / -67.52±4.26 / 69.86±3.05 / -9.88±0.40 / 30.32±4.17 / -48.44±4.16
MMP-7Y214L/APP-IP / -71.09±2.89 / -80.31±4.36 / 71.95±2.61 / -11.93±0.28 / 34.93±2.36 / -56.45±4.17
a ΔEele, electrostatic contribution; ΔEvdw, van der Waals contribution; ΔGGB, the polar contribution of desolvation; ΔGSA, nonpolar contribution of desolvation;–TΔS, the conformational entropy at temperature T.
Fig. S2Molecular surface of MMP-7Y214L with APP-IPMMP-7-Y214L (in blue) and APP-IPMMP-7 (in pink).
(a)(b)
Fig. S3 The comparisons of key residues of MMP-7 (a) and APP-IP (b) for MMP-7/APP-IP and MMP-7Y214L/APP-IP complexes.For MMP-7, only residues withenergy contribution differences larger than 0.5 kcal/molare shown.
Root-mean-square deviation (rmsd) of APP-IP in six systems
To explore the stability of the peptide, the RMSD values of the backbone atoms (CA, N, C) of APP-IP in the six complexes relative to the starting structureare calculated and shown in Fig. S4. It is clear that the RMSD values remain stable after the 8 ns MD simulation in all complexes. So it is equilibrated and reasonable to sampling from 8 to 12 ns.
Fig. S4Root-mean-square deviation (rmsd) of the backbone atoms (CA, N, C) of the APP-IP in six systems relative to the starting structure.
1