The enzyme model is constructed based on the X-ray structures
for the E. coli Alkaline Phosphatase (AP) mutant
R166S with bound inorganic phosphate at 2.05 Å resolution (PDB
code 3CMR(122 (link))). The substrate methyl p-nitrophenyl
phosphate (MpNPP–) is first “mutated”
to the orientation with the −OMe group oriented toward the
magnesium ion (denoted as α orientation) starting from the PDB
structure. All basic and acidic amino acids are kept in their physiological
protonation states except for Ser102 in AP, which is accepted to be
the nucleophile and deprotonated in the reactive complex. Water molecules
are added following the standard protocol of superimposing the system
with a water droplet of 25 Å radius centered at Zn12+ (see Figure6 for atomic labels) and removing
water molecules within 2.8 Å from any heavy atoms resolved in
the crystal structure. GSBP is used to treat long-range electrostatic
interactions in MD simulations. In the QM/MM simulations, as described
in details in our previous work,48 (link),49 (link) the QM region
includes the two zinc ions and their six ligands (Asp51, Asp369, His370,
Asp327, His412, His331), Ser102, and the substrate MpNPP–. Only side chains of protein residues are included in the QM region,
and link atoms are added between Cα and Cβ atoms. A NOE potential is added to the C–O bonds in Asp51
in AP to prevent artificial polarization.48 (link) The integration time step is 1 fs, and all bonds involving hydrogens
are constrained using SHAKE. The DFTB3/MM results are also compared
to MM simulations using a conventional nonbonded zinc model14 (link) (referred to as a Coulomb scheme) or short–long
effective functions (SLEF1)25 (link) model developed
by Zhang and co-workers. Protein atoms in the MM region are described
by the all-atom CHARMM22 force field, and water molecules are described
with the TIP3P model.
To further benchmark the applicability
of DFTB3/3OB to the reaction of interest, we also study an active
site model that includes all QM atoms in the QM/MM enzyme model. The
Cβ carbons are fixed at their positions in the crystal
structure during geometry optimization; the positions of the link
atoms used to saturate the valence of the Cβ atoms
in the active site model are also fixed. The reactant (Michaelis)
complex and transition state are located for MpNPP– (methyl p-nitrophenyl phosphate), MmNPP– (methyl m-nitrophenyl phosphate), and MPP– (methyl phenyl phosphate) using DFTB3 and B3LYP with the 6-31+G(d,p)
basis set. The minimum energy path (MEP) calculations are carried
out by one-dimensional adiabatic mapping at the DFTB3 level; the reaction
coordinate is the antisymmetric stretch involving the breaking and
forming P–O bonds. Subsequently, the saddle point is further
refined by conjugated peak refinement (CPR123 ) to obtain precise transition state structure. Single-point energy
calculations at DFTB3 and B3LYP geometries are then carried out using
B3LYP, M06,124 PBE, and MP2 methods using
a larger basis set of 6-311++G(d,p). The D3125 (link) dispersion corrections are added for B3LYP, M06, and PBE methods.
for the E. coli Alkaline Phosphatase (AP) mutant
R166S with bound inorganic phosphate at 2.05 Å resolution (PDB
code 3CMR(122 (link))). The substrate methyl p-nitrophenyl
phosphate (MpNPP–) is first “mutated”
to the orientation with the −OMe group oriented toward the
magnesium ion (denoted as α orientation) starting from the PDB
structure. All basic and acidic amino acids are kept in their physiological
protonation states except for Ser102 in AP, which is accepted to be
the nucleophile and deprotonated in the reactive complex. Water molecules
are added following the standard protocol of superimposing the system
with a water droplet of 25 Å radius centered at Zn12+ (see Figure
water molecules within 2.8 Å from any heavy atoms resolved in
the crystal structure. GSBP is used to treat long-range electrostatic
interactions in MD simulations. In the QM/MM simulations, as described
in details in our previous work,48 (link),49 (link) the QM region
includes the two zinc ions and their six ligands (Asp51, Asp369, His370,
Asp327, His412, His331), Ser102, and the substrate MpNPP–. Only side chains of protein residues are included in the QM region,
and link atoms are added between Cα and Cβ atoms. A NOE potential is added to the C–O bonds in Asp51
in AP to prevent artificial polarization.48 (link) The integration time step is 1 fs, and all bonds involving hydrogens
are constrained using SHAKE. The DFTB3/MM results are also compared
to MM simulations using a conventional nonbonded zinc model14 (link) (referred to as a Coulomb scheme) or short–long
effective functions (SLEF1)25 (link) model developed
by Zhang and co-workers. Protein atoms in the MM region are described
by the all-atom CHARMM22 force field, and water molecules are described
with the TIP3P model.
To further benchmark the applicability
of DFTB3/3OB to the reaction of interest, we also study an active
site model that includes all QM atoms in the QM/MM enzyme model. The
Cβ carbons are fixed at their positions in the crystal
structure during geometry optimization; the positions of the link
atoms used to saturate the valence of the Cβ atoms
in the active site model are also fixed. The reactant (Michaelis)
complex and transition state are located for MpNPP– (methyl p-nitrophenyl phosphate), MmNPP– (methyl m-nitrophenyl phosphate), and MPP– (methyl phenyl phosphate) using DFTB3 and B3LYP with the 6-31+G(d,p)
basis set. The minimum energy path (MEP) calculations are carried
out by one-dimensional adiabatic mapping at the DFTB3 level; the reaction
coordinate is the antisymmetric stretch involving the breaking and
forming P–O bonds. Subsequently, the saddle point is further
refined by conjugated peak refinement (CPR123 ) to obtain precise transition state structure. Single-point energy
calculations at DFTB3 and B3LYP geometries are then carried out using
B3LYP, M06,124 PBE, and MP2 methods using
a larger basis set of 6-311++G(d,p). The D3125 (link) dispersion corrections are added for B3LYP, M06, and PBE methods.