2023 年 23 巻 p. 14-25
The fragment molecular orbital (FMO) method enables quantum mechanical calculations for macromolecules by dividing the target into fragments. However, most calculations, even for metalloproteins, have been performed by removing metal ions from the structures registered in the Protein Data Bank (PDB). For more realistic and useful calculations, FMO calculations must be performed without removing the metal ions. In this study, we discuss the results obtained from FMO calculations performed using 6-31G* and model core potentials (MCPs) for metal proteins containing Zn and Mg ions. Subsequently, we analyze the differences in atomic charges and interactions.
In order to properly treat metalloprotein in fragment molecular orbital (FMO) calculations, we investigated the application with/without the model core potentials (MCPs)1,2 to metal ions in SARS-CoV-2 helicase (PDBID:7NN0)3. Helicases are important proteins that unwind double-stranded DNA into single strands. We have previously performed 60 calculations on helicase 4 and 60 results using auto-FMO protocol have been published in the FMODB (https://drugdesign.riken.jp/FMODB/)5,6; however, the calculations were performed with all metal ions removed. Therefore, in this study, FMO calculations were performed without removing the three Zn ions and one Mg ion contained in the complex between SARS-CoV-2 helicase and a phosphoaminophosphonic acid-adenylate ester (ANP). For basis functions, in addition to the standard 6-31G*, we used MCPs1,2 with a fixed number of inner-shell electrons for metal atoms. Subsequently, the atomic charges7 and pair interaction energy decomposition analyses (PIEDA)8,9 of each metal ion were compared.
Structural preprocessing was performed for the SARS-CoV-2-related proteins in a previous study4. The helicase complex was obtained by X-ray crystallographic data with a resolution of 3.04 Å (PDBID: 7NN0). The protein structure is shown in Figure 1. First, the complexes were prepared using “Structural Preparation” and “Protonate 3D” functions in the Molecular Operating Environment (MOE, v2020.09; Chemical Computing Group Inc., Montreal, QC, Canada), where missing atoms completion and hydrogen addition were also performed. Structural optimization was performed using AMBER10:EHT force field, where the restraint condition was that all heavy atoms were constrained with a tether weight (1.0 kcal/Å) and all hydrogen atoms unconstrained. The resulting structure is shown in Figure 1.
Figure 1. Structure of the SARS-CoV-2 Helicase and ANP complex including Zn2+ and Mg2+ ions (PDBID:7NN0)
The fragmentation was the same as that in the conventional FMO calculation, i.e., amino acid unit fragmentation. However, there were three Zn2+ ions and one Mg2+ ion in the PDB entry 7NN0. Considering the coordination bonds, the fragments around the four metal ions were reconstructed as shown in Figure 2. ZN702 contains two histidine residues (HIS33 and HIS39) and two coordinated cysteine residues (CYS16 and CYS19). The side chains of these four amino acid residues and ZN702 merged to form a single fragment (Figure 2(a)). Similarly, ZN703 and the side chains of CYS5, CYS8, CYS26, and CYS29 combined to form a single fragment (Figure 2(b)). ZN704 combined with the side chains of CYS50, CYS55, CYS72, and HIS75 to form a single fragment (Figure 2(c)). MG705 is adjacent to the 3-phosphate portion of the ligand ANP701 and three H2O molecules. After fragmenting ANP701 at the 3-phosphate moiety, MG701, 3-phosphate, and three H2O molecules combined to form a single fragment (Fig. 2(d)). As an example of the fragmentation scheme with bond detached atoms (BDAs) and bond attached atoms (BAAs), the constructions of fragments around ZN702 are shown in Figure 3.
Figure 2. Fragmentation around (a) ZN702, (b) ZN703, (c) ZN704, and (d) MG705. Atoms of the same color belong to the same fragment
Figure 3. Fragmentation around Zn2+ (702) with clearly stated BDA and BAA
FMO calculations were performed using ABINIT-MP, an application for quantum chemical calculations based on the fragment molecular orbital (FMO) method10,11. Based on FMO calculation at MP2/6-31G* level, two patterns of metal treatment were investigated: (1) metal ions were also applied to 6-31G*, and (2) MCPs were only applied to metal ions. The MCPs were used to read a file downloaded from ref. 12. In ABINIT-MP, multiple downloaded MCP files of Zn2+ and Mg2+ ions can be merged into a single file, and the MCPs can be applied to specific elements by specifying it in the input file, as shown in Figure 4. The format of the MCP file to be loaded into ABINIT-MP is described in the user manual. All calculations were performed on a "Fugaku" supercomputer at RIKEN. The FMO data of the complex with and without MCPs has been registered in the FMODB with the codes (FMODB ID) as 43QQN and Q8Q1Y, respectively.
Figure 4. Specifying MCPs in ABINIT-MP input file
The natural bond orbital (NBO) charges for each metal ion, calculated for 6-31G* and MCPs, are listed in Table 1. The coordination bond lengths of each metal ion and its surrounding fragments are shown in Table 2. The average atomic charge of the three Zn2+ ions was 1.558 e for 6-31G* and 1.923 e for the MCPs. In contrast, for the Mg2+ ions, the atomic charge was 1.662 e for 6-31G* and 1.995 e for MCP. For both metal ions, there was more electron influx from the surrounding fragments in 6-31G* than in the MCPs, moving them from a positively charged state closer to a neutral charge. These results were obtained because the MCPs replace the inner-shell electrons with a model nucleus, which suppresses charge transfer, unlike 6-31G*. Among all the metal ions, Zn2+ (703) showed the largest charge transfer, with a change of 0.525 e for 6-31G* and 0.108 e for MCPs.
Table 1. Natural bond orbital charges for each metal ion in 6-31G* and MCPs
|
Metal Ion (Residue number) |
Natural bond orbital charge [e] | |
| All 6-31G* | MCP | |
| Zn2+ (702) | 1.650 | 1.953 |
| Zn2+ (703) | 1.475 | 1.892 |
| Zn2+ (704) | 1.549 | 1.924 |
| Mg2+ (705) | 1.662 | 1.995 |
Table 2. Coordination bond lengths between each metal ion and its surrounding fragments
|
Metal Ion (Residue number) |
Distance （Å） | |
|---|---|---|
| Zn2+ (702) | SG (CYS16) | 1.921 |
| SG (CYS19) | 1.914 | |
| ND1 (His36) | 1.931 | |
| ND1 (His39) | 1.939 | |
| Zn2+ (703) | SG (CYS5) | 2.012 |
| SG (CYS8) | 1.968 | |
| SG (CYS26) | 1.983 | |
| SG (CYS29) | 1.959 | |
| Zn2+ (704) | SG (CYS50) | 1.959 |
| SG (CYS55) | 1,930 | |
| SG (CYS72) | 1.957 | |
| ND1 (HIS75) | 2.050 | |
| Mg2+ (705) | O1G (ANP701) | 1.786 |
| O2B (ANP701) | 1.897 | |
| O (HOH801) | 2.025 | |
| O (HOH803) | 2.023 | |
Because atomic charge transfer can cause changes in interactions, we analyzed the total inter-fragment interaction energy (IFIE) and conducted PIEDA for fragments containing a metal ion. The interaction energies of Zn2+ (702) fragment with each amino acid residue of the helicase are listed in Table 3, where differences of IFIEs between 6-31G* and MCP greater than ±0.2 kcal/mol are enumerated. The most significant change is in the interaction with Zn2+(702), with a difference of −1.1 kcal/mol. A change in the interaction was also observed for the charged amino acids ARG22 and ARG21. This was attributed to changes in the ES associated with atomic charge transfer. In Table 3, there are neutrally charged amino acids with a total IFIE greater than 100 kcal/mol between them and Zn2+(702). These include LYS40, GLY17, and ILE20. As shown in Figure 2(a), these are fragments neighboring the amino acids (HIS39, CYS16, and CYS19) that merge with Zn2+(702). This combination with Zn2+(702) was only performed for the side chains, and the main-chain moiety was a small independent fragment. The interaction between this small fragment with BDA and fragments with BAA (LYS40, GLY17, and ILE20; Figure 3) was significantly large, but this may be an artifact of fragmentation. Although it is not shown in Table 3, due to the small difference between 6-31G* and MCP, VAL34 is also similar to the three fragments above, with an IFIE of 133.0 kcal/mol with Zn2+(702). In ordinary fragmentation, the interactions between fragments involving the splitting of covalent bonds are not considered12,13. These results suggest that when metal ions are treated through special fragmentation, as in this study, their interactions with the fragments in the second neighbor to fragments containing a metal ion must be ignored. Here, the second neighbor fragment is a sequence behind one of the amino acid residues whose side chain is coordinated to the metal ion and an amino acid residue covalently bonded to the residue (Figure 3). The 3D structure around Zn702, including Arg21, Arg22, Val34, and Lys40, is shown in Figure 5. The results for the other fragments containing a metal ion are listed in Tables 4–6, and the values of fragments that should be ignored are shown in parentheses. Differences were observed in the interactions of each metal ion with other metal ions and charged amino acid residues. These changes are reasonable given the change in the atomic charges of the metal ions.
Table 3. Total IFIE and PIEDA of Zn2+ ion (702) fragment with each amino acid residue of the helicase
ES, EX, CT+mix, and DI in the table represent electrostatic, exchange repulsion, charge transfer, and dispersion interactions, respectively. The upper half represents each value calculated by 6-31G*. The lower half shows the results of each value calculated with 6-31G* minus those calculated with MCP. IFIEs with a difference between 6-31G* and MCP greater than ±0.2 kcal/mol were enumerated. Numbers with () are values that should be ignored (see text).
| 6-31G* | IFIE [kcal/mol] | PIEDA [kcal/mol] | Charge transfer [e] |
Distance [Å]† |
||||
|---|---|---|---|---|---|---|---|---|
| Fragment | Total | ES | EX | CT+mix | DI | q(I->J) | Main | |
| ZN2703 | 1.4 | 1.4 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 4.59 |
| ARG22 | -12.3 | -11.3 | 0.4 | -0.6 | -0.9 | 0.000 | ES | 2.76 |
| ARG21 | -8.6 | -7.7 | 0.3 | -0.5 | -0.7 | -0.003 | ES | 2.57 |
| VAL42 | 6.7 | 7.7 | 0.1 | -0.3 | -0.9 | 0.002 | ES | 4.25 |
| PHE24 | 0.7 | 1.0 | 0.0 | 0.0 | -0.2 | 0.000 | ES | 4.06 |
| ASP32 | -6.7 | -6.4 | 0.0 | -0.2 | -0.1 | 0.000 | ES | 4.65 |
| CYS30 | -4.2 | -2.2 | 2.3 | -2.1 | -2.2 | 0.016 | EX | 2.65 |
| ASP59 | 0.4 | 0.4 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 8.60 |
| ASN107 | -20.2 | -21.7 | 8.8 | -3.9 | -3.5 | 0.050 | ES | 1.81 |
| PRO23 | -5.3 | -5.2 | 5.5 | -2.6 | -3.0 | 0.003 | EX | 4.59 |
| ILE35 | 0.4 | 0.7 | 0.0 | -0.1 | -0.2 | 0.000 | ES | 2.08 |
| LYS28 | 0.5 | 0.5 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 4.21 |
| ASP113 | -2.9 | -2.9 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 7.60 |
| VAL34 | 133.0 | 23.0 | 19.9 | 94.5 | -4.5 | 0.101 | CT+mix | 7.22 |
| LEU43 | -2.2 | -2.2 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 1.54 |
| LYS40 | 143.7 | 27.6 | 16.5 | 102.2 | -2.5 | 0.116 | CT+mix | 5.83 |
| ARG15 | -7.7 | -7.3 | 0.0 | -0.1 | -0.3 | 0.000 | ES | 1.54 |
| THR111 | -14.7 | -10.7 | 2.3 | -2.4 | -3.9 | 0.024 | ES | 4.63 |
| ALA110 | 0.9 | 1.8 | 4.7 | -2.3 | -3.2 | 0.003 | EX | 2.57 |
| GLY17 | (131.7) | (19.6) | (20.2) | (98.2) | (-6.4) | (0.143) | CT+mix | 2.46 |
| ILE20 | (140.5) | (15.7) | (21.5) | (104.6) | (-1.4) | (0.093) | CT+mix | 1.53 |
| Difference in each value of the MCP data from the 6-31G* data | ||||||||
| Fragment | Total | ES | EX | CT+mix | DI | q(I->J) | ||
| ZN2703 | -1.1 | -1.1 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| ARG22 | -0.8 | -1.0 | 0.0 | 0.1 | 0.1 | 0.000 | ||
| ARG21 | -0.8 | -0.7 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| VAL42 | -0.7 | -0.5 | 0.0 | 0.0 | -0.2 | 0.000 | ||
| PHE24 | -0.4 | -0.4 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| ASP32 | -0.4 | -0.4 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| CYS30 | -0.4 | -0.3 | 0.0 | 0.0 | -0.1 | 0.000 | ||
| ASP59 | -0.2 | -0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| ASN107 | -0.2 | -0.2 | 0.0 | 0.0 | 0.0 | 0.001 | ||
| PRO23 | 0.2 | -0.2 | 0.2 | 0.3 | -0.1 | 0.000 | ||
| ILE35 | 0.2 | 0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| LYS28 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| ASP113 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| VAL34 | 0.3 | 0.3 | 0.0 | 0.0 | -0.1 | 0.001 | ||
| LEU43 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| LYS40 | 0.4 | 0.1 | -0.1 | 0.2 | 0.1 | 0.002 | ||
| ARG15 | 0.4 | 0.4 | 0.0 | 0.0 | 0.1 | 0.000 | ||
| THR111 | 0.5 | 0.3 | 0.0 | 0.2 | 0.0 | 0.001 | ||
| ALA110 | 0.8 | 0.1 | 0.3 | 0.6 | -0.1 | 0.001 | ||
| GLY17 | (1.9) | (2.0) | (1.3) | (-1.0) | (-0.4) | (0.005) | ||
| ILE20 | (4.4) | (-4.5) | (-0.3) | (6.2) | (2.9) | (-0.003) | ||
† Distance between nearest neighbor atoms of a fragment pair.
Figure 5. 3D structure diagram around Zn702 showing the positions of the amino acids, in particular Arg21, Arg22, Val34, and Lys40
Table 4. Total IFIE and PIEDA of Zn2+ ion (703) fragment with each amino acid residue of the helicase
ES, EX, CT+mix, and DI in the table represent electrostatic, exchange repulsion, charge transfer, and dispersion interactions, respectively. The upper half represents each value calculated by 6-31G*. The lower half shows the results of each value calculated with 6-31G* minus those calculated with MCP
IFIEs with a difference between 6-31G* and MCP greater than ±0.2 kcal/mol were enumerated. Numbers with () are values that should be ignored (see text).
| 6-31G* | IFIE [kcal/mol] | PIEDA [kcal/mol] | Charge transfer [e] |
Distance [Å]† |
||||
|---|---|---|---|---|---|---|---|---|
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | Main | |
| ZN1702 | 1.4 | 1.4 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 4.59 |
| GLY99 | -22.5 | -23.7 | 9.7 | -5.1 | -3.5 | 0.004 | ES | 2.57 |
| GLN11 | 5.0 | 5.5 | 1.6 | -0.9 | -1.2 | 0.009 | ES | 2.74 |
| LYS28 | (-104.8) | (-102.0) | (2.2) | (-2.2) | (-2.9) | -0.013 | ES | 3.15 |
| GLU136 | 38.6 | 38.6 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 13.67 |
| THR12 | 6.8 | 8.5 | 0.6 | -0.8 | -1.4 | 0.005 | ES | 3.03 |
| VAL98 | -15.1 | -13.3 | 0.7 | -0.9 | -1.7 | -0.002 | ES | 2.55 |
| PHE24 | -1.8 | -1.6 | 0.0 | 0.0 | -0.2 | 0.000 | ES | 4.75 |
| LEU25 | 5.3 | 7.1 | 4.7 | -2.9 | -3.6 | 0.003 | ES | 2.15 |
| VAL2 | -46.0 | -46.0 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 8.95 |
| CYS97 | 1.2 | -11.4 | 29.2 | -10.6 | -6.0 | -0.052 | EX | 1.99 |
| ASP101 | 57.5 | 57.5 | 0.0 | 0.0 | 0.0 | 0.001 | ES | 4.68 |
| ARG129 | -60.6 | -60.6 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 7.11 |
| ASP32 | 71.8 | 72.3 | 0.0 | -0.1 | -0.4 | 0.000 | ES | 4.78 |
| VAL6 | (118.8) | (12.1) | (19.4) | (93.1) | (-5.8) | 0.117 | CT+mix | 1.54 |
| SER10 | -11.0 | -11.0 | 12.5 | -6.8 | -5.6 | -0.002 | EX | 1.75 |
| ASN9 | (105.3) | (-4.0) | (21.0) | (94.6) | (-6.1) | 0.082 | CT+mix | 1.54 |
| LEU7 | -22.9 | -23.1 | 16.2 | -8.1 | -7.9 | 0.014 | ES | 2.41 |
| CYS27 | (120.8) | (3.2) | (21.0) | (102.0) | (-5.4) | 0.107 | CT+mix | 1.53 |
| Difference in each value of the MCP data from the 6-31G* data | ||||||||
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | ||
| ZN1702 | -1.1 | -1.1 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| GLY99 | -1.0 | -1.7 | 1.4 | -0.5 | -0.1 | 0.012 | ||
| GLN11 | -0.9 | -0.8 | 0.0 | 0.0 | -0.1 | 0.000 | ||
| LYS28 | (-0.7) | (-1.7) | (0.7) | (0.3) | (0.0) | (0.006 | ||
| GLU136 | -0.2 | -0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| THR12 | 0.2 | 0.1 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| VAL98 | 0.3 | -0.2 | 0.1 | 0.4 | 0.0 | 0.002 | ||
| PHE24 | 0.3 | 0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| LEU25 | 0.3 | -0.4 | 0.7 | 0.2 | -0.2 | 0.008 | ||
| VAL2 | 0.4 | 0.4 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| CYS97 | 0.4 | -1.3 | 2.0 | -0.4 | 0.2 | 0.009 | ||
| ASP101 | 0.5 | 0.2 | 0.0 | 0.3 | 0.0 | 0.001 | ||
| ARG129 | 0.6 | 0.6 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| ASP32 | 0.7 | 0.7 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| VAL6 | (0.9) | (1.5) | (0.4) | (-0.7) | (-0.3) | (0.004) | ||
| SER10 | 1.2 | 0.7 | -0.1 | 0.6 | 0.0 | 0.000 | ||
| ASN9 | (1.3 | (-0.1 | (-0.2 | (1.6 | (0.1 | -0.001 | ||
| LEU7 | 2.0 | -0.3 | 2.9 | -0.7 | 0.2 | 0.019 | ||
| CYS27 | (2.3) | (-0.9) | (1.1) | (1.5) | (0.6) | (0.000) | ||
† Distance between nearest neighbor atoms of a fragment pair.
Table 5. Total IFIE and PIEDA of Zn2+ ion (704) fragment with each amino acid residue of the helicase
ES, EX, CT+mix, and DI in the table represent electrostatic, exchange repulsion, charge transfer, and dispersion interactions, respectively. The upper half represents each value calculated by 6-31G*. The lower half shows the results of each value calculated with 6-31G* minus those calculated with MCP. IFIEs with a difference between 6-31G* and MCP greater than ±0.2 kcal/mol were enumerated. Numbers with () are values that should be ignored (see text).
| 6-31G* | IFIE (kcal/mol) | PIEDA [kcal/mol] | Charge transfer [e] |
Distance [Å]† |
||||
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | Main | |
| LYS76 | (96.3) | (-17.4) | (16.3) | (100.1) | (-2.8) | (0.090) | CT+mix | 1.54 |
| THR58 | -3.7 | -2.8 | 0.2 | -0.3 | -0.8 | 0.005 | ES | 3.20 |
| TYR64 | -8.9 | -8.7 | 0.0 | 0.0 | -0.2 | 0.000 | ES | 5.31 |
| GLY54 | -2.1 | -1.7 | 0.0 | -0.1 | -0.3 | 0.000 | ES | 4.48 |
| ASP59 | 26.3 | 26.3 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 6.05 |
| TYR70 | -2.9 | -2.9 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 5.87 |
| VAL49 | -1.0 | -0.4 | 0.0 | -0.2 | -0.4 | 0.000 | ES | 3.69 |
| PRO53 | -2.4 | -2.1 | 0.4 | 0.3 | -1.0 | 0.005 | ES | 3.10 |
| TYR71 | 0.8 | 1.2 | 5.4 | -2.4 | -3.4 | -0.004 | EX | 2.80 |
| LEU63 | 4.8 | 5.7 | 0.8 | -0.5 | -1.3 | 0.000 | ES | 2.63 |
| ASP56 | (180.9) | (66.0) | (22.0) | (98.7) | (-5.8) | (0.134) | CT+mix | 1.53 |
| ASN51 | (121.0) | (8.0) | (19.3) | (98.4) | (-4.6) | (0.115) | CT+mix | 1.54 |
| ALA52 | -15.1 | -15.9 | 15.6 | -7.6 | -7.3 | -0.006 | ES | 2.18 |
| LYS73 | (92.0) | (-27.8) | (20.4) | (104.1) | (-4.8) | (0.104) | CT+mix | 1.53 |
| Difference in each value of the MCP data from the 6-31G* data | ||||||||
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | ||
| LYS76 | (-0.5) | (-0.9) | (-0.1) | (0.4) | (0.1) | (0.001) | ||
| THR58 | -0.5 | -0.7 | 0.0 | 0.3 | -0.1 | 0.001 | ||
| TYR64 | -0.2 | -0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| GLY54 | 0.2 | 0.1 | 0.0 | 0.0 | 0.1 | 0.000 | ||
| ASP59 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| TYR70 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| VAL49 | 0.3 | 0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| PRO53 | 0.5 | -0.1 | 0.0 | 0.5 | 0.1 | 0.001 | ||
| TYR71 | 0.5 | -0.6 | 1.0 | 0.5 | -0.4 | 0.004 | ||
| LEU63 | 0.7 | 0.5 | 0.0 | 0.2 | 0.0 | 0.000 | ||
| ASP56 | (0.9) | (0.7) | (0.0) | (0.3) | (-0.1) | (0.001 | ||
| ASN51 | (1.1) | (1.1) | (0.7) | (-0.9) | (0.2) | 0.001 | ||
| ALA52 | 2.2 | 0.9 | 1.8 | -0.5 | 0.1 | -0.002 | ||
| LYS73 | (2.8) | (-3.1) | (0.6) | (4.0) | (1.3) | (0.001 | ||
† Distance between nearest neighbor atoms of a fragment pair.
Table 6. Total IFIE and PIEDA of Mg2+ ion (705) fragment with each amino acid residue of the helicase
ES, EX, CT+mix, and DI in the table represent electrostatic, exchange repulsion, charge transfer, and dispersion interactions, respectively. The upper half represents each value calculated by 6-31G*. The lower half shows the results of each value calculated with 6-31G* minus those calculated with MCP. IFIEs with a difference between 6-31G* and MCP greater than ±0.2 kcal/mol were enumerated. Numbers with () are values that should be ignored (see text).
| 6-31G* | IFIE [kcal/mol] | PIEDA [kcal/mol] | Charge transfer [e] |
Distance [Å]† |
||||
|---|---|---|---|---|---|---|---|---|
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | Main | |
| ASP374 | 0.3 | 4.0 | 27.2 | -18.9 | -12.0 | 0.244 | EX | 1.62 |
| SER289 | -53.6 | -79.0 | 49.7 | -18.2 | -6.1 | 0.024 | ES | 1.96 |
| GLU375 | 30.6 | 13.1 | 43.1 | -17.2 | -8.5 | 0.149 | EX | 1.59 |
| SER539 | -8.2 | -16.7 | 13.2 | -0.8 | -3.7 | 0.054 | ES | 1.72 |
| GLU540 | 102.3 | 103.0 | 2.3 | -1.3 | -1.7 | 0.017 | ES | 2.46 |
| ASP401 | 85.7 | 86.0 | 0.0 | 0.0 | -0.3 | 0.000 | ES | 4.60 |
| ALA316 | 0.3 | 1.0 | 0.0 | -0.2 | -0.5 | 0.002 | ES | 3.21 |
| GLY287 | -21.9 | -22.6 | 8.1 | -3.5 | -3.9 | -0.023 | ES | 2.29 |
| GLN404 | -13.1 | -16.6 | 9.7 | -3.1 | -3.2 | -0.039 | ES | 1.85 |
| LYS569 | -59.5 | -59.5 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 6.25 |
| THR286 | -7.9 | -14.4 | 1.8 | 7.2 | -2.5 | 0.009 | ES | 2.36 |
| ARG442 | -67.0 | -66.8 | 0.0 | 0.0 | -0.2 | 0.000 | ES | 3.59 |
| LYS465 | -58.7 | -58.7 | 0.0 | 0.0 | 0.0 | 0.000 | ES | 7.34 |
| LYS320 | -153.2 | -156.3 | 13.7 | -5.2 | -5.4 | -0.068 | ES | 2.19 |
| GLY285 | -27.4 | -33.0 | 18.2 | -6.3 | -6.3 | -0.043 | ES | 1.70 |
| GLY538 | -12.3 | -11.2 | 5.5 | -4.3 | -2.4 | -0.028 | ES | 2.62 |
| ARG567 | -144.3 | -158.4 | 27.1 | -7.0 | -5.9 | -0.082 | ES | 1.74 |
| HIS290 | -11.8 | -17.7 | 11.5 | -2.5 | -3.1 | -0.019 | ES | 1.89 |
| ARG443 | -196.7 | -213.8 | 43.8 | -14.0 | -12.7 | -0.104 | ES | 1.70 |
| Difference in each value of the MCP data from the 6-31G* data | ||||||||
| Fragment | total | ES | EX | CT+mix | DI | q(I->J) | ||
| ASP374 | -9.0 | 1.4 | 0.4 | -5.7 | -5.0 | 0.117 | ||
| SER289 | -5.5 | -11.4 | 9.2 | -6.9 | 3.6 | 0.013 | ||
| GLU375 | -2.6 | -0.9 | 0.1 | -0.8 | -1.0 | 0.012 | ||
| SER539 | -0.8 | -0.8 | -0.1 | 0.3 | -0.8 | 0.004 | ||
| GLU540 | -0.8 | -0.7 | 0.0 | 0.0 | -0.1 | 0.000 | ||
| ASP401 | -0.3 | -0.3 | 0.0 | 0.0 | -0.1 | 0.000 | ||
| ALA316 | -0.3 | 0.0 | 0.0 | 0.0 | -0.2 | 0.001 | ||
| GLY287 | 0.2 | 0.3 | 0.0 | 0.0 | -0.1 | 0.001 | ||
| GLN404 | 0.2 | 0.2 | 0.0 | 0.0 | 0.0 | 0.001 | ||
| LYS569 | 0.2 | 0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| THR286 | 0.2 | 0.3 | 0.0 | -0.1 | 0.0 | 0.000 | ||
| ARG442 | 0.3 | 0.2 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| LYS465 | 0.3 | 0.3 | 0.0 | 0.0 | 0.0 | 0.000 | ||
| LYS320 | 0.3 | 0.1 | 0.1 | 0.1 | 0.1 | 0.004 | ||
| GLY285 | 0.4 | 0.4 | 0.0 | 0.0 | -0.1 | 0.002 | ||
| GLY538 | 0.4 | -0.1 | 1.1 | -0.9 | 0.4 | -0.004 | ||
| ARG567 | 1.1 | 1.0 | 0.0 | 0.0 | 0.0 | 0.003 | ||
| HIS290 | 1.6 | -1.0 | -0.1 | 1.1 | 1.7 | 0.018 | ||
| ARG443 | 2.0 | 1.6 | 0.3 | -0.1 | 0.2 | 0.003 | ||
† Distance between nearest neighbor atoms of a fragment pair.
The computation time for 6-31G* was 10590.3 s (approximately 3 h), using 48 nodes of "Fugaku" at RIKEN, whereas that of the MCPs was 10572.6 s, using the same computing resources; thus, there was no significant difference in the computation times between them.
As mentioned above, the MCPs works in the direction of suppressing atomic charge transfers. Therefore, we also attempted a calculation in which fragmentation was performed as usual and all metal ions were treated as independent fragments. However, the calculations did not converge in either the 6-31G* or MCPs case. Because the calculation of fragment dimers containing Zn2+ ions that form coordination bonds did not converge, the coordination bonds should be merged to form a single fragment, as in this paper, rather than fragmenting the coordination bonds.
We performed FMO calculations for the SARS-CoV-2 helicase and ANP complex including Zn2+ and Mg2+ ions (PDBID: 7NN0) using 6-31G* and MCP basis functions for metal ions. Then the difference in the atomic charge of the metal ions between the isolated and complex forms was milder with the MCPs. MCP seems to be more consistent with chemically intuition, since the atomic charges are closer to the valence of the ions, but it is difficult to determine which is more desirable. Results also revealed that the difference in the interaction energies with the change in the atomic charge due to 6-31G* and MCP basis functions was generally less than 1 kcal/mol and was not affect the qualitative discussion of interfragment interactions.
This study was conducted as a part of the FMO Drug Design Consortium (FMODD). We thank Professor Yuji Mochizuki of Rikkyo University and Drs. Tatsuya Nakano and Yoshio Okiyama of the National Institute of Health Sciences for providing the Fugaku version of ABINIT-MP and for general discussions on the FMO calculations. We also thank Drs. Teruki Honma, Daisuke Takaya, and Kikuko Kamisaka at RIKEN for their helpful comments and technical assistance regarding FMODB registration. The FMO calculations were performed using Fugaku (Project ID: hp220143) at RIKEN. This work was supported in part by the Japan Agency for Medical Research and Development (AMED) under the Drug Discovery and Life Science Research Support Platform Project (BINDS) (Grant No. JP22ama121030). CW acknowledges the JST PRESTO Research Grant (JPMJPR18GD).
