115
chemical shift mapping and subsequent molecular docking ( Affinity and Discovery
modules of Insight II software, ESFF force field [116]), the two potential binding
were proposed. Site I includes amino acid residues Leu22, Phe25, Glu26, Asn28,
Ile33, Lys35, Asn37, site II—residues Lys60, Leu62, Gly64, Gly66 and Ala70, consequently. Both discovered sites are located on template DNA binding interface
of the 8-kDa domain, and SQMG incorporation into any of them would prevent
interaction with the DNA template [104]. The above-described binding sites for
LCA and its derivates, as well as for NA, are situated on the template DNA binding
interface, too [82, 114, 115]. Thus, one can conclude that all pol β inhibitors identified by industrious japanese researchers as interacting with 8-kDA domain have a
very similar structural mechanism of action. These compounds interact with the
DNA binding area and compete with the DNA template in a similar fashion.
Similar strategy combining spectroscopic and computational approaches was applied by Hazan et al. [93] to identify the pamoic acid binding site on the surface of
DNA polymerase β. The ability of pamoic acid to interact with the 8-kDa templateprimer binding domain was previously demonstrated by Hu et al. [78]. The molecular docking of pamoic acid to the 8 kDa domain of pol β was carried out using
AutoDock 3.0.5 software [117, 118].
Structures generated by AutoDock have been ranked according to their binding
energy and 100 lowest energy structures were selected for the further analysis. With
the force field used by AutoDock, the energy values for the best ligands varied from
− 9.58 to − 8.96 kcal/mol. Systematic analysis of the 100 best docked structures
revealed that all of them were located at a single site, although pamoic acid could
move freely around the 8 kDa domain during docking. Close atomic contacts between pairs of protein–ligand atoms (with a distance cutoff of 2 Å) were computed.
Nine residues—His34, Lys35, Asn37, Ala38, Lys41 on helix 2 and Gly64, Gly66,
Lys68, Lys69 on helix 4—were frequently found to be close to the pamoic acid.
In fact, in more than 50 % of the resulting conformations, at least one proton of
Ala38, Lys68 and Ile69 was located within 2 Å from the pamoic acid. For residues
His34, Lys35, Asn37, Lys41, Gly64 and Gly66, over 20 % of the 100 best docked
structures contained a pair of protein–ligand atoms with a separation below 2 Å.
Mapping these residues onto the 8 kDa domain structure indicated that they form
a single positively charged groove at the protein surface (Fig. 4.5a). Interestingly,
Lys35, Lys60 and Lys68, which have been shown to be responsible for singlestranded DNA binding by site-directed mutagenesis [119] are located in the groove
where pamoic acid binds to. As this groove is the one where DNA binds, pamoic
acid is likely to interfere with single-stranded DNA recognition.
Clustering the 100 best ligand structures has been performed using the RMSD
(root mean square deviation) cutoff value of 2 Å. The five resulting clusters indicated that the ligands adopt five different ensembles of conformations in the binding
site described above [93]. Calculated conformation ensembles were verified using NMR chemical shift mapping [120], NOE (Nuclear Overhauser Enhancement)
spectroscopy and STD (Saturation Transfer Difference) experiments [121, 122].
Only one reconstructed complex between the 8 kDa domain and pamoic acid was
consistent with the entire NMR data (Fig. 4.5b).
4 DNA Dependent DNA Polymerases as Targets for Low-Weight …
chemical shift mapping and subsequent molecular docking ( Affinity and Discovery
modules of Insight II software, ESFF force field [116]), the two potential binding
were proposed. Site I includes amino acid residues Leu22, Phe25, Glu26, Asn28,
Ile33, Lys35, Asn37, site II—residues Lys60, Leu62, Gly64, Gly66 and Ala70, consequently. Both discovered sites are located on template DNA binding interface
of the 8-kDa domain, and SQMG incorporation into any of them would prevent
interaction with the DNA template [104]. The above-described binding sites for
LCA and its derivates, as well as for NA, are situated on the template DNA binding
interface, too [82, 114, 115]. Thus, one can conclude that all pol β inhibitors identified by industrious japanese researchers as interacting with 8-kDA domain have a
very similar structural mechanism of action. These compounds interact with the
DNA binding area and compete with the DNA template in a similar fashion.
Similar strategy combining spectroscopic and computational approaches was applied by Hazan et al. [93] to identify the pamoic acid binding site on the surface of
DNA polymerase β. The ability of pamoic acid to interact with the 8-kDa templateprimer binding domain was previously demonstrated by Hu et al. [78]. The molecular docking of pamoic acid to the 8 kDa domain of pol β was carried out using
AutoDock 3.0.5 software [117, 118].
Structures generated by AutoDock have been ranked according to their binding
energy and 100 lowest energy structures were selected for the further analysis. With
the force field used by AutoDock, the energy values for the best ligands varied from
− 9.58 to − 8.96 kcal/mol. Systematic analysis of the 100 best docked structures
revealed that all of them were located at a single site, although pamoic acid could
move freely around the 8 kDa domain during docking. Close atomic contacts between pairs of protein–ligand atoms (with a distance cutoff of 2 Å) were computed.
Nine residues—His34, Lys35, Asn37, Ala38, Lys41 on helix 2 and Gly64, Gly66,
Lys68, Lys69 on helix 4—were frequently found to be close to the pamoic acid.
In fact, in more than 50 % of the resulting conformations, at least one proton of
Ala38, Lys68 and Ile69 was located within 2 Å from the pamoic acid. For residues
His34, Lys35, Asn37, Lys41, Gly64 and Gly66, over 20 % of the 100 best docked
structures contained a pair of protein–ligand atoms with a separation below 2 Å.
Mapping these residues onto the 8 kDa domain structure indicated that they form
a single positively charged groove at the protein surface (Fig. 4.5a). Interestingly,
Lys35, Lys60 and Lys68, which have been shown to be responsible for singlestranded DNA binding by site-directed mutagenesis [119] are located in the groove
where pamoic acid binds to. As this groove is the one where DNA binds, pamoic
acid is likely to interfere with single-stranded DNA recognition.
Clustering the 100 best ligand structures has been performed using the RMSD
(root mean square deviation) cutoff value of 2 Å. The five resulting clusters indicated that the ligands adopt five different ensembles of conformations in the binding
site described above [93]. Calculated conformation ensembles were verified using NMR chemical shift mapping [120], NOE (Nuclear Overhauser Enhancement)
spectroscopy and STD (Saturation Transfer Difference) experiments [121, 122].
Only one reconstructed complex between the 8 kDa domain and pamoic acid was
consistent with the entire NMR data (Fig. 4.5b).
4 DNA Dependent DNA Polymerases as Targets for Low-Weight …