Figure 17. IRC calculations of the ligand-AB3dendritic-prodrug conjugate.
To investigate changes in interaction energy between various functional components along the IRC path, we performed GAMESS/FMO calculations at the M06-2x/6-31+G(d) level. FMO2 calculations were performed for the seven fragments corresponding to seven functional components in order to obtain four snapshots along the PES of the R’ state to the TS. Figure 18 shows the interacting/bonding energies of fragments 1-2 and 3-4 remained similar along the reaction path (~−30.0 Hartree), although the bonding energies of the linker–drug fragments (5-4, 5-3, 6-4, 6-5, 7-3, and 7-4) showed relatively large fluctuations at different stages of the reaction pathway. For example, the interacting/bonding energies of fragments 7-3 and 7-4 were −31.61 Hartree and −15.18 Hartree in R’ (Figure 18a) and −0.01 Hartree and −52.89 Hartree at the IRC path’s middle point (Figure 18b). For the spacer–linker fragment (2-3), the interacting/bonding energies were −29.36 Hartree in R’ (Figure 18a) and −16.16 Hartree during the TS (Figure 18d). These changes indicated that the drug components in the ligand-AB3 dendritic-prodrug conjugate were more flexible and experienced various structural changes relative to the other components, such as the ligand and spacer. Furthermore, the C–N(H) bond of spacer–linker fragment 2-3 can undergo heterolytic dissociation in the presence of a catalyst. According to a previous study,89 enzymatic cleavage of the bond at Phe–Lys interaction can trigger dissociation of the conjugate and promote triple elimination to release the three paclitaxel molecules from each dendritic platform. Although full-stack calculations have not yet been performed for the entire therapeutic release mechanism, FMO, combined with TS calculations, represent a useful tool for studying the drug-release mechanism.