these mutations reduce the susceptibility of IN to both RAL and EVG , while susceptibility to MK 0536 and DTG is only modestly affected . N155 is positioned such that it makes Anastrozole Van der Waals contacts with both D64 and E152 of the catalytic triad and likely reduces susceptibility to RAL and EVG by altering the binding geometry around the Mg2 ions . As with the G140S/Q148H IN, MK 0536 and DTG retain activity against N155H. Even though this set of resistance mutations is limited, the effects of these mutations on susceptibility to these four compounds are distt. This illustrates the importance of developing accurate methods for predicting the antiviral activity of INSTIs against both wild type and drugresistant forms of HIV 1 IN.
The development of early INSTIs was driven by structureactivity relationship studies and in silico screening based on structures of fragments of HIV 1 IN . The recent crystal structures of PFV IN DNA INSTI complexes have provided valuable new insights into the binding modes of the inhibitors. Using the PFV IN structures, we developed a homology model of HIV 1 IN with similarities S1P Receptors to that of Krishnan This is one of two recently published models that rely on tures of HIV 1 IN fragments to develop a full length model . The available HIV 1 IN fragment structures differ in the arrangement of interdomain linkers and dimer interfaces. Our model was built using the crystal structure of full length PFV IN and the available structures of the HIV 1 fragments; however, in our model the structure of the interdomain linkers is based on the PFV IN structure.
We have used our model in a docking and molecular dynamics protocol that correlates the interaction potential of the Mg2 ions in several nonpositivist INSTI intasome complexes with their in vitro activities. This correlation ludes the RALresistant mutants Y143R, N155H, and G140S/Q148H. Additional mutants can be easily orporated into the analysis. This is important because, when INSTIs under investigation are approved for clinical use, therapy with these new INSTIs will select for new resistance mutations, in addition to those seen with RAL treatment. We also present a series of binding energy calculations that can be applied to examine changes in activity between compounds or between mutants. Energy differences can be calculated for the internal or interaction potential for the ligand, Mg2 ions, protein residues, and the DNA’s terminal dinucleotide.
These approaches can significantly reduce the amount of chemical synthesis needed to develop new INSTIs by identifying compounds predicted to have weak activity and focusing synthesis on compounds with favorable thermodynamic profiles. MATERIALS AND METHODS Docking INSTIs. The coordinates of our HIV 1 IN model are based on the PFV IN crystal structure that served as the initial template . The sequences of the HIV 1 and PFV IN catalytic core domains were aligned with an array of other retroviral integrases. Using the CCD alignment as a starting point, sequence alignments were performed for the NTD and CTD. A homology model was generated using MOE 2009.10 based on the full length alignment, and energy was minimized with the AMBER99 force field with relative field solvation, as recommended by the manufacturer .