Chemistry in Silico

 

Research in the Parish group focuses on understanding the dynamical behavior of interesting molecular systems.  The tools of quantum mechanics, conformational searching and free energy simulations are used to answer questions about the structure, energy and dynamics of these systems.  We develop new tools as needed and refine and optimize existing tools. Research funding for these projects has been obtained from the NSF, ACS-PRF and the Camille and Henry Dreyfus Foundation.

 

Polyhedral Oligomeric Silsesquioxanes (POSS):  Silsesquioxanes are macromolecules that contain silicon and oxygen in a 1:1.5 atom ratio and refer to random structures, ladder structures, cage structures, and partial cage structures, as illustrated in Figure 1.

Although first reported in 1946, between that time and the late 1990s, silsesquioxanes received little attention in the literature.  Interest in these systems has exploded recently due to their commercial availabilty and their potential for use in nanoscale devices, however very little is known about their molecular behavior.  Our research group has been investigating the electronic structure, conformational flexibility and dynamical behavior of silsesquioxanes and their derivatizes in order to correlate structure with anticipated nanomolecular function.  We are perfoming molecular dynamics simulations on caged structures to investigate the “breathability” of these systems and to determine how the electronic structure of the derivatized systems effects the pore size.  We are also using conformational searching methods to probe the flexibility of partial cage and ladder structures and to determine the suitability of these systems for use in host-guest chemistry.

 

HIV Protease Inhibitor Drugs:  The human immunodeficiency virus (HIV) encodes an aspartyl protease enzyme (HIV-PR) that cleaves viral polyprotein precursors, allowing the maturation of the HIV virus that causes the autoimmune deficiency disease (AIDS).  Protease inhibitors occupy the basket-shaped active site, interfering with the functioning of the enzyme and preventing the maturation of the virus.  Our work uses conformational searching and free energy simulation methods to study different FDA-approved HIV protease inhibitor drugs such as indinavir, saquinavir, nelfinavir, amprenavir, lopinavir and ritonavir to determine the similarities and differences in the way these drugs interact with the protease active site.  For instance, functional group interactions and conformational flexibility are thought to be important in the binding of the inhibitor to the protease.  Models have been proposed for specific inhibitor-protease interactions, however there has not been a systematic comparison of known inhibitors ref.  This study will address the following questions: What common structural motifs do active HIV-PR inhibitors share?  Are hydrogen bonding patterns the same among all inhibitors or unique to individual inhibitors?  If unique, is this due to flexibility on the part of the protease?  What role do solvent interactions play?  This information will be used to develop a “prescription” or template for describing molecular behavior that is necessary in the design of new and more effective protease inhibitors.  We have completed an in-depth analysis of two recently developed conformational searching methods and we are confident that Low Mode:Monte Carlo method will allow exhaustive coverage of the conformational space of each of the inhibitors in this study in solvent and in the space of the active site.

 

 

Anticancer Enediyne Warhead Drugs:  Naturally occurring anticancer agents, such as Dynemicin A, contain reactive, electron rich enediyne moieties shown in red.  Under the proper conditions, the enediyne group undergoes a Bergman cyclization that results in a p-benzyne diradical that can abstract hydrogen atoms from DNA resulting in cancer cell death.  In order to design anticancer agents or other endiynes of biological interest it is important to understand the factors that affect the rate of Bergman cyclization.  This study will use quantum mechanical and reaction field theory methods to understand how the rate of Bergman cyclization in ten-membered enediynes is influenced by various electronic factors.  Using appropriate model systems we will investigate the effects of pH, molecular charge distribution, tautomerism and solvation.  High level ab initio calculations will be performed on each of the enediynes prepared in our collaborator’s laboratory (KC Russell, Northern Kentucky University) in order to evaluate their structure, stability, cyclization barriers, thermochemistry and singlet-triplet energy differences.  Our theoretical results will be verified by comparison with experimentally obtained temperature dependent rate constants obtained in the Russell group.  Together, these studies will provide molecular level insight into the competing factors influencing cyclization.  An understanding of how these factors effect the rate of cyclization will allow the development of more effective anticancer prodrug warheads – enediynes that cyclize and cause cell death only under certain, controlled conditions.

 

 

Recent Results

 

Comparing the Conformational Behavior of a Series of Cyclic Urea HIV-1 Protease Inhibitors  The conformational flexibility of a series of diastereomeric cyclic urea HIV–1 protease inhibitors has been examined using the Low Mode:Monte Carlo conformational search method. (Journal of Medicinal Chemistry, 2004, in press)  An ensemble of low energy structures was generated using OPLSAA/GBSA(water) and used to compare the molecular shape and flexibility of each diastereomer to the experimentally determined binding affinities and crystal structures of closely related systems. The results indicate that diastereomeric solution-phase energetic stability, conformational rigidity and ability to adopt a chair conformation correlate strongly with experimental binding affinities. Rigid body docking suggests that all of the diastereomers adopt solution-phase conformations suitable for alignment with the HIV-1 protease; however; these results indicate that the binding affinities are dependent upon subtle differences in the P1/P1' and P2/P2' substituent orientations.

Analysis of Oligomeric Models of Synthetic, Silicon-based Enzyme Catalysts:  We have investigated the conformational flexibility of hybrid organosiloxane oligomers that contain aminopyridine groups in various states of protonation. (Journal of Inorganic and Organometallic Polymers, 2002 12(1-2), 31-47.)   We are interested in these systems because they are models for larger synthetic oligomers that display acyltransferases enzyme-like properties.  In this work, an ensemble of low energy structures was generated and used to investigate the dependence of molecular shape and flexibility on protonation.  The results, as measured by the number of unique conformations, the end-to-end or longest intramolecular distance and the radius of gyration of the conformational point cloud, indicate that the number of protonated pyridines plays a significant role in the overall molecular shape; i.e., molecules with multiple pyridinium ions are, on average, significantly larger, less flexible and more extended than molecules without charge.  These results help to explain the molecular behavior of the synthetic catalytic systems and should guide future synthesis and molecular design of synthetic enzymes.

 

Optimizing Conformational Search Methods:  We have recently completed the evaluation and comparison of two widely used methods for searching the conformational space of complex molecules (Journal of Molecular Graphics and Modeling 2002, 21, 129-150).  We compared the ability of Low Mode, Monte Carlo and a hybrid Low Mode-Monte Carlo method to search the potential energy surface of three diverse molecular systems.  We considered a number of factors in the selection of test compounds and decided to include three systems which present a range of challenges to the methods; a non-cyclic system with six searchable degrees of freedom, a 7-membered heterocycle with 14 searchable degrees of freedom, and a macrocycle containing 34 searchable degrees of freedom.  We found that either method, or a combination of the methods, is equally efficient at searching the conformational space of the smaller molecular systems while a 50:50 hybrid of Low Mode and Monte Carlo is most efficient at searching the space of the larger molecular system.  This work provides a significant roadmap for other researchers in their selection of optimal conformational search protocols.