Ligand-based Drug Design: Homework
Replication of retroviruses relies on enzyme reverse transcriptase that synthesizes a double-stranded DNA intermediate using single-stranded viral RNA as a template. The DNA intermediate is integrated by the viral integrase into the genome of the host cell. The transcription of the integrated viral genome yields viral RNA molecules and proteins that are needed for the assembly of new virions. Reverse transcriptase is unique to retroviruses and thus constitutes a good target for antiviral drugs. The first AIDS drug to reach the market, Zidovudine (AZT), inhibits the reverse transcription.
While offering hope to AIDS patient when it was approved in 1987, AZT had serous drawbacks. The chemical and metabolic insability of AZT required that large dozes of drug be taken every four hours, even at nighttime. While AZT did not bind well to nuclear DNA polymerases, it had a noticable affinity against mitochondrial DNA polymerase gamma. The drug thus interfered with mitochondrial DNA synthesis and repair, especially in rapidly dividing cells, or in cells that were subject to oxidative damage. AIDS patients who took AZT often developed mitochondrial myopathy and lactic acidosis due to the accumulation of mutations in mitochondrial DNA.
A search for better HIV reverse transcriptase inhibitors quickly ensured. As documented in the paper "Correlation between preferred sugar ring conformation and activity of nucleoside analogues against human immunodeficiency virus" by Van Roey, Salerno, Chu, and Schinazi, the molecules with similar chemical structures showed wide range of antiviral activities. In particular, it appeared that the three-dimensional conformation of the analogs had a significant effect on the ability of drugs to inhibit the reverse transcription of viral RNA. Use their paper as a starting point to understand the conformational requirements for effective HIV reverse transcriptase inhibitors, and propose new structures that would show desired conformational properties while being stable toward acid-catalyzed depurination.
With the help of literature searches and/or computer modeling, answer the following questions:
- Provide a rational explanation how AZT and other 2',3'-dideoxynucleosides inhibit synthesis of the DNA intermediate from the viral RNA.
- HIV reverse transcriptase uses nucleoside triphosphates as substrates but does not bind nucleosides well. Provide a rational explanation how AZT and other 2',3'-dideoxynucleosides could even inhibit a reverse transcription without binding to reverse transcriptase.
- With the help of appropriate illustrations, explain what are C2'endo/C3'exo and C2'exo/C3'endo conformations in 2',3'-deoxyribofuranoside. If you used an image created by someone else, provide appropriate reference.
- Discuss in general terms what is known about the gauche effect in chemistry. How is the gauche effect possibly related to the endo/exo conformational preference in ribonucleosides, deoxyribonucleosides, and 2'3'dideoxynucleosides. What substituents, and in which position are expected to stabilize the C2'endo/C3'exo conformation of 2'3'-dideoxynucleosides while increasing the chemical stability toward acid-catalyzed depurination? Provide literature references where appropriate.
- Generate 3D structure of the simplified 2',3'dideoxyribofuranose using a method of your choice. Because Tinker MM3 force field does not know how to handle unusual nucleosides, the recommended simplified model should have 1'-methoxy group. It is not critical to include the 5'-phosphate; a 5'-hydroxy or 5'-deoxy analog will be sufficient for modeling. For now, use hydrogens as 2' and 3' substituents.
- Alter the structure of your model such that it would adopt 2'exo/3'endo conformation. Minimize this structures with MOLDEN's Force Field optimizer tool and MM3 Tinker force field. Note that you may need to re-type the furanose oxygen as ether (O: C-O-H, C-O-C) for the minimization to work. Make sure to give a unique name, such as dd_OCH3_C3endo before minimizing the structure. The minimized structure will be saved in file dd_OCH3_C3endo.xyz_2. If you choose so, you are free to use programs other than MOLDEN for this task.
- Alter the structure of your model such that it would adopt 2'endo/3'exo conformation. Minimize this structures with MOLDEN's Force Field optimizer tool and MM3 Tinker force field. Note that you may need to re-type the furanose oxygen as ether (O: C-O-H, C-O-C) for the minimization to work. Make sure to give a unique name, such as dd_OCH3_C2endo before minimizing the structure. The minimized structure will be saved in file dd_OCH3_C2endo.xyz_2. If you choose so, you are free to use programs other than MOLDEN for this task.
- Perform further conformational analysis of the two conformers TINKER's scan program. In this stage, the methoxy group (and 5'hydroxyl, if present) rotations are performed but TINKER's scan program cannot easily cyhange the sugar pucker.
- Use the TINKER's program analyze to calculate Total Potential Energy of each conformer (this is also listed during the scan in the last column). Note that you can give the answers to questions TINKER asks on the command line. For example, to calculate the Total Potential Energy of C2endo conformer 2, type analyze dd_OCH3_C2endo.002 E.
- Create a table showing three lowest-energy C2endo conformers and three lowest-energy C3endo conformers. For each conformer, give the name, the value of the cyclic C-C-C-C dihedral, the Total Potential Energy, and the relative energy with respect to the lowest energy conformer.
- Based on the relative energies of each of the six conformers in the previous table, calculate the probability of each of these conformers. The probabilities can be calculated from Boltzmann's distribution. Assume room temperature (kT = 0.593 kcal/mol) and no degeneracy (gi=1 as this is asymmetric molecule).
- Modify structure of the furanose ring in the in a rational attempt to make the 2'endo/3'exo conformation more favorable. Keep in mind that the MM3 force field cannot handle substituents that are too unusual. Also, long flexible substituents will quickly increase the number of conformations to unmanageable numbers. Make the same change in 2'exo/3'endo structure. Remember that substitution at 2' or 3' positions create a new stereocenter and for each modification you can have two diastereomers in this already chiral molecule. Diastereomers have different physical, chemical, and biological properties, so you need to evaluate both stereoisomers. In other words, your goal is to find the substituent, and its absolute stereochemistry that makes the 2'endo/3'exo ring conformation favorable.
- Minimize the 2'endo/3'exo and 2'exo/3'endo conformers of of your modified nucleoside using MM3 force field and MOLDEN.
- Perform conformational analysis of your modified molecules using TINKER's scan program. You may observe that the search from one ring pucker also finds minima with a different ring pucker. This is the expected behavior for conformational analysis programs.
- Use the TINKER's program analyze to calculate Total Potential Energy of each conformer (this is also listed during the scan in the last column).
- Create a table showing two lowest-energy 2'endo/3'exo conformers and two lowest-energy 2'exo/3'endo conformers of your modified compound. For each conformer, give the name, the value of the cyclic C-C-C-C dihedral, the Total Potential Energy, and the relative energy with respect to the lowest energy conformer.
- Based on the relative energies of each of the four conformers in the previous table, calculate the probability of each of these conformers. The probabilities can be calculated from Boltzmann's distribution. When evaluating the partition function, you may ignore the presence of other high-energy conformers (this means that i = 1, 2, 3, 4). Assume room temperature (kT = 0.593 kcal/mol) and no degeneracy (gi=1 as this is asymmetric molecule).
- Overlay one of your best modified analog with the structure of the bioactive conformer of AZT from PDB file 1KDT using a program of your choice (Chimera & Sybyl are available in the SGI lab). You can convert the Tinker's XYX file to PDB file using either MOLDEN or TINKER's xyzpdb tool.
- Discuss if your attempt to increase the fraction of bioactive 2'endo/3'exo ring pucker conformers by modifying the dideoxyfuranose ring was successful.
Please follow this link to see homework for Winter 2011 course.