Abstract/Details

Using computational modeling techniques to rationalize and predict metabolism and inhibition involved in CYP3A4 enyzme


2007 2007

Other formats: Order a copy

Abstract (summary)

CYP3A4 is the most abundant human hepatic CYP isoform and is responsible for the metabolism of about 50% of therapeutic agents, which vary widely in molecular structure, volume or size. Accurate predictions of the interactions between CYP3A4 and its substrates or inhibitors may help increase productivity in drug discovery efforts and could also reduce the risk of progressing lead compounds in a drug discovery program that may cause clinical drug-drug interactions and patient safety issues. In this research project, we successfully investigated the interaction between CYP3A4 and its substrates and inhibitors using computational techniques.

In the first study, two computational techniques, the MetaSite and docking methodologies, that may be used to predict sites of metabolism were successfully applied and evaluated for a set of 227 known CYP3A4 substrates using both the CYP3A4 crystal structure and a CYP3A4 homology model. The MetaSite methodology is automated, rapid, and has relatively accurate predictions compared to the docking methodology used in this study. The MetaSite method with its reactivity factor-enabled achieved 78% prediction accuracy. The docking method had a relatively lower prediction accuracy (∼57%). However it may also provide useful insights into interactions between a ligand and the CYP3A4 protein, especially for uncommon reactions.

In the second study, a CYP3A4 QSAR inhibition model was successfully derived using 2562 structurally diverse compounds and a support vector machine approach. The SVM model achieved 77% selectivity, 85% specificity and an overall 83% prediction accuracy. In addition, it was demonstrated that the distance between each compound and the separating surface in the feature space could be used as valuable confidence index for a prediction. The high classification accuracy indicates that the SVM binary classification model can be used as a high throughput computational filter for library design and for identifying CYP3A4 inhibition liability at early stages of drug discovery.

In the third study, we successfully simulated the interactions between midazolam and testosterone in the active site of CYP3A4 using a CYP3A4 homology model and the docking program GLUE. The results demonstrated that both midazolam and testosterone could simultaneously occupy the CYP3A4 active site. In the presence of testosterone, the 4-hydroxylation orientation of midazolam is preferred. This simulation matched well with experimental findings and suggests a new effector site for potential experimental follow up.

Indexing (details)


Subject
Pharmacology
Classification
0491: Pharmacology
Identifier / keyword
Pure sciences; CYP3A4; Drug discovery; Midazolam; Testosterone
Title
Using computational modeling techniques to rationalize and predict metabolism and inhibition involved in CYP3A4 enyzme
Author
Zhou, Diansong
Number of pages
143
Publication year
2007
Degree date
2007
School code
1379
Source
DAI-B 68/05, Dissertation Abstracts International
Place of publication
Ann Arbor
Country of publication
United States
ISBN
9780549039396
University/institution
University of the Sciences in Philadelphia
University location
United States -- Pennsylvania
Degree
Ph.D.
Source type
Dissertations & Theses
Language
English
Document type
Dissertation/Thesis
Dissertation/thesis number
3265587
ProQuest document ID
304700252
Copyright
Database copyright ProQuest LLC; ProQuest does not claim copyright in the individual underlying works.
Document URL
http://search.proquest.com/docview/304700252
Access the complete full text

You can get the full text of this document if it is part of your institution's ProQuest subscription.

Try one of the following:

  • Connect to ProQuest through your library network and search for the document from there.
  • Request the document from your library.
  • Go to the ProQuest login page and enter a ProQuest or My Research username / password.