Prediction of molecular polar surface area and bioabsorption

Details Prediction of molecular polar surface area and bioabsorption Prediction of molecular polar surface area and bioabsorption

2000-04-19

2003-02-18

Marschel, Ardin H. (Department: 1631)

Data processing: measuring, calibrating, or testing

Measurement system in a specific environment

Biological or biochemical

C702S022000, C702S027000, C702S030000, C422S068100

Reexamination Certificate

active

06522975

ABSTRACT:

SUMMARY OF THE INVENTION
1. Field of the Invention
The invention relates to computational methods of pharmaceutical discovery. Specifically, the invention relates to the prediction of membrane permeability and physiological absorption of molecules.
2. Description of the Related Art
In the development of pharmaceutical compounds, it is well known that identifying a compound with a desired biological activity is not itself sufficient to determine that compound's suitability as a drug. Not only must the compound exhibit the necessary biological activity, but it must also be deliverable to the target tissue(s), preferably in a cost effective and convenient manner such as oral administration. This has been a problem with several treatment protocols. For example, a wide variety of peptide molecules have been shown to have useful pharmacological activity, but their generally limited capacity to diffuse through biomembranes such as the human gastrointestinal epithelium has limited their clinical development. Effective oral administration requires that a drug be absorbed through the intestinal membranes to enter systemic circulation and if such absorption is limited, a compound's promise for clinical development is poor.
Not only is intestinal absorption an important concern during drug development, but the ability of a candidate compound to penetrate the blood-brain barrier is also of significant interest. The blood-brain barrier (BBB) is a cellular system that separates the fluids of the central nervous system from the circulatory system. Drugs intended for targets in the central nervous system should be able to penetrate the BBB. On the other hand, drugs intended for other target tissues may cause unwanted side effects if they freely pass into the fluids of the central nervous system.
In vivo animal testing for bioabsorption and blood-brain barrier penetration has long been practiced. In addition, a cell based in vitro assay using human intestinal caco-2 cells is in widespread use to measure the biomembrane permeability of drug candidate compounds. Because both of these protocols are slow, expensive, and labor intensive, computational methods to predict the potential for gastrointestinal absorption and blood-brain barrier penetration based on more easily obtained molecular characteristics have been developed. Also, such computational methods are of great interest for the in silico prediction of absorption and blood-brain barrier penetration for virtual libraries of compounds which have not been synthesized, for the purposes of determining which compounds should be synthesized. In these computational models, a formula for estimating either the % human intestinal absorption (%HIA) or the logarithm of the steady state ratio of the compound's concentration in the central nervous system and the blood (often called logBB) is constructed. The formula typically uses molecular properties and parameters that may be derived from the molecular structure of the compound. Using these formulas, %HIA and logBB may be estimated without the need to perform in vivo experiments.
Many models focus on molecular characteristics related to hydrogen bonding, lipophilicity, and molecular weight to predict propensity for intestinal absorption or blood-brain barrier penetration. A sigmoidal relationship between the polar surface area (PSA) of a molecule and its %HIA has been observed, with high polar surface area correlated to low %HIA. This is shown in Palm, et al. Polar Molecular Surface Properties Predict the Intestinal Absorption of Drugs in Humans,
Pharmaceutical Research,
Vol. 14, No. 5, p. 568 (1997). It has also been observed that molecules having either especially high or especially low octanol/water partition coefficients (logP), which is a measure of lipophilicity, are associated with low %HIA. Palm, supra, and Wils, et al., High Lipophilicity Decreases Drug Transport Across Intestinal Epithelial Cells,
The Journal of Pharmacology and Experimental Therapeutics,
Volume 269, No. 2, p. 654 (1994). The disclosures of both the Palm and Wils articles are hereby incorporated by reference in their entireties.
Accordingly, molecular PSA has been suggested as a parameter which can be used to distinguish between well absorbed and poorly absorbed compounds, with 140 square angstroms being proposed as a cutoff value. Clark, Rapid Calculation of Polar Molecular Surface Area and its Application to the Prediction of Transport Phenomena 1. Prediction of Intestinal Absorption,
Journal of Pharmaceutical Sciences,
Vol. 88, No. 8, p. 807 (1999). PSA has also been used as a variable in linear formulas for predicting membrane permeability and logBB. In some cases, logP is used along with PSA in such linear formulas. Clark, Rapid Calculation of Polar Molecular Surface Area and its Application to the Prediction of Transport Phenomena 2. Prediction of Blood-Brain Barrier Penetration,
Journal of Pharmaceutical Sciences,
Vol. 88, No. 8, p. 815 (1999) and Winiwarter, et al. Correlation of Human Jejunal Permeability (in Vivo) of Drugs with Experimentally and Theoretically Derived Parameters. A Multivariate Data Analysis Approach,
Journal of Medicinal Chemistry
41, p. 4939 (1998), both of which are hereby incorporated by reference in their entireties.
Although these models have improved the speed of the drug candidate evaluation process by reducing reliance on in vivo and in vitro chemical testing, they remain computationally expensive, and in many cases, the strict linear modeling limits their predictive value. PSA calculations have required the calculation of energy minimized three dimensional molecular structures, which requires 10-15 seconds of CPU time on a Sun or SGI-R1000 workstation. The effective application of these techniques to large libraries of candidate compounds requires techniques which reduce the computation time required for each molecule.
SUMMARY OF THE INVENTION
In one embodiment, the invention comprises a method of estimating the polar surface area of a molecule comprising making separate counts of the number of atoms or atom clusters in the molecule which fall within the definition of each of a plurality of atomic classes, and estimating a polar surface area of the molecule using at least some of the separate counts.
Computed polar surface areas are useful in computer implemented compound analysis methods, and the invention also comprises a method of predicting the propensity of a molecule for membrane permeability comprising computing a polar surface area for the molecule without reference to an energy minimized three dimensional structure of the molecule and using the computed polar surface area to predict the propensity.
Methods of drug discovery are also provided. In one embodiment, such a method comprises selecting a subset of molecules from a database, wherein the selecting comprises numerically estimating the PSA and logP of each molecule in the database, determining the position of each compound in the database in a PSA-logP plane, and categorizing compounds based on their position on the PSA-logP plane.


REFERENCES:
Van de Waterbeemd et al.; “Estimation of Blood-Brain Barrier Crossing of Drugs Using Molecular Size and Shape, and H-Bonding Descriptors.” Journal of Drug Targeting, 1998, vlol. 6, No. 2, pp. 151-165.*
Egan et al.; “Prediction of Drug Absorption Using Multivariate Statistics.” Journal of Medicinal Chemistry, 2000, vol. 43, pp. 3867-3877.*
Egan et al.; “Outlier Detection i Multivariate Analytical Chemical Data.” Analytical Chemistry, 1998, vol. 70, pp. 2372-2379.*
Ghose, et al.,J. Phys. Chem. A, vol. 102, pp. 3762-3772, 1998 “Prediction of Hydrophobic (Lipophilic) Properties of Small Organic Molecules Using Fragmental Methods: An Anlaysis of ALOGP and CLOGP Methods”.
Hall, et al.,J. Chem. Inf. Comput. Sci., vol. 35, pp. 1039-1045, 1995 “Electrotopological State Indices for Atom Types: A Novel Combination of Electronic, Topoloical, and Valence State Information”.
Winiwarter, et al.,J. Med. Chem., vol. 41, pp. 4939-4949, 1998 “Correlation of Human Jejunal Permeab

Affiliated with

Also associated with

Rating

Prediction of molecular polar surface area and bioabsorption does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Prediction of molecular polar surface area and bioabsorption, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Prediction of molecular polar surface area and bioabsorption will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3168699