A performance evaluation of multiple classification models of human PEPT1 inhibitors and non-inhibitors

Structure-based model profiles affinity constant of drugs with hPEPT1 for rapid virtual screening of hPEPT1's substrate

The human proton-coupled peptide transporter (hPEPT1) with broad substrates is an important route for improving the pharmacokinetic performance of drugs. Thus, it is essential to predict the affinity constant between drug molecule and hPEPT1 for rapid virtual screening of hPEPT1's substrate during lead optimization, candidate selection and hPEPT1 prodrug design. Here, a structure-based in silico model for 114 compounds was constructed based on eight structural parameters. This model was built by the multiple linear regression method and satisfied all the prerequisites of the regression models. For the entire data set, the r(2) and adjusted r(2) values were 0.74 and 0.72, respectively. Then, this model was used to perform substrate/non-substrate classification. For 29 drugs from DrugBank database, all were correctly classified as substrates of hPEPT1. This model was also used to perform substrate/non-substrate classification for 18 drugs and their prodrugs; this QSAR model also can distinguish between the substrate and non-substrate. In conclusion, the QSAR model in this paper was validated by a large external data set, and all results indicated that the developed model was robust, stable, and can be used for rapid virtual screening of hPEPT1's substrate in the early stage of drug discovery.

Classification models for safe drug molecules

Frequent failure of drug candidates during development stages remains the major deterrent for an early introduction of new drug molecules. The drug toxicity is the major cause of expensive late-stage development failures. An early identification/optimization of the most favorable molecule will naturally save considerable cost, time, human efforts and minimize animal sacrifice. (Quantitative) Structure Activity Relationships [(Q)SARs] represent statistically derived predictive models correlating biological activity (including desirable therapeutic effect and undesirable side effects) of chemicals (drugs/toxicants/environmental pollutants) with molecular descriptors and/or properties. (Q)SAR models which categorize the available data into two or more groups/classes are known as classification models. Numerous techniques of diverse nature are being presently employed for development of classification models. Though there is an increasing use of classification models for prediction of either biological activity or toxicity, the future trend will naturally be towards the development of classification models capable of simultaneous prediction of biological activity, toxicity, and pharmacokinetic parameters so as to accelerate development of bioavailable safe drug molecules.

Human intestinal transporter database: QSAR modeling and virtual profiling of drug uptake, efflux and interactions

PURPOSE

Membrane transporters mediate many biological effects of chemicals and play a major role in pharmacokinetics and drug resistance. The selection of viable drug candidates among biologically active compounds requires the assessment of their transporter interaction profiles.

METHODS

Using public sources, we have assembled and curated the largest, to our knowledge, human intestinal transporter database (>5,000 interaction entries for >3,700 molecules). This data was used to develop thoroughly validated classification Quantitative Structure-Activity Relationship (QSAR) models of transport and/or inhibition of several major transporters including MDR1, BCRP, MRP1-4, PEPT1, ASBT, OATP2B1, OCT1, and MCT1.

RESULTS

QSAR models have been developed with advanced machine learning techniques such as Support Vector Machines, Random Forest, and k Nearest Neighbors using Dragon and MOE chemical descriptors. These models afforded high external prediction accuracies of 71-100% estimated by 5-fold external validation, and showed hit retrieval rates with up to 20-fold enrichment in the virtual screening of DrugBank compounds.

CONCLUSIONS

The compendium of predictive QSAR models developed in this study can be used for virtual profiling of drug candidates and/or environmental agents with the optimal transporter profiles.

Discovery of ligands for the human intestinal Di-/Tripeptide transporter (hPEPT1) using a QSAR-assisted virtual screening strategy

The discovery of novel ligands for the hPEPT1 transporter is reported. By exploiting a fast and rigorously validated QSAR model in combination with the distance in activity-centered chemical space (DACCS) approach, a database of commercially available compounds (Sigma-Aldrich) was screened for virtual hits. Twelve compounds were then purchased and characterized in an apical [14C]Gly-Sar uptake competition assay. Four compounds displayed affinity in the medium-to-high range. A simple benzophenone derivative displayed high affinity with a sub-millimolar binding constant (Ki=0.24 mM). The results of this study will serve as starting points for future projects, including the design and synthesis of compound libraries that seek to systematically explore the fundamental requirements for binding and transport by hPEPT1.

QSAR models for the human H(+)/peptide symporter, hPEPT1: affinity prediction using alignment-independent descriptors

A data set comprising the major known chemical classes of hPEPT1 ligands was compiled from the literature. For these compounds, alignment-independent descriptors (VolSurf, GRIND/Almond, and MOE) were computed. Using hierarchical partial least-squares projection to latent structures (H-PLS), a one-component model with r2 = 0.77 and q2 = 0.75 was obtained. The model satisfied a set of rigorous validation criteria and performed well in the prediction of an external test set. Mechanistic interpretation of the model reveals polarity properties to be the dominant factors in determining hPEPT1 affinity, with hydrophobic interactions contributing to a lesser extent. The model is superior to previously reported models due to its combination of quality and speed. Accordingly, it is suitable for ligand-based virtual screening, such as QSAR-based database mining.