Prediction of the in vitro permeability determined in Caco-2 cells by using artificial neural networks
Introduction
Oral administration of drugs, due to its ease and patient compliance, is the preferred route and a major goal in the development of new drug entities. It is also traditionally one of the reasons for either discontinuation or prolongation of the development time of compounds. These problems lead to a new paradigm, a multivariate approach, in compound selection and optimization (Venkatesh and Lipper, 2000). In this context, and as a consequence of combinatorial chemistry, initial screening of compounds in a number of thousands is done typically by using in silico approaches. In vitro tests are responsible to reduce the number of compounds from hundreds to dozens and in vivo animal models to 1–5 finally potential drugs that are included in clinical trials.
Various in silico models already exist to predict drug absorption potential. One of the most known is the “Lipinski rule of 5” (Lipinski et al., 2001), based on the analysis of successful drugs, in which indications for poor absorbed compounds are based on the number of H-bond donors and acceptors, molecular weight and ClogP. After that initial effort, various models were proposed to quantitatively predict oral absorption. These were based on the search to correlate some molecular descriptors to measures of bioavailability by means of multivariate regression tools. Models were developed based on the in vivo Human Intestinal Absorption (Abraham et al., 2002, Butina, 2004, Klopman et al., 2002, Norinder and Osterberg, 2001, Zhao et al., 2001), in vivo absorption rate constants (Linnankoski et al., 2006), in vivo jejunal effective permeability (Winiwarter et al., 2003, Winiwarter et al., 1998), in vitro apparent permeability in artificial membranes (Fujikawa et al., 2005, Fujikawa et al., 2007, Nakao et al., 2009, Verma et al., 2007) and in vitro apparent permeability (Papp) in Caco-2 cells (Castillo-Garit et al., 2008, Degim, 2005, Di Fenza et al., 2007, Fujiwara et al., 2002, Hou et al., 2004, Nordqvist et al., 2004, Ponce et al., 2004, Santos-Filho and Hopfinger, 2008, Yamashita et al., 2002a).
In vivo based data, although obviously the target for the lead drug, presents some drawbacks. In vivo global bioavailability results from different physical−chemical and biological processes that are difficult to isolate. To be absorbed, a drug needs first to be in its soluble form. Passive diffusion may be the principal driving force for drug absorption, but either active transport or efflux mechanisms may condition the final bioavailability. Instability in the gastro-intestinal fluids and metabolization may also reduce the amount of drug that effectively reaches the systemic circulation. For these reasons, calculation of the absorption rate constants, which uses pharmacokinetic data after intravenous and oral administration, may underestimate the absorption. Collection of data on Human Intestinal Absorption is also experimentally demanding. It requires the evaluation of the fraction of dose, either as parent or metabolites, eliminated by the faeces and urine which may easily result in underestimation of the actual fraction of dose absorbed. Finally, data on jejunal effective permeability, which isolates the absorption process but is experimentally complex, is limited to only a small number of drugs. Another drawback in these three approaches is that prediction confirmation is only possible when the drug is administered to humans in the end stages of drug discovery, leaving no room for “fine-tuning” during the development phase.
In vitro tests, on the contrary, are routinely performed by pharmaceutical companies in an early stage of drug development. For this reason, a large number of Papp values for different molecules are being produced and are/will be available to build databases with a large span of values from high to low bioavailabilities. Parallel artificial membrane permeation assays (PAMPA) are a very promising tool to predict absorption, but there are various experimental variables to be considered and in practice it appears to produce the same outcome that a simple Log D7.4 assay in predicting high and low absorption (Galinis-Luciani et al., 2007). Caco-2 cells, on the contrary, are a widely preformed in vitro test with interesting properties when extrapolating results to bioavailability. It is a cell system, characterized by easy handling and at the same time resemble morphological and biochemical characteristics of the intestinal cells (Vogel, 2006), that shows a sigmoid relationship between the fraction absorbed in humans and the Papp across the cells (Artursson and Karlsson, 1991).
A number of statistical multivariate methods for developing in silico models are currently used. Artificial neural networks (ANN), however, have shown superior performance to the linear multivariate class of methods in various QSAR models (Fujiwara et al., 2002, Paixao et al., 2009, Sutherland et al., 2004, Votano et al., 2004). To the best of our knowledge, only three studies were previously reported to predict in vitro apparent permeability (Papp) in Caco-2 cells using ANN and molecular descriptors. Of these, two studies (Degim, 2005, Fujiwara et al., 2002) were based on a small number of drugs, did not make any particulate approach to avoid overfitting, nor made a convenient search of optimal molecular descriptors to predict the in vitro Caco-2 permeability. The other study (Di Fenza et al., 2007) resolved some of these problems, but was based on proprietary analogue drugs. Since the proposed model was built with the use of molecules contained in a very similar chemical space, it was proved to be less competent in predicting Log Papp values in a more chemical diverse external dataset.
With the purpose to overcome these limitations, we present a backpropagation ANN model using early stop to predict in vitro Caco-2 Log Papp based on 296 structurally different drugs and drug-like molecules collected in the literature. This large dataset would provide a broad molecular space, ideal to enlarge the applicability of the proposed model. Various molecular descriptors, encoding different molecular characteristics were initially used. Reduction of the number of possible molecular descriptors was made by a simple pruning procedure, allowing to point out some possible factors influencing the drugs absorption process. Use of early stop during the whole modelling procedure also ensured that optimal prediction capabilities were maintained in the final model. This model, and proposed methodology, may be a valuable tool in early drug development.
Section snippets
Data collection
In vitro Log Papp values (Table 1) for 296 different drugs were obtained from published studies of drug absorption in Caco-2 cells. Examples of drugs absorbed by passive diffusion using either the paracellular or transcellular route, and cases of active transport by different transporters were included. Examples of drugs suffering efflux mechanisms were also included. In all cases when non-linear relations between Papp and apical concentrations were described, Papp values were collected in the
Results
Optimization of the ANN model was made as described under methods. Regarding the reduction of the molecular descriptors space, the removal of correlated and non-discriminatory descriptors reduced the number of descriptors to 79. The next two procedures were performed using an ANN with an architecture of n-3-2-1, n being the number of molecular descriptors. This ANN structure was chosen as a compromise between simplicity, in order to avoid memorisation, and complexity to allow an adequate
Model data
Drug absorption in the GIT is a complex process. Various physical, chemical and biological processes are known to be involved in the overall oral bioavailability of drugs that may condition the movement of compounds from the intestinal lumen to the blood circulation. Permeation in the epithelium lining is, however, one of the major factors governing the drug bioavailability, and the in vitro Caco-2 cell system has been shown to be a suitable model for studies on intestinal drug absorption (
Conclusions
In conclusion, we presented an ANN methodology based on a pruning procedure to reduce the descriptors space, and using an early stop approach that produced a robust and logical model with good predictive abilities. Comparison of our model with the previously reported ANN models also stresses its characteristics. It was build with more drugs than the method of Fujiwara et al. (Fujiwara et al., 2002), with an external validation with similar statistics of the authors own “leave-one-out”
Acknowledgment
This work was partially supported by project number SFRH/BD/28545/2006 from Fundação para a Ciência e a Tecnologia.
References (114)
- et al.
Application of hydrogen bonding calculations in property based drug design
Drug Discov. Today
(2002) - et al.
Biopharmaceutical characterization of sotalol-containing oral immediate release drug products
Eur. J. Pharm. Biopharm.
(2004) - et al.
Correlation between oral drug absorption in humans and apparent drug permeability coefficients in human intestinal epithelial (Caco-2) cells
Biochem. Biophys. Res. Commun.
(1991) - et al.
Do cell culture conditions influence the carrier-mediated transport of peptides in Caco-2 cell monolayers?
Eur. J. Pharm. Sci.
(2003) - et al.
Estimation of ADME properties in drug discovery: predicting Caco-2 cell permeability using atom-based stochastic and non-stochastic linear indices
J. Pharm. Sci.
(2008) The influence of P-glycoprotein on morphine transport in Caco-2 cells. Comparison with paclitaxel
Eur. J. Pharmacol.
(2002)- et al.
Relationships between structure and high-throughput screening permeability of diverse drugs with artificial membranes: application to prediction of Caco-2 cell permeability
Bioorg. Med. Chem.
(2005) - et al.
QSAR study on permeability of hydrophobic compounds with artificial membranes
Bioorg. Med. Chem.
(2007) - et al.
Prediction of Caco-2 cell permeability using a combination of MO-calculation and neural network
Int. J. Pharm.
(2002) - et al.
Is PAMPA a useful tool for discovery?
J. Pharm. Sci.
(2007)
Modelling and prediction of soil sorption coefficients of non-ionic organic pesticides by molecular descriptors
Chemosphere
Caco-2 versus Caco-2/HT29-MTX co-cultured cell lines: permeabilities via diffusion, inside- and outside-directed carrier-mediated transport
J. Pharm. Sci.
Functional expression of P-glycoprotein in apical membranes of human intestinal Caco-2 cells. Kinetics of vinblastine secretion and interaction with modulators
J. Biol. Chem.
MDCK (Madin–Darby canine kidney) cells: a tool for membrane permeability screening
J. Pharm. Sci.
Paracellular drug transport across intestinal epithelia: influence of charge and induced water flux
Eur. J. Pharm. Sci.
Combined application of parallel artificial membrane permeability assay and Caco-2 permeability assays in drug discovery
J. Pharm. Sci.
ADME evaluation. 2. A computer model for the prediction of intestinal absorption in humans
Eur. J. Pharm. Sci.
Effect of experimental pH on the in vitro permeability in intact rabbit intestines and Caco-2 monolayer
Eur. J. Pharm. Sci.
Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings
Adv. Drug Deliv. Rev.
Validation of the 96 well Caco-2 cell culture model for high throughput permeability assessment of discovery compounds
Int. J. Pharm.
Optimizing Caco-2 cell monolayers to increase throughput in drug intestinal absorption analysis
J. Pharmacol. Toxicol. Methods
Solubility parameter and oral absorption
Eur. J. Pharm. Biopharm.
Usefulness of a novel Caco-2 cell perfusion system. I. In vitro prediction of the absorption potential of passively diffused compounds
J. Pharm. Sci.
Comparison of in vitro models for the prediction of compound absorption across the human intestinal mucosa
J. Biomol .Screen
Transepithelial permeation of tolbutamide across the human intestinal cell line, Caco-2
Drug Metab. Pharmacokinet.
Theoretical calculation and prediction of drug transport processes using simple parameters and partial least squares projections to latent structures (PLS) statistics. The use of electrotopological state indices
J. Pharm. Sci.
Link between drug absorption solubility and permeability measurements in Caco-2 cells
J. Pharm. Sci.
Prediction of drug distribution within blood
Eur. J. Pharm. Sci.
Correlation of drug absorption with molecular surface properties
J. Pharm. Sci.
Automatic early stopping using cross validation: quantifying the criteria
Neural Networks
Combined 4D-fingerprint and clustering based membrane-interaction QSAR analyses for constructing consensus Caco-2 cell permeation virtual screens
J. Pharm. Sci.
Permeability characteristics and membrane affinity of flavonoids and alkyl gallates in Caco-2 cells and in phospholipid vesicles
Arch. Biochem. Biophys.
Application of ALOGPS to predict 1-octanol/water distribution coefficients, logP, and logD, of AstraZeneca in-house database
J. Pharm. Sci.
Pharmacokinetic parameter prediction from drug structure using artificial neural networks
Int. J. Pharm.
Role of the development scientist in compound lead selection and optimization
J. Pharm. Sci.
Asymptotic statistical theory of overtraining and cross-validation
IEEE Trans. Neural Network
Interactions of the dipeptide ester prodrugs of acyclovir with the intestinal oligopeptide transporter: competitive inhibition of glycylsarcosine transport in human intestinal cell line-Caco-2
J. Pharmacol. Exp. Ther.
Selective paracellular permeability in two models of intestinal absorption: cultured monolayers of human intestinal epithelial cells and rat intestinal segments
Pharm. Res.
Delineation of human peptide transporter 1 (hPepT1)-mediated uptake and transport of substrates with varying transporter affinities utilizing stably transfected hPepT1/Madin–Darby canine kidney clones and Caco-2 cells
J. Pharmacol. Exp. Ther.
The influence of culture time and passage number on the morphological and physiological development of Caco-2 cells
Proc. Soc. Exp. Biol. Med.
Performance of Kier-hall E-state descriptors in quantitative structure activity relationship (QSAR) studies of multifunctional molecules
Molecules
In vitro permeability of eight beta-blockers through Caco-2 monolayers utilizing liquid chromatography/electrospray ionization mass spectrometry
J. Mass Spectrom.
Estimation of permeability by passive diffusion through Caco-2 cell monolayers using the drugs’ lipophilicity and molecular weight
Eur. J. Pharm. Sci.
No free lunch for early stopping
Neural Comput.
Transepithelial transport of the fluoroquinolone ciprofloxacin by human airway epithelial Calu-3 cells
Antimicrob. Agents Chemother.
In vitro and in situ absorption of SDZ-RAD using a human intestinal cell line (Caco-2) and a single pass perfusion model in rats: comparison with rapamycin
Pharm. Res.
Short term Caco-2/TC7 cell culture: comparison between conventional 21-d and a commercially available 3-d system
Biol. Pharm. Bull.
In Silico Prediction of Physicochemical Properties. JRC Scientific and Technical Reports
Prediction of permeability coefficients of compounds through Caco-2 cell monolayer using artificial neural network analysis
Drug Dev. Ind. Pharm.
Caco-2 cell permeability modelling: a neural network coupled genetic algorithm approach
J. Comput. Aided Mol. Des.
Cited by (53)
Toxicokinetics and tissue distribution of phalloidin in mice
2023, Food and Chemical ToxicologyFormation of a Bile Salt-Drug Hydrogel to Predict Human Intestinal Absorption
2019, Journal of Pharmaceutical SciencesExploring a Kinetic Model Approach in Biopharmaceutics: Estimating the Fraction Absorbed of Orally Administered Drugs in Humans
2018, Journal of Pharmaceutical SciencesVirtual Screening of Phytochemicals
2018, Computational PhytochemistryQuantitative structure-retention relationships of flavonoids unraveled by immobilized artificial membrane chromatography
2016, European Journal of Pharmaceutical Sciences