Development of Classification Models for Identifying “True” P-glycoprotein (P-gp) Inhibitors Through Inhibition, ATPase Activation and Monolayer Efflux Assays
Abstract
:1. Introduction
2. Results and Discussion
2.1. Leave-One-Out cross Validation and Test Set Prediction
2.2. Consensus Modelling and Prediction Task on an Additional External Set
3. Experimental Section
3.1. Dataset and Molecular Descriptors
3.2. Classification Methods
3.3. Statistical Validation
3.4. Applicability Domain
4. Conclusions
Supplementary Information
ijms-13-06924-s001.pdfAcknowledgments
References
- Dean, M.; Hamon, Y.; Chimini, G. The human ATP-binding cassette (ABC) transporter superfamily. J. Lipid. Res 2001, 42, 1007–1017. [Google Scholar]
- Bates, E.S.; Robey, R.; Knutsen, T.; Honjo, Y.; Litman, T.; Dean, M. New ABC transporters in multi-drug resistance. Emerging Ther. Targets 2000, 4, 561–580. [Google Scholar]
- Gottesman, M.M.; Fojo, T.; Bates, S.E. Multidrug resistance in cancer: Role of ATP-dependent transporters. Nat. Rev. Cancer 2002, 2, 48–58. [Google Scholar]
- Schneider, E.; Paul, D.; Ivy, P.; Cowan, K.H. Multidrug resistance. Cancer Chemother. Biol. Response Modif 1999, 18, 152–177. [Google Scholar]
- Glavinas, H.; Krajcsi, P.; Cserepes, J.; Sarkadi, B. The role of ABC transporters in drug resistance, metabolism and toxicity. Curr. Drug Deliv. 2004, 1, 27–42. [Google Scholar]
- Hyde, S.C.; Emsley, P.; Hartshorn, M.J.; Mimmack, M.M.; Gileadi, U.; Pearce, S.R.; Gallagher, M.P.; Gill, D.R.; Hubbard, R.E.; Higgins, C.F. Structural model of ATP-binding proteins associated with cystic fibrosis, multidrug resistance and bacterial transport. Nature 1990, 346, 362–365. [Google Scholar]
- Quinton, P.M. Physiological basis of cystic fibrosis: A historical perspective. Physiol. Rev 1999, 79, S3–S22. [Google Scholar]
- Remaley, A.T.; Rust, S.; Rosier, M.; Knapper, C.; Naudin, L.; Broccardo, C.; Peterson, K.M.; Koch, C.; Arnould, I.; Prades, C.; et al. Human ATP-binding cassette transporter 1 (ABC1): genomic organization and identification of the genetic defect in the original Tangier disease kindred. Proc. Natl. Acad. Sci. USA 1996, 96, 12685–12690. [Google Scholar]
- Brooks-Wilson, A.; Marcil, M.; Clee, S.M.; Zhang, L.H.; Roomp, K.; van Dam, M.; Yu, L.; Brewer, C.; Collins, J.A.; Molhuizen, H.O.; et al. Mutations in ABC1 in Tangier disease and familial high-density lipoprotein deficiency. Nat. Genet 1999, 22, 336–345. [Google Scholar]
- Rapposelli, S.; Digiacomo, M.; Balsamo, A. P-gp transporter and its role in neurodegenerative diseases. Curr. Top. Med. Chem 2009, 9, 209–217. [Google Scholar]
- Krishna, R.; Mayer, L.D. Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs. Eur. J. Pharmacol. Sci 2000, 11, 265–283. [Google Scholar]
- Polli, J.W.; Wring, S.A.; Humphreys, J.E.; Huang, L.; Morgan, J.B.; Webster, L.O.; Serabjit-Singh, C.S. Rational use of in vitro P-glycoprotein assays in drug discovery. J. Pharmacol. Exp. Ther 2001, 299, 620–628. [Google Scholar]
- Colabufo, N.A.; Berardi, F.; Perrone, R.; Rapposelli, S.; Digiacomo, M.; Vanni, M.; Balsamo, A. Synthesis and biological evaluation of (hetero)arylmethyloxy- and arylmethylamine-phenyl derivatives as potent P-glycoprotein modulating agents. J. Med. Chem 2008, 51, 1415–1422. [Google Scholar]
- Colabufo, N.A.; Berardi, F.; Perrone, R.; Rapposelli, S.; Digiacomo, M.; Balsamo, A. Arylmethyloxyphenyl derivatives: Small molecules displaying P-glycoprotein inhibition. J. Med. Chem. 2006, 49, 6607–6613. [Google Scholar]
- Colabufo, N.A.; Berardi, F.; Perrone, R.; Rapposelli, S.; Digiacomo, M.; Vanni, M.; Balsamo, A. 2-[(3-Methoxyphenylethyl)phenoxy]-based ABCB1 inhibitors: Effect of different basic side-chains on their biological properties. J. Med. Chem 2008, 51, 7602–7613. [Google Scholar]
- Berardi, F.; Colabufo, N.A.; Perrone, R.; Balsamo, A.; Rapposelli, S.; Digiacomo, M. 1-phenylalcoy-2-beta-phenylethyl Derivatives as P-glycoprotein (P-gp) Inhibitors Useful in Drug Resistance Events. U.S. Patent Application US0,093,493,9, April 2009. [Google Scholar]
- OECD Home Page. Available online: http://www.oecd.org/ehs accessed on 6 June 2012.
- Palm, K.; Stenberg, P.; Luthman, K.; Artursson, P. Polar molecular surface properties predict the intestinal absorption of drugs in humans. Pharm. Res 1997, 14, 568–571. [Google Scholar]
- Gadhe, C.G.; Madhavan, T.; Kothandan, G.; Cho, S.J. In Silico Quantitative Structure-Activity Relationship Studies on P-gp Modulators of Tetrahydroisoquinoline-Ethyl-Phenylamine Series. BMC Struct. Biol 2011, 11, 5–19. [Google Scholar]
- Fernandes, J.; Gattas, C.R. Topological Polar Surface Area Defines Substrate Transport by Multidrug Resistance Associated Protein 1 (MRP1/ABCC1). J. Med. Chem 2009, 52, 1214–1218. [Google Scholar]
- Schwaha, R.; Ecker, G.F. Similarity based descriptors—Useful classification of substrates of the human multidrug transporter P-glycoprotein? QSAR Comb. Sci 2009, 28, 834–839. [Google Scholar]
- Cabrera, M.A.; González, I.; Fernandez, C.; Navarro, C.; Bermejo, M. A topological substructural approach for the prediction of P-glycoprotein substrates. J. Pharm. Sci 2006, 95, 589–606. [Google Scholar]
- Tropsha, A. Best practices for QSAR model development, validation, and exploitation. Mol. Inf 2010, 29, 476–488. [Google Scholar]
- Yap, C.W. PaDEL-Descriptor: An open source software to calculate molecular descriptors and fingerprints. J. Comput. Chem 2010, 32, 1466–1474. [Google Scholar]
- PubChem Substructure Fingerprint Web Site. Available onlone: ftp://ftp.ncbi.nlm.nih.gov/pubchem/specifications/pubchem_fingerprints.txt accessed on 6 June 2012.
- Gobbi, A.; Lee, M.-L. DISE: Directed Sphere Exclusion. J. Chem. Inf. Comput. Sci 2003, 43, 317–323. [Google Scholar]
- Coi, A.; Massarelli, I.; Saraceno, M.; Carli, N.; Testai, L.; Calderone, V.; Bianucci, A.M. Quantitative structure-activity relationship models for predicting biological properties, developed by combining structure- and ligand-based approaches: an application to the human Ether-a-go-go Related Gene potassium channel inhibition. Chem. Biol. Drug. Des 2009, 74, 416–433. [Google Scholar]
- Hall, M.A. Correlation-based Feature Subset Selection for Machine Learning Hamilton. Ph.D. Dissertation, The University of Waikato, Hamilton, New Zealand, April 1999. [Google Scholar]
- Hall, M.; Frank, E.; Holmes, G.; Pfahringer, B.; Reutemann, P.; Witten, I.H. The WEKA data mining software: An update. SIGKDD Explor 2009, 11, 10–18. [Google Scholar]
- Quinlan, R. C4.5: Programs for Machine Learning; Morgan Kaufmann Publishers: San Mateo, CA, USA, 1993; pp. 55–77. [Google Scholar]
- Matthews, B.V. Comparison of the predicted and observed secondary structure of T4 phage lysozyme. Biochim. Biophys. Acta 1975, 405, 442–451. [Google Scholar]
- Smeeton, N.C. Early history of the kappa statistic. Biometrics 1985, 41, 795. [Google Scholar]
- Fawcett, T. An introduction to ROC analysis. Pattern Recognit. Lett 2006, 27, 861–874. [Google Scholar]
- Netzeva, T.I.; Worth, A.P.; Aldenberg, T.; Benigni, R.; Cronin, M.T.D.; Gramatica, P.; Jaworska, J.S.; Kahn, S.; Klopman, G.; Marchant, C.A.; et al. Current status of methods for defining the applicability domain of (quantitative) structure-activity relationships. ATLA 2005, 33, 155–173. [Google Scholar]
Definition | P-gp Inhibition | ATPase activation | Efflux |
---|---|---|---|
“True” inhibitor | Y | N | N |
Substrate | Y | Y | Y |
Non-substrate | N | N | N |
Model | LOO-cross validation statistics | Test set statistics | |||||||
---|---|---|---|---|---|---|---|---|---|
TP | TN | Acc | MCC | K | AUC | TP | TN | Acc | |
RT(S9 K10) | 62.5 | 87.0 | 76.9 | 0.51 | 0.51 | 0.75 | 90.0 | 50.0 | 70.0 |
RT(S10 K11) | 68.8 | 82.6 | 76.9 | 0.52 | 0.52 | 0.76 | 60.0 | 80.0 | 70.0 |
RT(S1 K11) | 68.8 | 78.3 | 74.4 | 0.47 | 0.47 | 0.74 | 90.0 | 70.0 | 80.0 |
C4.5 | 37.5 | 65.2 | 53.8 | 0.19 | 0.48 | 0.52 | 60.0 | 50.0 | 55.0 |
Model | LOO-cross validation statistics | Test set statistics | |||||||
---|---|---|---|---|---|---|---|---|---|
TP | TN | Acc | MCC | K | AUC | TP | TN | Acc | |
RT(S5 K3) | 84.2 | 80 | 82.1 | 0.64 | 0.64 | 0.82 | 80.0 | 60.0 | 70.0 |
RT(S10 K2) | 73.7 | 80 | 76.9 | 0.54 | 0.54 | 0.77 | 60.0 | 80.0 | 70.0 |
RT(S10000 K8) | 73.7 | 70 | 71.8 | 0.44 | 0.44 | 0.72 | 70.0 | 80.0 | 75.0 |
C4.5 | 89.5 | 75 | 82.1 | 0.65 | 0.64 | 0.86 | 60.0 | 50.0 | 55.0 |
Model | LOO-cross validation statistics | Test set statistics | |||||||
---|---|---|---|---|---|---|---|---|---|
TP | TN | Acc | MCC | K | AUC | TP | TN | Acc | |
RT(S80 K15) | 79.2 | 60 | 71.8 | 0.40 | 0.40 | 0.70 | 80.0 | 70.0 | 75.0 |
RT(S20 K4) | 83.3 | 66.7 | 76.9 | 0.51 | 0.51 | 0.75 | 80.0 | 60.0 | 70.0 |
RT(S80 K14) | 83.3 | 60 | 74.4 | 0.45 | 0.44 | 0.72 | 70.0 | 60.0 | 65.0 |
RT(S30 K4) | 70.3 | 73.3 | 71.8 | 0.44 | 0.43 | 0.72 | 80.0 | 50.0 | 65.0 |
RT(S1000 K14) | 79.2 | 66.7 | 74.4 | 0.46 | 0.46 | 0.73 | 60.0 | 70.0 | 65.0 |
C4.5 | 75.0 | 40.0 | 61.5 | 0.16 | 0.16 | 0.41 | 100.0 | 40.0 | 70.0 |
Inhibition | |
---|---|
Model | n° of Molecular Descriptors involved |
RT(S9 K10) | 6 |
RT(S10 K11) | 5 |
RT(S1 K11) | 5 |
Molecular descriptor | n° of models in which the descriptor is involved |
XLogP | 3 |
AMR | 3 |
nBondsS3 | 3 |
Ghose-Crippen LogKow | 3 |
TopoPSA | 2 |
PubchemFP544 | 1 |
C1SP2 | 1 |
ATPase activation | |
Model | n° of Molecular Descriptors involved |
RT(S5 K3) | 10 |
RT(S10 K2) | 9 |
RT(S10000 K8) | 7 |
Molecular descriptor | n° of models in which the descriptor is involved |
TopoPSA | 3 |
MLFER_E | 3 |
n6Ring | 3 |
nHBDon | 3 |
nHBAcc | 3 |
nT6Ring | 3 |
nRing | 2 |
PubchemFP256 | 2 |
PubchemFP392 | 1 |
Pubchem FP437 | 1 |
PubchemFP495 | 1 |
PubchemFP592 | 1 |
PubchemFP607 | 1 |
Monolayer efflux | |
Model | n° of Molecular Descriptors involved |
RT(S30 K4) | 10 |
RT(S20 K4) | 9 |
RT(S80 K14) | 8 |
RT(S80 K15) | 7 |
RT(S1000 K14) | 6 |
Molecular descriptor | n° of models in which the descriptor is involved |
nBondsS3 | 5 |
MLFER_E | 5 |
XLogP | 4 |
C3SP2 | 4 |
MLFER_A | 4 |
Ghose-Crippen LogKow | 4 |
AMR | 3 |
C2SP2 | 2 |
nHBAcc | 2 |
nRing | 2 |
Mannhold LogP | 1 |
C2SP3 | 1 |
PubchemFP299 | 1 |
PubchemFP737 | 1 |
SubFPC275 | 1 |
LOO-cross validation statistics | Test set statistics | ||||||||
---|---|---|---|---|---|---|---|---|---|
Model | TP (Y) | TN (N) | Acc | MCC | K | AUC | TP (Y) | TN (N) | Acc |
Inhibition | 68.8 | 95.7 | 84.6 | 0.67 | 0.67 | 0.82 | 90.0 | 70.0 | 80.0 |
ATPase activation | 78.9 | 75 | 76.9 | 0.54 | 0.54 | 0.84 | 70.0 | 80.0 | 75.0 |
Monolayer efflux | 83.3 | 66.7 | 76.9 | 0.44 | 0.51 | 0.75 | 80.0 | 70.0 | 75.0 |
External set statistics | |||
---|---|---|---|
Model | TP (Y) | TN (N) | Accuracy |
Inhibition | 77.1 | 83.3 | 77.8 |
ATPase activation | 75.0 | 71.4 | 72.3 |
Monolayer efflux | 74.4 | 100.0 | 76.6 |
Compound | Inhibitor | ATPase activator | Efflux | Compound | Inhibitor | ATPase activator | Efflux |
---|---|---|---|---|---|---|---|
Amantadine | N | N | N | Testosterone | Y | N | N |
Chlorpheniramine | N | N | N | Chlorpromazine | Y | Y | N |
Doxorubicin | N | N | N | Ketoconazole | Y | Y | N |
Itraconazole | N | N | N | Mebendazole | Y | Y | N |
Lidocaine | N | N | N | Midazolam | Y | Y | N |
Mannitol | N | N | N | Nicardipine | Y | Y | N |
Methotrexate | N | N | N | Nifedipine | Y | Y | N |
Practolol | N | N | N | Nitrendipine | Y | Y | N |
Propranolol | N | N | N | Verapamil | Y | Y | N |
Pyridostigmine | N | N | N | Chloroquine | N | N | Y |
Ranitidine | N | N | N | Cimetidine | N | N | Y |
Sumatriptan | N | N | N | Colchicine | N | N | Y |
Triamterene | N | N | N | Daunorubicin | N | N | Y |
Yohimbine | N | N | N | Dexamethasone | N | N | Y |
Amprenavir | Y | Y | Y | Etoposide | N | N | Y |
Diltiazem | Y | Y | Y | Hoechst 33342 | N | N | Y |
Dipyridamole | Y | Y | Y | Mitoxantrone | N | N | Y |
Loperamide | Y | Y | Y | Neostigmine | N | N | Y |
Loratadine | Y | Y | Y | Puromycin | N | N | Y |
Monensin | Y | Y | Y | Vincristine | N | N | Y |
Nelfinavir | Y | Y | Y | Vinorelbine | N | N | Y |
Prazosin | Y | Y | Y | Clarythromycin | N | Y | Y |
Quinidine | Y | Y | Y | Eletriptan | N | Y | Y |
Reserpine | Y | Y | Y | Emetine | N | Y | Y |
Ritonavir | Y | Y | Y | Erythromycin | N | Y | Y |
Saquinavir | Y | Y | Y | Indinavir | N | Y | Y |
Terfenadine | Y | Y | Y | Taxol | N | Y | Y |
Vinblastine | Y | Y | Y | Trimethoprim | N | Y | Y |
Elacridar | Y | N | N | Cyclosporin A | Y | N | Y |
GW420867 | Y | N | N |
Compound | Inhibitor | ATPase activator | Efflux | Compound | Inhibitor | ATPase activator | Efflux |
---|---|---|---|---|---|---|---|
I_8b | Y | N | Y | III_7c | Y | Y | Y |
II_11a | Y | N | N | III_7d | Y | Y | Y |
II_11b | Y | N | Y | III_8a | Y | Y | Y |
II_13a | Y | N | Y | III_8b | Y | Y | Y |
II_13b | Y | N | Y | III_8c | Y | Y | Y |
II_14a | Y | N | N | III_8d | Y | Y | Y |
II_14b | Y | Y | Y | III_9a | Y | N | Y |
II_15a | Y | N | N | III_9b | Y | N | Y |
II_15b | Y | N | Y | III_9c | Y | N | Y |
II_16a | Y | N | Y | III_9d | Y | N | Y |
II_16b | Y | N | Y | III_10a | Y | N | Y |
II_17b | Y | Y | N | III_10b | Y | N | Y |
II_23 | Y | N | Y | III_10c | Y | N | Y |
II_25 | Y | N | Y | III_10d | Y | N | Y |
II_26 | Y | N | Y | III_11a | Y | N | Y |
II_27 | Y | N | Y | III_11b | Y | N | Y |
III_5a | Y | N | Y | III_11c | Y | N | Y |
III_5b | Y | N | Y | III_11d | Y | N | Y |
III_5c | Y | N | Y | III_12a | Y | Y | Y |
III_5d | Y | N | Y | III_12b | Y | N | Y |
III_6b | Y | N | Y | III_12c | Y | N | Y |
III_6c | Y | N | Y | III_12d | Y | Y | Y |
III_7a | N | Y | Y | MV181 | N | N | Y |
III_7b | Y | Y | Y |
Descriptor Type | Descriptor ID | Class |
---|---|---|
AcidicGroupCount | nAcid | 2D |
ALOGP | ALogP, ALogP2, AMR | 2D |
APol | apol | 2D |
Aromatic atoms count | naAromAtom | 2D |
Aromatic bonds count | nAromBond | 2D |
Atom count | nAtom, nHeavyAtom, nH, nB, nC, nN, nO, nS, nP, nF, nCl, nBr, nI | 2D |
BasicGroupCount | nBase | 2D |
BondCount | nBonds, nBonds2, nBondsS, nBondsS2, nBondsS3, nBondsD, nBondsD2, nBondsT, nBondsQ | 2D |
BPol | bpol | 2D |
Carbon types | C1SP1, C2SP1, C1SP2, C2SP2, C3SP2, C1SP3, C2SP3, C3SP3, C4SP3 | 2D |
HBondAcceptorCount | nHBAcc, nHBAcc2, nHBAcc3, nHBAcc_Lipinski | 2D |
HBondDonorCount | nHBDon, nHBDon_Lipinski | 2D |
LargestChain | nAtomLC | 2D |
LargestPiSystem | nAtomP | 2D |
LongestAliphaticChain | nAtomLAC | 2D |
MannholdLogP | MLogP | 2D |
McGowanVolume | McGowan_Volume | 2D |
MLFER | MLFER_A, MLFER_BH, MLFER_BO, MLFER_S, MLFER_E, MLFER_L | 2D |
Ring count | nRing, n3Ring, n4Ring, n5Ring, n6Ring, n7Ring, n8Ring, n9Ring, n10Ring, n11Ring, n12Ring, nG12Ring, nFRing, nF4Ring, nF5Ring, nF6Ring, nF7Ring, nF8Ring, nF9Ring, nF10Ring, nF11Ring, nF12Ring, nFG12Ring, nTRing, nT4Ring, nT5Ring, nT6Ring, nT7Ring, nT8Ring, nT9Ring, nT10Ring, nT11Ring, nT12Ring, nTG12Ring | 2D |
Rotatable bonds count | nRotB | 2D |
Rule of five | LipinskiFailures | 2D |
Topological polar surface area | TopoPSA | 2D |
van der Waals volume | VABC | 2D |
Weight | MW | 2D |
XLogP | XLogP | 2D |
Charged partial surface area | PPSA-1, PPSA-2, PPSA-3, PNSA-1, PNSA-2, PNSA-3, DPSA-1, DPSA-2, DPSA-3, FPSA-1, FPSA-2, FPSA-3, FNSA-1, FNSA-2, FNSA-3, WPSA-1, WPSA-2, WPSA-3, WNSA-1, WNSA-2, WNSA-3, RPCG, RNCG, RPCS, RNCS, THSA, TPSA, RHSA, RPSA | 3D |
Moment of inertia | MOMI-X, MOMI-Y, MOMI-Z, MOMI-XY, MOMI-XZ, MOMI-YZ, MOMI-R | 3D |
Pubchem fingerprint | Hierarchal element counts Rings in a canonic Extended Smallest Set of Smallest Rings (ESSSR) ring set Simple atom pairs Simple atom nearest neighbours Detailed atom neighbourhoods Simple SMARTS patterns Complex SMARTS patterns | fingerprint |
Substructure fingerprint | - | fingerprint |
Substructure fingerprint count | - |
Inhibition | ATPase activation | Monolayer efflux |
---|---|---|
Vinblastine | Dipyridamole | Vinblastine |
Terfenadine | Vinblastine | Taxol |
Ritonavir | Taxol | Ritonavir |
Loratadine | Ritonavir | Clarythromycin |
Monensin | Clarithromycin | Indinavir |
Reserpine | Monensin | Emetine |
Nelfinavir | Amprenavir | Dipyridamole |
Dipyridamole | Reserpine | Monensin |
Ketoconazole | Trimethoprim | Reserpine |
Loperamide | Prazosin | Colchicine |
Vincristine | Doxorubicin | Itraconazole |
Taxol | Vincristine | Verapamil |
Vinorelbine | Mitoxantrone | Nicardipine |
Clarithromycin | Etoposide | Yohimbine |
Itraconazole | Methotrexate | Chlorpromazine |
Etoposide | Puromycin | Midazolam |
Daunorubicin | Vinorelbine | Nifedipine |
Mitoxantrone | Triamterene | Methotrexate |
Hoechst 33342 | Mannitol | Testosterone |
Emetine | Cimetidine | Practolol |
© 2012 by the authors; licensee Molecular Diversity Preservation International, Basel, Switzerland. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/).
Share and Cite
Rapposelli, S.; Coi, A.; Imbriani, M.; Bianucci, A.M. Development of Classification Models for Identifying “True” P-glycoprotein (P-gp) Inhibitors Through Inhibition, ATPase Activation and Monolayer Efflux Assays. Int. J. Mol. Sci. 2012, 13, 6924-6943. https://doi.org/10.3390/ijms13066924
Rapposelli S, Coi A, Imbriani M, Bianucci AM. Development of Classification Models for Identifying “True” P-glycoprotein (P-gp) Inhibitors Through Inhibition, ATPase Activation and Monolayer Efflux Assays. International Journal of Molecular Sciences. 2012; 13(6):6924-6943. https://doi.org/10.3390/ijms13066924
Chicago/Turabian StyleRapposelli, Simona, Alessio Coi, Marcello Imbriani, and Anna Maria Bianucci. 2012. "Development of Classification Models for Identifying “True” P-glycoprotein (P-gp) Inhibitors Through Inhibition, ATPase Activation and Monolayer Efflux Assays" International Journal of Molecular Sciences 13, no. 6: 6924-6943. https://doi.org/10.3390/ijms13066924