Machine learning for predicting lifespan-extending chemical compounds.

Increasing age is a risk factor for many diseases; therefore developing pharmacological interventions that slow down ageing and consequently postpone the onset of many age-related diseases is highly desirable. In this work we analyse data from the DrugAge database, which contains chemical compounds and their effect on the lifespan of model organisms. Predictive models were built using the machine learning method random forests to predict whether or not a chemical compound will increase Caenorhabditis elegans' lifespan, using as features Gene Ontology (GO) terms annotated for proteins targeted by the compounds and chemical descriptors calculated from each compound's chemical structure. The model with the best predictive accuracy used both biological and chemical features, achieving a prediction accuracy of 80%. The top 20 most important GO terms include those related to mitochondrial processes, to enzymatic and immunological processes, and terms related to metabolic and transport processes. We applied our best model to predict compounds which are more likely to increase C. elegans' lifespan in the DGIdb database, where the effect of the compounds on an organism's lifespan is unknown. The top hit compounds can be broadly divided into four groups: compounds affecting mitochondria, compounds for cancer treatment, anti-inflammatories, and compounds for gonadotropin-releasing hormone therapies.

Effect of OATP-binding on the prediction of biliary excretion.

1. Biliary excretion of compounds is dependant on several transporter proteins for the active uptake of compounds from the blood into the hepatocytes. Organic anion-transporting polypeptides (OATPs) are some of the most abundant transporter proteins in the sinusoidal membrane and have been shown to have substrate specificity similar to the structural characteristics of cholephilic compounds. 2. In this study, we sought to use measures of OATP binding as predictors of biliary excretion in conjunction with molecular descriptors in a quantitative structure-activity relationship (QSAR) study. Percentage inhibitions of three subtypes of OATPs were used as surrogate indicators of OATP substrates. Several statistical modelling techniques were incorporated including classification and regression trees, boosted trees, random forest and multivariate adaptive regression splines (MARS) in order to first develop QSARs for the prediction of OATP inhibition of compounds. The predicted OATP percentage inhibition using selected models were then used as features of the QSAR models for the prediction of biliary excretion of compounds in rat. 3. The results indicated that incorporation of predicted OATP inhibition improves accuracy of biliary excretion models. The best result was obtained from a simple regression tree that used predicted OATP1B1 percentage inhibition at the root node of the tree.

Simultaneous Prediction of four ATP-binding Cassette Transporters' Substrates Using Multi-label QSAR.

Efflux by the ATP-binding cassette (ABC) transporters affects the pharmacokinetic profile of drugs and it has been implicated in drug-drug interactions as well as its major role in multi-drug resistance in cancer. It is therefore important for the pharmaceutical industry to be able to understand what phenomena rule ABC substrate recognition. Considering a high degree of substrate overlap between various members of ABC transporter family, it is advantageous to employ a multi-label classification approach where predictions made for one transporter can be used for modeling of the other ABC transporters. Here, we present decision tree-based QSAR classification models able to simultaneously predict substrates and non-substrates for BCRP1, P-gp/MDR1 and MRP1 and MRP2, using a dataset of 1493 compounds. To this end, two multi-label classification QSAR modelling approaches were adopted: Binary Relevance (BR) and Classifier Chain (CC). Even though both multi-label models yielded similar predictive performances in terms of overall accuracies (close to 70 %), the CC model overcame the problem of skewed performance towards identifying substrates compared with non-substrates, which is a common problem in the literature. The models were thoroughly validated by using external testing, applicability domain and activity cliffs characterization. In conclusion, a multi-label classification approach is an appropriate alternative for the prediction of ABC efflux.

Substrate-specific effects of pirinixic acid derivatives on ABCB1-mediated drug transport.

Pirinixic acid derivatives, a new class of drug candidates for a range of diseases, interfere with targets including PPARα, PPARγ, 5-lipoxygenase (5-LO), and microsomal prostaglandin and E2 synthase-1 (mPGES1). Since 5-LO, mPGES1, PPARα, and PPARγ represent potential anti-cancer drug targets, we here investigated the effects of 39 pirinixic acid derivatives on prostate cancer (PC-3) and neuroblastoma (UKF-NB-3) cell viability and, subsequently, the effects of selected compounds on drug-resistant neuroblastoma cells. Few compounds affected cancer cell viability in low micromolar concentrations but there was no correlation between the anti-cancer effects and the effects on 5-LO, mPGES1, PPARα, or PPARγ. Most strikingly, pirinixic acid derivatives interfered with drug transport by the ATP-binding cassette (ABC) transporter ABCB1 in a drug-specific fashion. LP117, the compound that exerted the strongest effect on ABCB1, interfered in the investigated concentrations of up to 2µM with the ABCB1-mediated transport of vincristine, vinorelbine, actinomycin D, paclitaxel, and calcein-AM but not of doxorubicin, rhodamine 123, or JC-1. In silico docking studies identified differences in the interaction profiles of the investigated ABCB1 substrates with the known ABCB1 binding sites that may explain the substrate-specific effects of LP117. Thus, pirinixic acid derivatives may offer potential as drug-specific modulators of ABCB1-mediated drug transport.

Enzastaurin inhibits ABCB1-mediated drug efflux independently of effects on protein kinase C signalling and the cellular p53 status.

The PKCß inhibitor enzastaurin was tested in parental neuroblastoma and rhabdomyosarcoma cell lines, their vincristine-resistant sub-lines, primary neuroblastoma cells, ABCB1-transduced, ABCG2-transduced, and p53-depleted cells. Enzastaurin IC50s ranged from 3.3 to 9.5 µM in cell lines and primary cells independently of the ABCB1, ABCG2, or p53 status. Enzastaurin 0.3125 µM interfered with ABCB1-mediated drug transport. PKCα and PKCß may phosphorylate and activate ABCB1 under the control of p53. However, enzastaurin exerted similar effects on ABCB1 in the presence or absence of functional p53. Also, enzastaurin inhibited PKC signalling only in concentrations ≥ 1.25 µM. The investigated cell lines did not express PKCß. PKCα depletion reduced PKC signalling but did not affect ABCB1 activity. Intracellular levels of the fluorescent ABCB1 substrate rhodamine 123 rapidly decreased after wash-out of extracellular enzastaurin, and enzastaurin induced ABCB1 ATPase activity resembling the ABCB1 substrate verapamil. Computational docking experiments detected a direct interaction of enzastaurin and ABCB1. These data suggest that enzastaurin directly interferes with ABCB1 function. Enzastaurin further inhibited ABCG2-mediated drug transport but by a different mechanism since it reduced ABCG2 ATPase activity. These findings are important for the further development of therapies combining enzastaurin with ABC transporter substrates.

Karanjin interferes with ABCB1, ABCC1, and ABCG2.

PURPOSE: The prominent ATP-binding cassette (ABC) transporters ABCB1, ABCC1, and ABCG2 are involved in substance transport across physiological barriers and therefore in drug absorption, distribution, and elimination. They also mediate multi-drug resistance in cancer cells. Different flavonoids are known to interfere with different ABC transporters. Here, the effect of the furanoflavonol karanjin, a potential drug with antiglycaemic, gastroprotective, antifungal, and antibacterial effects, was investigated on ABCB1, ABCC1, and ABCG2-mediated drug transport in comparison to the flavonoids apigenin, genistein, and naringenin. METHODS: Cells expressing the relevant transporters (ABCB1: UKF-NB-3(ABCB1), UKF-NB-3(r)VCR¹°; ABCC1: G62, PC-3(r)VCR²°; ABCG2: UKF-NB-3(ABCG2)) were used in combination with specific fluorescent and cytotoxic ABC transporter substrates and ABC transporter inhibitors to study ABC transporter function. Moreover, the effects of the investigated flavonoids were determined on the ABC transporter ATPase activities. RESULTS: Karanjin interfered with drug efflux mediated by ABCB1, ABCC1, and ABCG2 and enhanced the ATPase activity of all three transporters. Moreover, karanjin exerted more pronounced effects than the control flavonoids apigenin, genistein, and naringenin on all three transporters. Most notably, karanjin interfered with ABCB1 at low concentrations being about 1 µM. CONCLUSIONS: Taken together, these findings should be taken into account during further consideration of karanjin as a potential drug for different therapeutic indications. The effects on ABCB1, ABCC1, and ABCG2 may affect the pharmacokinetics of co-administered drugs.

Differential effects of the oncogenic BRAF inhibitor PLX4032 (vemurafenib) and its progenitor PLX4720 on ABCB1 function.

PURPOSE: The clinically approved oncogenic BRAF inhibitor PLX4032 (vemurafenib) was shown to be a substrate of the ATP-binding cassette (ABC) transporter ABCB1. Here, we compared PLX4032 and its structurally closely related precursor compound PLX4720 for their interference with ABCB1 and the ABCB1-mediated compound transport using docking and cell culture experiments. METHODS: For the docking study of PLX4032 and PLX4720 with ABCB1, we analysed binding of both compounds to mouse Abcb1a and to human ABCB1 using a homology model of human ABCB1 based on the 3D structure of Abcb1a. Naturally ABCB1 expressing cells including V600E BRAF-mutated and BRAF wild-type melanoma cells and cells transduced with a lentiviral vector encoding for ABCB1 were used as cell culture models. ABCB1 expression and function were studied by the use of fluorescent and cytotoxic ABCB1 substrates in combination with ABCB1 inhibitors. RESULTS: Docking experiments predicted PLX4032 to interact stronger with ABCB1 than PLX4720. Experimental studies using different cellular models and structurally different ABCB1 substrates confirmed that PLX4032 interfered stronger with ABCB1 function than PLX4720. For example, PLX4032 (20 µM) induced a 4-fold enhanced rhodamine 123 accumulation compared to PLX4720 (20 µM) in ABCB1-transduced UKF-NB-3 cells and reduced the IC50 for the cytotoxic ABCB1 substrate vincristine in this model by 21-fold in contrast to a 9-fold decrease induced by PLX4720. CONCLUSIONS: PLX4032 exerted stronger effects on ABCB1-mediated drug transport than PLX4720. This indicates that small changes in a molecule can substantially modify its interaction with ABCB1, a promiscuous transporter that transports structurally different compounds.

Testing of SNS-032 in a Panel of Human Neuroblastoma Cell Lines with Acquired Resistance to a Broad Range of Drugs.

Novel treatment options are needed for the successful therapy of patients with high-risk neuroblastoma. Here, we investigated the cyclin-dependent kinase (CDK) inhibitor SNS-032 in a panel of 109 neuroblastoma cell lines consisting of 19 parental cell lines and 90 sublines with acquired resistance to 14 different anticancer drugs. Seventy-three percent of the investigated neuroblastoma cell lines and all four investigated primary tumor samples displayed concentrations that reduce cell viability by 50% in the range of the therapeutic plasma levels reported for SNS-032 (<754 nM). Sixty-two percent of the cell lines and two of the primary samples displayed concentrations that reduce cell viability by 90% in this concentration range. SNS-032 also impaired the growth of the multidrug-resistant cisplatin-adapted UKF-NB-3 subline UKF-NB-3(r)CDDP(1000) in mice. ABCB1 expression (but not ABCG2 expression) conferred resistance to SNS-032. The antineuroblastoma effects of SNS-032 did not depend on functional p53. The antineuroblastoma mechanism of SNS-032 included CDK7 and CDK9 inhibition-mediated suppression of RNA synthesis and subsequent depletion of antiapoptotic proteins with a fast turnover rate including X-linked inhibitor of apoptosis (XIAP), myeloid cell leukemia sequence 1 (Mcl-1), baculoviral IAP repeat containing 2 (BIRC2; cIAP-1), and survivin. In conclusion, CDK7 and CDK9 represent promising drug targets and SNS-032 represents a potential treatment option for neuroblastoma including therapy-refractory cases.

Solubility prediction of paracetamol in water-ethanol-propylene glycol mixtures at 25 and 30 degrees C using practical approaches.

The solubility of paracetamol in water-ethanol-propylene glycol binary and ternary mixtures at 25 and 30 degrees C was determined using flask shake method. The generated data extended the solubility database for further computational investigations and also was used to assess the prediction capability of the Jouyban-Acree model. A new version of the model was proposed for modeling the solubility data in water-cosolvent mixtures with the cosolvent concentration of <50% which is required in pharmaceutical formulations. The accuracy of the predicted solubilities was evaluated by the mean percentage deviation (MPD) between the predicted and experimental solubilities. The overall MPD of the Jouyban-Acree model and the log-linear model of Yalkowsky for the entire composition range of the cosolvents were 11.0+/-8.7 and 55.4+/-17.8%, respectively; the corresponding values for the predicted solubilities in mixtures having a cosolvent concentration of <50% were 12.0+/-9.1 and 22.0+/-11.0%.

Comparison of electrotopological-state indices versus atomic charge and superdelocalisability indices in a QSAR study of the receptor binding properties of halogenated estradiol derivatives.

The structural requirements for the binding of steroids to the oestrogen receptor (ER) are important both for the design of new drugs and to assess the health risks of chemicals with ER affinity. In the present QSAR study of receptor binding affinity (RBA) for estradiol derivatives, the atom-level electrotopological state indices have been compared with molecular orbital derived, atom-level parameters for superdelocalisability and atomic charge. The AM1 Hamiltonian was used to calculate molecular orbital parameters. The predictive power of the QSARs indicated that the superdelocalisability indices provide a better model than the electrotopological state indices, and that the atomic charges resulted in the highest prediction error. The most accurate predictions were achieved when other molecular descriptors were also employed in the construction of the QSAR.

The effect of penetration enhancers on drug delivery through skin: a QSAR study.

Skin penetration enhancers are used to allow formulation of transdermal delivery systems for drugs that are otherwise insufficiently skin-permeable. A full understanding of the mode of action could be beneficial for the design of potent enhancers and for the choice of the enhancer to be used in the topical formulation of a special drug. In this study, the structural requirements of penetration enhancers have been investigated using the Quantitative Structure-Activity Relationship (QSAR) technique. Activities of naturally occurring terpenes, pyrrolidinone and N-acetylprolinate derivatives on the skin penetration of 5-fluorouracil, diclofenac sodium (DFS), hydrocortisone (HC), estradiol and benazepril have been considered. The resulting QSARs indicated that for 5-fluorouracil and diclofenac sodium, less hydrophobic enhancers were the most active. More precisely, molecular descriptors in the corresponding QSARs indicated the possible involvement of intermolecular electron donor-acceptor interactions. This was in contrast to the skin permeation promotion of hydrocortisone, estradiol and benazepril by enhancers, where a linear relationship between enhancement activity and n-octanol/water partition coefficients of enhancers was evident. The possible mechanisms of penetration enhancement as suggested by the QSARs will be discussed.