Deep Artificial Neural Networks Reveal a Distributed Cortical Network Encoding Propositional Sentence-Level Meaning.

Understanding how and where in the brain sentence-level meaning is constructed from words presents a major scientific challenge. Recent advances have begun to explain brain activation elicited by sentences using vector models of word meaning derived from patterns of word co-occurrence in text corpora. These studies have helped map out semantic representation across a distributed brain network spanning temporal, parietal, and frontal cortex. However, it remains unclear whether activation patterns within regions reflect unified representations of sentence-level meaning, as opposed to superpositions of context-independent component words. This is because models have typically represented sentences as "bags-of-words" that neglect sentence-level structure. To address this issue, we interrogated fMRI activation elicited as 240 sentences were read by 14 participants (9 female, 5 male), using sentences encoded by a recurrent deep artificial neural-network trained on a sentence inference task (InferSent). Recurrent connections and nonlinear filters enable InferSent to transform sequences of word vectors into unified "propositional" sentence representations suitable for evaluating intersentence entailment relations. Using voxelwise encoding modeling, we demonstrate that InferSent predicts elements of fMRI activation that cannot be predicted by bag-of-words models and sentence models using grammatical rules to assemble word vectors. This effect occurs throughout a distributed network, which suggests that propositional sentence-level meaning is represented within and across multiple cortical regions rather than at any single site. In follow-up analyses, we place results in the context of other deep network approaches (ELMo and BERT) and estimate the degree of unpredicted neural signal using an "experiential" semantic model and cross-participant encoding.SIGNIFICANCE STATEMENT A modern-day scientific challenge is to understand how the human brain transforms word sequences into representations of sentence meaning. A recent approach, emerging from advances in functional neuroimaging, big data, and machine learning, is to computationally model meaning, and use models to predict brain activity. Such models have helped map a cortical semantic information-processing network. However, how unified sentence-level information, as opposed to word-level units, is represented throughout this network remains unclear. This is because models have typically represented sentences as unordered "bags-of-words." Using a deep artificial neural network that recurrently and nonlinearly combines word representations into unified propositional sentence representations, we provide evidence that sentence-level information is encoded throughout a cortical network, rather than in a single region.

Predictability of meaning in grammatical encoding: Optional plural marking.

The markedness principle plays a central role in linguistic theory: marked grammatical categories (like plural) tend to receive more linguistic encoding (e.g., morphological marking), while unmarked categories (like singular) tend to receive less linguistic encoding. What precisely makes a grammatical category or meaning marked, however, remains unclear. One prominent proposal attributes markedness to the frequency or predictability of meanings: infrequent or less predictable meanings are more likely to receive extra linguistic encoding than frequent or more predictable meanings. Existing support for the predictability account is limited to correlational evidence, leaving open whether meaning predictability can cause markedness patterns. We present two miniature language learning experiments that directly assess effects of predictability on morphological plural marking. We find that learners preferentially produce plural marking on nouns that are less probable to occur with plural meaning-despite the fact that no such pattern was present in learners' input. This suggests that meaning predictability can cause the markedness patterns like those that are cross-linguistically observed.

Multiple Regions of a Cortical Network Commonly Encode the Meaning of Words in Multiple Grammatical Positions of Read Sentences.

Deciphering how sentence meaning is represented in the brain remains a major challenge to science. Semantically related neural activity has recently been shown to arise concurrently in distributed brain regions as successive words in a sentence are read. However, what semantic content is represented by different regions, what is common across them, and how this relates to words in different grammatical positions of sentences is weakly understood. To address these questions, we apply a semantic model of word meaning to interpret brain activation patterns elicited in sentence reading. The model is based on human ratings of 65 sensory/motor/emotional and cognitive features of experience with words (and their referents). Through a process of mapping functional Magnetic Resonance Imaging activation back into model space we test: which brain regions semantically encode content words in different grammatical positions (e.g., subject/verb/object); and what semantic features are encoded by different regions. In left temporal, inferior parietal, and inferior/superior frontal regions we detect the semantic encoding of words in all grammatical positions tested and reveal multiple common components of semantic representation. This suggests that sentence comprehension involves a common core representation of multiple words' meaning being encoded in a network of regions distributed across the brain.

Absorption, distribution, metabolism and elimination of 14C-ETX0914, a novel inhibitor of bacterial type-II topoisomerases in rodents.

1. ETX0914 is a novel bacterial topoisomerase inhibitor that has a novel mode-of-inhibition and is in clinical development for the treatment of infections caused by Neisseria gonorrhoeae. 2. The in vitro biotransformation studies of ETX0914 using mouse, rat, dog and human hepatocytes showed moderate intrinsic clearance in mouse and rat and low intrinsic clearance in dog and human. 3. Following intravenous administration of [14C]-ETX0914 in rats, the mean recovery of administered dose in urine, bile and feces was approximately 15%, 55% and 24%, respectively. Unchanged ETX0914 recovered in urine and bile was less than 5% of the dose, indicating that ETX0914 underwent extensive metabolism in rats. Metabolites M1, M2, M4, M6 and M12 detected in both rat and mouse urine samples were not detected in mouse urine when predosed with 1-aminobenzotriazole, indicating that these metabolites were cytochrome P450 mediated products. The major fecal metabolites observed in rats were not formed when ETX0914 was incubated with fresh feces from germ free rats under sterile condition or in incubations with rat intestinal microsome and cytosol, suggesting that most likely ETX0914 was directly excreted into gut lumen where metabolites were formed as intestinal microflora-mediated products. The major sites of metabolism by CYP enzymes were in the morpholine and oxazolidinone rings while it was benzisoxazole reduction with the gut microflora.

Fusidic Acid Inhibits Hepatic Transporters and Metabolic Enzymes: Potential Cause of Clinical Drug-Drug Interaction Observed with Statin Coadministration.

Fusidic acid (FA), which was approved in the 1960s in many European and Asian countries, has gained renewed interest due to its continued effectiveness against methicillin-resistant Staphylococcus aureus As rhabdomyolysis has been reported upon coadministration of FA with statins, we aimed to elucidate the underlying molecular mechanisms that contribute to FA-statin drug-drug interactions. Because of the association between rhabdomyolysis and increased exposure to statins, we investigated if cytochrome P450 (CYP) enzymes and transporters involved in the disposition of various statins are inhibited by FA. FA was found to inhibit BCRP and OATP1B1 but not P-gp. In overexpressing cell systems, FA inhibited BCRP-mediated efflux (50% inhibitory concentration [IC50], ∼50 to 110 µM) and OATP1B1-mediated uptake (IC50, ∼4 to 35 µM) of statins at clinically relevant concentrations achievable in the intestine and liver (based on a 550-mg oral dose of FA, the expected maximum theoretical gastrointestinal concentration is ∼4 mM, and the maximum total or unbound concentration in the inlet to the liver was reported to be up to 223 µM or 11 µM, respectively, upon multiple dosing). Similarly, FA inhibited metabolism of statins in human liver microsomes (IC50, ∼17 to 195 µM). These data suggest that FA inhibits at least 3 major dispositional pathways (BCRP, OATP1B1, and CYP3A) and thus affects the clearance of several statins. We confirmed that FA is eliminated via phase 1 metabolism (primarily via CYP3A); however, there is also some phase 2 metabolism (mediated primarily by UGT1A1). Taken together, these data provide evidence for molecular mechanisms that may explain the occurrence of rhabdomyolysis when FA is administered with statins.

Pharmacokinetics, metabolism and excretion of [(14)C]-lanicemine (AZD6765), a novel low-trapping N-methyl-d-aspartic acid receptor channel blocker, in healthy subjects.

1.(1S)-1-phenyl-2-(pyridin-2-yl)ethanamine (lanicemine; AZD6765) is a low-trapping N-methyl-d-aspartate (NMDA) channel blocker that has been studied as an adjunctive treatment in major depressive disorder. The metabolism and disposition of lanicemine was determined in six healthy male subjects after a single intravenous infusion dose of 150 mg [(14)C]-lanicemine. 2.Blood, urine and feces were collected from all subjects. The ratios of Cmax and AUC(0-∞) of lanicemine to plasma total radioactivity were 84 and 66%, respectively, indicating that lanicemine was the major circulating component with T1/2 at 16 h. The plasma clearance of lanicemine was 8.3 L/h, revealing that lanicemine is a low-clearance compound. The mean recovery of radioactivity from urine was 93.8% of radioactive dose. 3.In urine samples, 10 metabolites of lanicemine were identified. Among which, an O-glucuronide conjugate (M1) was the most abundant metabolite (∼11% of the dose in excreta). In plasma, the circulatory metabolites were identified as a para-hydroxylated metabolite (M1), an O-glucuronide (M2), an N-carbamoyl glucuronide (M3) and an N-acetylated metabolite (M6). The average amount of each of metabolite was less than 4% of total radioactivity detected in plasma or urine. 4.In conclusion, lanicemine is a low-clearance compound. The unchanged drug and metabolites are predominantly eliminated via urinary excretion.

Metabolism of a G protein-coupled receptor modulator, including two major 1,2,4-oxadiazole ring-opened metabolites and a rearranged cysteine-piperazine adduct.

Metabolites of a G protein-coupled receptor modulator containing 1,2,4-oxadiazole and piperazine substructures were identified in vitro in human, rat, and dog hepatocyte incubates and in vivo in rat plasma, bile, urine, and feces by using 14C-radiolabeled parent compound. Exposure coverage for the major circulating metabolites in humans at steady state and in preclinical species used in drug safety assessments was determined by using pooled plasma samples collected from a human multiple ascending dose study and a 3-month rat toxicokinetic study. Metabolites M1 and M2, which were formed by opening of the 1,2,4-oxadiazole ring, were observed as major metabolites both in vitro and in vivo across species. The carboxylic acid metabolite M2 was presumably formed through reductive N-O bond cleavage of the oxadiazole ring and subsequent hydrolysis. However, the mechanism for the formation of the unusual N-cyanoamide metabolite M1 remains uncertain. Neither M1 nor M2 had any target activity, as did parent drug P. In rat bile, rearranged Cys-piperazine and Gly-Cys-piperazine adducts, involving the formation of a five-membered heteroaromatic imidazole derivative from a six-membered piperazine ring, were observed as minor metabolites. These findings support a previously reported mechanism regarding glutathione detoxification for piperazine bioactivation products.

A clinical study to assess CYP1A2 and CYP3A4 induction by AZD7325, a selective GABA(A) receptor modulator - an in vitro and in vivo comparison.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT: • AZD7325 is an orally administered, potent, selective gamma-amino-butyric acid (GABA(A) ) α2,3 receptor modulator intended for the treatment of anxiety. • The induction effects of AZD7325 on CYP1A2 and CYP3A4 have not been systematically studied. WHAT THIS STUDY ADDS: • The in vitro studies showed that AZD7325 was a moderate CYP1A2 inducer and potent CYP3A4 inducer. • The follow-up clinical studies in healthy volunteers demonstrated that the expected efficacious daily dose of AZD7325 only weakly induced the pharmacokinetics of the CYP3A4 sensitive substrate, midazolam, and had no effect on the pharmacokinetics of the CYP1A2 substrate, caffeine. There was no apparent change in AZD7325 exposure following co-administration of midazolam or caffeine compared with AZD7325 alone. • The study demonstrated that clinical exposure of the inducer plays a critical role in the determination of cytochrome P450 induction risk of a drug candidate. AIM(S): To investigate the potential of AZD7325 to induce CYP1A2 and CYP3A4 enzyme activities. METHODS: Induction of CYP1A2 and CYP3A4 by AZD7325 was first evaluated using cultured human hepatocytes. The effect of multiple doses of 10 mg AZD7325 on the pharmacokinetics of midazolam and caffeine was then examined in healthy subjects. RESULTS: The highest CYP1A2 and CYP3A4 induction responses were observed in human hepatocytes treated with 1 or 10 µm of AZD7325, in the range of 17.9%-54.9% and 76.9%-85.7% of the positive control responses, respectively. The results triggered the further clinical evaluation of AZD7325 induction potential. AZD7325 reached a plasma C(max) of 0.2 µm after 10 mg daily dosing to steady-state. AZD7325 decreased midazolam geometric mean AUC by 19% (0.81-fold, 90% CI 0.77, 0.87), but had no effect on midazolam C(max) (90% CI 0.82, 0.97). The mean CL/F of midazolam increased from 62 l h(-1) (midazolam alone) to 76 l h(-1) when co-administered with AZD7325. The AUC and C(max) of caffeine were not changed after co-administration of AZD7325, with geometric mean ratios (90% CI) of 1.17 (1.12, 1.23) and 0.99 (0.95, 1.03), respectively. CONCLUSIONS: While AZD7325 appeared to be a potent CYP3A4 inducer and a moderate CYP1A2 inducer from in vitro studies, the expected efficacious dose of AZD7325 had no effect on CYP1A2 activity and only a weak inducing effect on CYP3A4 activity. This comparison of in vitro and in vivo results demonstrates the critical role that clinical exposure plays in evaluating the CYP induction risk of a drug candidate.

In vitro and in vivo metabolism of a selective δ-opioid receptor.

4-({4-[(2-hydroxy-ethyl)-methyl-carbamoyl]-phenyl}-quinolin-8-yl-methylene)-1-thiazol-4-ylmethyl-piperidinium (compound I) is a selective agonist of δ-opioid receptor developed for the treatment of depressive and anxiety disorders. The in vitro biotransformation studies using rat, dog, and human hepatocytes showed that the metabolites detected in human hepatocytes were also found in either rat or dog hepatocytes. M1 (N-dealkylation), M2 (N-demethylation), and M4 (carboxylic acid metabolite) were major phase I metabolites observed in all three species. Human CYP3A4/5 isoenzymes were identified to be the primary enzymes responsible for the formation of M1 and M2 in human liver microsomes. After single oral administration of [¹4C]compound I, the major elimination route for [(¹4C]compound I and its metabolites in rat was through feces with 92.9% recovery. The results from the bile duct-cannulated study revealed that a minimum of 51% of administered dose was absorbed in rats. The pharmacokinetic analysis using unlabeled parent drug showed that compound I was rapidly absorbed and exhibited a mean apparent terminal half-life of approximately 2.7 h. A total of 15 metabolites of compound I were detected and profiled in rat urine, bile, and feces. In rat bile, compound I accounted for <1.5% of the excreted dose, suggesting that compound I underwent extensive metabolism before elimination. The structures of metabolites were elucidated by high-resolution tandem mass spectrometry. M1, M4, and M6 were the most abundant metabolites observed in rat bile. Only a low level of parent [¹4C]compound I was observed in rat plasma.

Disposition and metabolism of ticagrelor, a novel P2Y12 receptor antagonist, in mice, rats, and marmosets.

Ticagrelor is a reversibly binding and selective oral P2Y(12) antagonist, developed for the prevention of atherothrombotic events in patients with acute coronary syndromes. The disposition and metabolism of [(14)C]ticagrelor was investigated in mice, rats, and marmosets to demonstrate that these preclinical toxicity species showed similar metabolic profiles to human. Incubations with hepatocytes or microsomes from multiple species were also studied to compare with in vivo metabolic profiles. The routes of excretion were similar for both oral and intravenous administration in mice, rats, and marmosets with fecal excretion being the major elimination pathway accounting for 59 to 96% of the total radioactivity administered. Urinary excretion of drug-related material accounted for only 1 to 15% of the total radioactivity administered. Milk samples from lactating rats displayed significantly higher levels of total radioactivity than plasma after oral administration of ticagrelor. This demonstrated that ticagrelor and/or its metabolites were readily transferred into rat milk and that neonatal rats could be exposed to ticagrelor-related compounds via maternal milk. Ticagrelor and active metabolite AR-C124910 (loss of hydroxyethyl side chain) were the major components in plasma from all species studied and similar to human plasma profiles. The primary metabolite of ticagrelor excreted in urine across all species was an inactive metabolite, AR-C133913 (loss of difluorophenylcyclopropyl group). Ticagrelor, AR-C124910, and AR-C133913 were the major components found in feces from the three species examined. Overall, in vivo metabolite profiles were qualitatively similar across all species and consistent with in vitro results.

Liquid chromatography-tandem mass spectrometry method for measurement of nicotine N-glucuronide: a marker for human UGT2B10 inhibition.

Nicotine is considered to be a specific substrate for UGT2B10, an isoform of human uridine diphosphate glucuronosyltransferase (UGT). In the present study, a sensitive and selective liquid chromatography/tandem mass spectrometry (LC-MS-MS) method for quantification of nicotine N-glucuronide in pooled human liver microsomal incubates was developed and validated. Proteins in a 200µL aliquot of incubation solution were precipitated by adding 40µL 35% perchloric acid. The overall extraction efficiency was greater than 98%. Nicotine N-glucuronide and internal standard were recorded using selected reaction monitoring in positive ion electrospray with ion transitions of m/z 339-163 and m/z 342-166, respectively. The linear calibration curve was obtained over the concentration range of 10-1000nM, with a lower limit of quantification of 10nM. The intra-day and inter-day precision (% CV) and accuracy (% bias) of the method were within 15% at all quality control levels. Nicotine glucuronide in processed samples was stable for 24h at room temperature and 48h at 4°C based on the stability experiments performed in this study. This established method was employed to evaluate the inhibitory effects of five target compounds including amitriptyline, hecogenin, imipramine, lamotrigine, and trifluoperazine on enzymatic activity of UGT2B10. IC(50) values for inhibition of nicotine N-glucuronidation by amitriptyline, imipramine, lamotrigine, and trifluoperazine were calculated. Trifluoperazine was found to be a non-substrate inhibitor for human UGT2B10.

In vitro metabolism of α7 neuronal nicotinic receptor agonist AZD0328 and enzyme identification for its N-oxide metabolite.

1. AZD0328 was pharmacologically characterized as a α7 neuronal nicotinic receptor agonist intended for treatment of Alzheimer's disease. In vitro AZD0328 cross species metabolite profile and enzyme identification for its N-oxide metabolite were evaluated in this study. 2. AZD0328 was very stable in the human hepatocyte incubation, whereas extensively metabolized in rat, dog and guinea pig hepatocyte incubations. The N-oxidation metabolite (M6) was the only metabolite detected in human hepatocyte incubations, and it also appeared to be the major in vitro metabolic pathway in a number of preclinical species. In addition, N-glucuronide metabolite of AZD0328 was observed in human liver microsomes. 3. Other metabolic pathways in the preclinical species include hydroxylation in azabicyclo octane or furopyridine part of the molecule. Pyridine N-methylation of AZD0328 (M2) was identified as a dog specific metabolite, not observed in human or other preclinical species. 4. Multiple enzymes including CYP2D6, CYP3A4/5, FMO1 and FMO3 catalyzed AZD0328 metabolism. The potential for AZD0328 to be inhibited clinically by co-administered drugs or genetic polymorphism is relative low.

Absorption, excretion, and metabolism of a potential GABA-A α2/3 receptor modulator in rats.

4-Amino-8-(2,5-dimethoxyphenyl)-N-propylcinnoline-3-carboxamide (AZD6280) is a selective GABA-A(α2/3) receptor modulator under development for the treatment of generalized anxiety disorders. Absorption, metabolism, and excretion of [(14)C]-AZD6280 was studied in rats following a single oral (7 mg/kg) or intravenous (i.v., 1 mg/kg) administration of [(14)C]-AZD6280. The results from the bile duct-cannulated study revealed that AZD6280 was well-absorbed in rats. The pharmacokinetic analysis was conducted using unlabelled parent drug that was rapidly absorbed (plasma T(max) ~1 h) and exhibited a mean apparent terminal half-life of ~4.2 h. The overall mean recoveries in the excreta were 98.6% and 100.3% after oral and i.v. administration of [(14)C]-AZD6280, respectively. The major route for elimination of [(14)C]-AZD6280 and its metabolites was through faeces. The radiochromatographic analysis of the excreta demonstrated that AZD6280 underwent extensive biotransformation. A total of 28 metabolites of AZD6280 were detected and profiled in urine, bile, and faeces in this study. The structures of metabolites were elucidated by high-resolution tandem mass spectrometry. Similar metabolite profiles were observed in rats given AZD6280 orally or intravenously.

In vitro evaluation of potential drug-drug interactions with ticagrelor: cytochrome P450 reaction phenotyping, inhibition, induction, and differential kinetics.

Ticagrelor is an orally administered, antiplatelet agent that inhibits the prothrombotic effects of ADP on the platelet by antagonizing the P2Y(12) receptor. Ticagrelor is a reversibly binding direct-acting P2Y(12) antagonist and does not require metabolic activation to achieve its antiplatelet effect. CYP3A4 and CYP3A5 appear to be the enzymes predominantly responsible for the formation of the ticagrelor active and inactive metabolites, AR-C124910XX and AR-C133913XX. The apparent K(m) values in human liver microsomes are 27.0 and 38.8 µM, with V(max) values of 730 and 417 pmol/min/mg for AR-C124910XX and AR-C133913XX, respectively. Ticagrelor moderately inhibited CYP2C9 activity in human liver microsomes with an IC(50) of 10.5 µM, while exhibiting little or no inhibition of CYP1A2, CYP2B6, CYP2C8, CYP2C19, CYP2D6, and CYP2E1. In human liver microsomes, ticagrelor inhibited midazolam 4-hydroxylation with an IC(50) of 8.2 µM, while activating 1'-hydroxylation of midazolam. Studies with recombinant enzymes suggested that cytochrome b(5) and CYP3A4 interactions play a significant role in this differential kinetic behavior. Evaluated in fresh human hepatocytes at concentration up to 20 µM, ticagrelor was not an inducer of CYP1A2 or CYP3A4. Although ticagrelor exhibited a tendency for CYP2B6 and CYP2C9 induction, its potential to cause drug interactions via the induction of these enzymes is low when its exposure at a therapeutic dose is considered.

In vitro assessment of metabolic drug­drug interaction potential of AZD2624, neurokinin-3 receptor antagonist, through cytochrome P(450) enzyme identification, inhibition, and induction studies.

AZD2624 was pharmacologically characterized as a NK3 receptor antagonist intended for treatment of schizophrenia. The metabolic drug-drug interaction potential of AZD2624 was evaluated in in vitro studies. CYP3A4 and CYP3A5 appeared to be the primary enzymes mediating the formation of pharmacologically active ketone metabolite (M1), whereas CYP3A4, CYP3A5, and CYP2C9 appeared to be the enzymes responsible for the formation of the hydroxylated metabolite (M2). The apparent K(m) values were 1.5 and 6.3 µM for the formation of M1 and M2 in human liver microsomes, respectively. AZD2624 exhibited an inhibitory effect on microsomal CYP3A4/5 activities with apparent IC(50) values of 7.1 and 19.8 µM for midazolam and testosterone assays, respectively. No time-dependent inactivation of CYP3A4/5 activity (midazolam 1'-hydroxylation) by AZD2624 was observed. AZD2624 demonstrated weak to no inhibition of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, and CYP2D6. AZD2624 was not an inducer of CYP1A2 or CYP2B6. Although AZD2624-induced CYP3A4 activity in hepatocytes, the potential of AZD2624 to cause inductive drug interactions of this enzyme was low at relevant exposure concentration. Together with targeted low efficacious concentration, the results of this study demonstrated AZD2624 has a relatively low metabolic drug-drug interaction potential towards co-administered drugs. However, metabolism of AZD2624 might be inhibited when co-administrated with potent CYP3A4/5 inhibitors.

Expression and characterization of dog cytochrome P450 2A13 and 2A25 in baculovirus-infected insect cells.

Dog CYP2A13 and CYP2A25 were coexpressed with dog NADPH-cytochrome P450 reductase (OR) in baculovirus-infected Sf9 insect cells. CYP2A13 effectively catalyzed 7-ethoxycoumarin (7EC) deethylation and coumarin hydroxylation with apparent K(m) values of 4.8 and 2.1 microM, respectively, similar to those observed using dog liver microsomes (7.5 and 0.75 microM, respectively). CYP2A25 exhibited much lower affinity toward 7EC, with an apparent K(m) value of 150 microM, which indicates that CYP2A13 plays a more significant role in the metabolism of these CYP2A substrates. Similar to the dog CYP1A2 enzyme, CYP2A13 efficiently catalyzed phenacetin deethylation with a K(m) value of 3.9 microM, which suggests that phenacetin is not a selective probe for dog CYP1A2 activity. Both dog CYP2A13 and CYP2A25 exhibited little or no catalytic activity toward other common cytochrome P450 probe substrates, including bupropion, amodiaquine, diclofenac, S-mephenytoin, bufuralol, dextromethorphan, midazolam, and testosterone. These results provided additional information about the selectivity of these commonly used probe substrates.

Role of human UGT2B10 in N-glucuronidation of tricyclic antidepressants, amitriptyline, imipramine, clomipramine, and trimipramine.

The role of human UDP glucuronosyltransferase (UGT) 2B10 in the N-glucuronidation of a number of tricyclic antidepressants was investigated and compared with that of UGT1A4 in both the Sf9 expressed system and human liver microsomes. The apparent K(m) (S(50)) values for the formation of quaternary N-glucuronides of amitriptyline, imipramine, clomipramine, and trimipramine were 2.60, 16.8, 14.4, and 11.2 microM in UGT2B10 and 448, 262, 112, and 258 microM in UGT1A4, respectively. The kinetics of amitriptyline and imipramine glucuronidation in human liver microsomes exhibited a biphasic character, where the high- and low-affinity components were in good agreement with our results in expressed UGT2B10 and UGT1A4, respectively. The kinetics of clomipramine and trimipramine glucuronidation in human liver microsomes were sigmoidal in nature, and the S(50) values were similar to those found for expressed UGT1A4. The in vitro clearances (CL(int) or CL(max)) were comparable between UGT2B10 and UGT1A4 for glucuronidation of imipramine, clomipramine, and trimipramine, whereas CL(int) of amitriptyline glucuronidation by UGT2B10 was more than 10-fold higher than that by UGT1A4. Nicotine was found to selectively inhibit UGT2B10 but not UGT1A4 activity. At a low tricyclic antidepressant concentration, nicotine inhibited their glucuronidation by 33 to 50% in human liver microsomes. Our results suggest that human UGT2B10 is a high-affinity enzyme for tricyclic antidepressant glucuronidation and is likely to be a major UGT isoform responsible for the glucuronidation of these drugs at therapeutic concentrations in vivo.

The conduct of in vitro studies to address time-dependent inhibition of drug-metabolizing enzymes: a perspective of the pharmaceutical research and manufacturers of America.

Time-dependent inhibition (TDI) of cytochrome P450 (P450) enzymes caused by new molecular entities (NMEs) is of concern because such compounds can be responsible for clinically relevant drug-drug interactions (DDI). Although the biochemistry underlying mechanism-based inactivation (MBI) of P450 enzymes has been generally understood for several years, significant advances have been made only in the past few years regarding how in vitro time-dependent inhibition data can be used to understand and predict clinical DDI. In this article, a team of scientists from 16 pharmaceutical research organizations that are member companies of the Pharmaceutical Research and Manufacturers of America offer a discussion of the phenomenon of TDI with emphasis on the laboratory methods used in its measurement. Results of an anonymous survey regarding pharmaceutical industry practices and strategies around TDI are reported. Specific topics that still possess a high degree of uncertainty are raised, such as parameter estimates needed to make predictions of DDI magnitude from in vitro inactivation parameters. A description of follow-up mechanistic experiments that can be done to characterize TDI are described. A consensus recommendation regarding common practices to address TDI is included, the salient points of which include the use of a tiered approach wherein abbreviated assays are first used to determine whether NMEs demonstrate TDI or not, followed by more thorough inactivation studies for those that do to define the parameters needed for prediction of DDI.

Comparison of methods for the prediction of the metabolic sites for CYP3A4-mediated metabolic reactions.

Predictions of the metabolic sites for new chemical entities, synthesized or only virtual, are important in the early phase of drug discovery to guide chemistry efforts in the synthesis of new compounds with reduced metabolic liability. This information can now be obtained from in silico predictions, and therefore, a thorough and unbiased evaluation of the computational techniques available is needed. Several computational methods to predict the metabolic hot spots are emerging. In this study, metabolite identification using MetaSite and a docking methodology, GLUE, were compared. Moreover, the published CYP3A4 crystal structure and computed CYP3A4 homology models were compared for their usefulness in predicting metabolic sites. A total of 227 known CYP3A4 substrates reported to have one or more metabolites adding up to 325 metabolic pathways were analyzed. Distance-based fingerprints and four-point pharmacophore derived from GRID molecular interaction fields were used to characterize the substrate and protein in MetaSite and the docking methodology, respectively. The CYP3A4 crystal structure and homology model with the reactivity factor enabled achieved a similar prediction success (78%) using the MetaSite method. The docking method had a relatively lower prediction success (approximately 57% for the homology model), although it still may provide useful insights for interactions between ligand and protein, especially for uncommon reactions. The MetaSite methodology is automated, rapid, and has relatively accurate predictions compared with the docking methodology used in this study.

Effects of cytochrome P450 3A modulators ketoconazole and carbamazepine on quetiapine pharmacokinetics.

AIMS: To explore the potential for drug interactions on quetiapine pharmacokinetics using in vitro and in vivo assessments. METHODS: The CYP enzymes responsible for quetiapine metabolite formation were assessed using recombinant expressed CYPs and CYP-selective inhibitors. P-glycoprotein (Pgp) transport was tested in MDCK cells expressing the human MDR1 gene. The effects of CYP3A4 inhibition were evaluated clinically in 12 healthy volunteers that received 25 mg quetiapine before and after 4 days of treatment with ketoconazole 200 mg daily. To assess CYP3A4 induction in vivo, 18 patients with psychiatric disorders were titrated to steady-state quetiapine levels (300 mg twice daily), then titrated to 600 mg daily carbamazepine for 2 weeks. RESULTS: CYP3A4 was found to be responsible for formation of quetiapine sulfoxide and N- and O-desalkylquetiapine and not a Pgp substrate. In the clinical studies, ketoconazole increased mean quetiapine plasma C(max) by 3.35-fold, from 45 to 150 ng ml(-1) (mean C(max) ratio 90% CI 2.51, 4.47) and decreased its clearance (Cl/F) by 84%, from 138 to 22 l h(-1) (mean ratio 90% CI 0.13, 0.20). Carbamazepine decreased quetiapine plasma C(max) by 80%, from 1042 to 205 ng ml(-1) (mean C(max) ratio 90% CI 0.14, 0.28) and increased its clearance 7.5-fold, from 65 to 483 l h(-1) (mean ratio 90% CI 6.04, 9.28). CONCLUSIONS: Cytochrome P450 3A4 is a primary enzyme responsible for the metabolic clearance of quetiapine. Quetiapine pharmacokinetics were affected by concomitant administration of ketoconazole and carbamazepine, and therefore other drugs and ingested natural products that strongly modulate the activity or expression of CYP3A4 would be predicted to change exposure to quetiapine.