P450-mediated O-demethylated metabolite is responsible for rat hepatobiliary toxicity of pyridyltriazine-containing PI3K inhibitors.

The dysregulation of phosphatidylinositol 3-kinase (PI3K)-dependent pathways is implicated in several human cancers making it an attractive target for small molecule PI3K inhibitors. A series of potent pyridyltriazine-containing inhibitors of class Ia PI3Ks were synthesized and a subset of compounds was evaluated in exploratory repeat-dose rat toxicology studies. Daily oral dosing of compound 1: in Sprague Dawley rats for four consecutive days was associated with hepatobiliary toxicity that included biliary epithelial hyperplasia and hypertrophy, periductular edema, biliary stasis, and acute peribiliary inflammatory infiltrates. These histological changes were associated with clinical pathology changes that included increased serum liver enzymes, total bile acids, and bilirubin. The predominant clearance pathway of 1: was shown in vitro and in a bile-duct cannulated rat (14)C-ADME study to be P450-mediated oxidative metabolism. An O-demethylated pyridine metabolite, M3: , was identified as a candidate proximal metabolite that caused the hepatotoxicity. Co-administration of the pan-P450 inhibitor 1-aminobenzotriazole with 1: to rats significantly reduced the formation of M3: and prevented liver toxicity, whereas direct administration of M3: reproduced the toxicity. Structural changes were introduced to 1: to make the methoxypyridine ring less susceptible to P450 oxidation (compound 2: ), and addition of a methyl group to the benzylic carbon (compound 3: ) improved the pharmacokinetic profile. These changes culminated in the successful design of a clinical candidate 3: (AMG 511) that was devoid of liver toxicity in a 14-day rat toxicity study. Herein, we describe how a metabolism-based structure-activity relationship analysis allowed for the successful identification of a PI3K inhibitor devoid of off-target toxicity.

Selective class I phosphoinositide 3-kinase inhibitors: optimization of a series of pyridyltriazines leading to the identification of a clinical candidate, AMG 511.

The phosphoinositide 3-kinase family catalyzes the phosphorylation of phosphatidylinositol-4,5-diphosphate to phosphatidylinositol-3,4,5-triphosphate, a secondary messenger which plays a critical role in important cellular functions such as metabolism, cell growth, and cell survival. Our efforts to identify potent, efficacious, and orally available phosphatidylinositol 3-kinase (PI3K) inhibitors as potential cancer therapeutics have resulted in the discovery of 4-(2-((6-methoxypyridin-3-yl)amino)-5-((4-(methylsulfonyl)piperazin-1-yl)methyl)pyridin-3-yl)-6-methyl-1,3,5-triazin-2-amine (1). In this paper, we describe the optimization of compound 1, which led to the design and synthesis of pyridyltriazine 31, a potent pan inhibitor of class I PI3Ks with a superior pharmacokinetic profile. Compound 31 was shown to potently block the targeted PI3K pathway in a mouse liver pharmacodynamic model and inhibit tumor growth in a U87 malignant glioma glioblastoma xenograft model. On the basis of its excellent in vivo efficacy and pharmacokinetic profile, compound 31 was selected for further evaluation as a clinical candidate and was designated AMG 511.

Sequential metabolism of AMG 487, a novel CXCR3 antagonist, results in formation of quinone reactive metabolites that covalently modify CYP3A4 Cys239 and cause time-dependent inhibition of the enzyme.

CYP3A4-mediated biotransformation of (R)-N-(1-(3-(4-ethoxyphenyl)-4-oxo-3,4-dihydropyrido[2,3-d]pyrimidin-2-yl)ethyl)-N-(pyridin-3-ylmethyl)-2-(4-(trifluoromethoxy)phenyl)acetamide (AMG 487) was previously shown to generate an inhibitory metabolite linked to dose- and time-dependent pharmacokinetics in humans. Although in vitro activity loss assays failed to demonstrate CYP3A4 time-dependent inhibition (TDI) with AMG 487, its M2 phenol metabolite readily produced TDI when remaining activity was assessed using either midazolam or testosterone (K(I) = 0.73-0.74 µM, k(inact) = 0.088-0.099 min(-1)). TDI investigations using an IC(50) shift method successfully produced inhibition attributable to AMG 487, but only when preincubations were extended from 30 to 90 min. The shift magnitude was ∼3× for midazolam activity, but no shift was observed for testosterone activity. Subsequent partition ratio determinations conducted for M2 using recombinant CYP3A4 showed that inactivation was a relatively inefficient process (r = 36). CYP3A4-mediated biotransformation of [(3)H]M2 in the presence of GSH led to identification of two new metabolites, M4 and M5, which shifted focus away from M2 being directly responsible for TDI. M4 (hydroxylated M2) was further metabolized to form reactive intermediates that, upon reaction with GSH, produced isomeric adducts, collectively designated M5. Incubations conducted in the presence of [(18)O]H(2)O confirmed incorporation of oxygen from O(2) for the majority of M4 and M5 formed (>75%). Further evidence of a primary role for M4 in CYP3A4 TDI was generated by protein labeling and proteolysis experiments, in which M4 was found to be covalently bound to Cys239 of CYP3A4. These investigations confirmed a primarily role for M4 in CYP3A4 inactivation, suggesting that a more complex metabolic pathway was responsible for generation of inhibitory metabolites affecting AMG 487 human pharmacokinetics.

Novel cytochrome p450 bioactivation of a terminal phenyl acetylene group: formation of a one-carbon loss benzaldehyde and other oxidative products in the presence of N-acetyl cysteine or glutathione.

Compounds 1 (N1-(3-ethynylphenyl)-6-methyl-N5-(3-(6-(methylamino)pyrimidin-4-yl)pyridin-2-yl) isoquinoline-1,5-diamine) and 2 (N-(3-ethynylphenyl)-6,7-bis(2-methoxyethoxy)quinazolin-4-amine; Erlotinib/Tarceva) are kinase inhibitors that contain a terminal phenyl acetylene moiety. When incubated in the presence of P450 and NADPH, the anticipated phenyl acetic acid metabolite was formed. When 10 mM of N-acetyl-l-cysteine was added to the incubation mixtures, the phenyl acetic acid product was reduced and at 25 mM or higher concentration of NAC, formation of the phenyl acetic acid was abolished. Instead, the phenyl acetylene moiety lost a carbon and formed a benzaldehyde product. Other oxidation products incorporating one or more equivalents of NAC were also observed. The identities of the metabolites were characterized by MS and NMR. Addition of deferoxamine or ascorbic acid diminished the formation of the NAC influenced products. Similar products were also observed when 1 or 2 were incubated in P450 reactions supplemented with GSH, in Fenton reactions supplemented with NAC or GSH, and in peroxidase reactions supplemented with NAC. We propose the thiols act as a pro-oxidant readily undergoing a one-electron oxidation to form thiyl radicals which in turn initiates the formation of other peroxy radicals that drive the reaction to the observed products. These in vitro findings suggest that one-electron oxidation of thiols may promote the cooxidation of xenobiotic substrates.

Substrate redox potential controls superoxide production kinetics in the cytochrome bc complex.

The Q-cycle mechanism of the cytochrome bc(1) complex maximizes energy conversion during the transport of electrons from ubiquinol to cytochrome c (or alternate physiological acceptors), yet important steps in the Q-cycle are still hotly debated, including bifurcated electron transport, the high yield and specificity of the Q-cycle despite possible short-circuits and bypass reactions, and the rarity of observable intermediates in the oxidation of quinol. Mounting evidence shows that some bypass reactions producing superoxide during oxidation of quinol at the Q(o) site diverge from the Q-cycle rather late in the bifurcated reaction and provide an additional means of studying initial reactions of the Q-cycle. Bypass reactions offer more scope for controlling and manipulating reaction conditions, e.g., redox potential, because they effectively isolate or decouple the Q-cycle initial reactions from later steps, preventing many complications and interactions. We examine the dependence of oxidation rate on substrate redox potential in the yeast cytochrome bc(1) complex and find that the rate limitation occurs at the level of direct one-electron oxidation of quinol to semiquinone by the Rieske protein. Oxidation of semiquinone and reduction of cyt b or O(2) are subsequent, distinct steps. These experimental results are incompatible with models in which the transfer of electrons to the Rieske protein is not a distinct step preceding transfer of electrons to cytochrome b, and with conformational gating models that produce superoxide by different rate-limiting reactions from the normal Q-cycle.

In vitro drug-drug interaction screens for canine veterinary medicines: evaluation of cytochrome P450 reversible inhibition.

Little information regarding the metabolic pathways of pharmaceutical agents administered to dogs, or the inhibition of those metabolic pathways, is available. Without this information, it is difficult to assess how combinations of drugs, whether new or old or approved or nonapproved, may increase the risk for metabolic drug-drug interactions in dogs. Because mammalian xenobiotic metabolism pathways often involve the hepatic cytochrome P450 (P450) monooxgenases, canine liver microsome P450 inhibition screens were tested to evaluate the potential metabolic drug interaction risk of commonly used veterinary medicines. A probe substrate cocktail was developed for four of the five major hepatic canine P450s and used to evaluate their inhibition by 45 canine therapeutic agents in a single-point IC(50) screen. Moderate inhibitors (>25%) were further characterized with an automated ninepoint IC(50) assay that identified ketoconazole, clomipramine, and loperamide as submicromolar CYP2D15 inhibitors. Additional inhibitors belonged to the antiemetic, antimitotic, and anxiolytic therapeutic classes. According to the marker activities, the relative frequency of P450 inhibition by isoform followed the sequence CYP2D15 > CYP2B11 > CYP2C21/41 > CYP3A12/26 > CYP1A1/2. The findings presented suggest there is some overlap in canine and human P450 inhibition specificity. However, occasional differences may give human drugs used off-label in dogs unexpected P450 inhibition profiles and, therefore, cause an unexpected drug-drug interaction risk.