Advancing the use of new approach methodologies for assessing teratogenicity: Building a tiered approach.

Many New Approach Methodologies (NAMs) have been developed for the safety assessment of new ingredients. Research into reproductive toxicity and teratogenicity is a particularly high priority, especially given their mechanistic complexity. Forty-six non-teratogenic and 39 teratogenic chemicals were screened for teratogenic potential using the in silico DART model from the OECD QSAR Toolbox; the devTox quickPredict™ (devTox assay) test and the Zebrafish Embryotoxicity Test (ZET). The sensitivity and specificity were 94.7% and 84.1%, respectively, for the DART tree (83 chemicals), 86.1% and 35.6% for the devTox (81 chemicals) and 77.8% and 76.7% for the ZET (57 chemicals). Fifty-three chemicals were tested in all three assays and when results were combined and based on a "2 out of 3 rule", the sensitivity and specificity were 96.0% and 71.4%, respectively. The specificity of the devTox assay for a sub-set of 43 chemicals was increased from 26.1% to 82.6% by incorporating human plasma concentrations into the assay interpretation. When all 85 chemicals were assessed in a decision tree approach, there was an excellent predictivity and assay robustness of 90%. In conclusion, all three models exhibited a good sensitivity and specificity, especially when outcomes from all three were combined or used in "2 out of 3" or a tiered decision tree approach. The latter is an interesting predictive approach for evaluating the teratogenic potential of new chemicals. Future investigations will extend the number of chemicals tested, as well as explore ways to refine the results and obtain a robust Integrated Testing Strategy to evaluate teratogenic potential.

Comparison of the metabolism of 10 chemicals in human and pig skin explants.

Skin metabolism is important to consider when assessing local toxicity and/or penetration of chemicals and their metabolites. If human skin supply is limited, pig skin can be used as an alternative. To identify any species differences, we have investigated the metabolism of 10 chemicals in a pig and human skin explant model. Phase I metabolic pathways in skin from both species included those known to occur via cytochrome P450s, esterases, alcohol dehydrogenases and aldehyde dehydrogenases. Common Phase II pathways were glucuronidation and sulfation but other conjugation pathways were also identified. Chemicals not metabolized by pig skin (caffeine, IQ and 4-chloroaniline) were also not metabolized by human skin. Six chemicals metabolized by pig skin were metabolized to a similar extent (percentage parent remaining) by human skin. Human skin metabolites were also detected in pig skin incubations, except for one unidentified minor vanillin metabolite. Three cinnamyl alcohol metabolites were unique to pig skin but represented minor metabolites. There were notable species differences in the relative amounts of common metabolites. The difference in the abundance of the sulfate conjugates of resorcinol and 4-amino-3-nitrophenol was in accordance with the known lack of aryl sulfotransferase activity in pigs. In conclusion, while qualitative comparisons of metabolic profiles were consistent between pig and human skin, there were some quantitative differences in the percentage of metabolites formed. This preliminary assessment suggests that pig skin is metabolically competent and could be a useful tool for evaluating potential first-pass metabolism before testing in human-derived tissues.

Comparison of protocols measuring diffusion and partition coefficients in the stratum corneum.

Partition (K) and diffusion (D) coefficients are important to measure for the modelling of skin penetration of chemicals through the stratum corneum (SC). We compared the feasibility of three protocols for the testing of 50 chemicals in our main studies, using three cosmetics-relevant model chemicals with a wide range of logP values. Protocol 1: SC concentration-depth profile using tape-stripping (measures KSC/v and DSC /HSC2 , where HSC is the SC thickness); Protocol 2A: incubation of isolated SC with chemical (direct measurement of KSC/v only) and Protocol 2B: diffusion through isolated SC mounted on a Franz cell (measures KSC/v and DSC /HSC2 , and is based on Fick's laws). KSC/v values for caffeine and resorcinol using Protocol 1 and 2B were within 30% of each other, values using Protocol 2A were ~two-fold higher, and all values were within 10-fold of each other. Only indirect determination of KSC/v by Protocol 2B was different from the direct measurement of KSC/v by Protocol 2A and Protocol 1 for 7-EC. The variability of KSC/v for all three chemicals using Protocol 2B was higher compared to Protocol 1 and 2A. DSC /HSC2 values for the three chemicals were of the same order of magnitude using all three protocols. Additionally, using Protocol 1, there was very little difference between parameters measured in pig and human SC. In conclusion, KSC/v, and DSC values were comparable using different methods. Pig skin might be a good surrogate for human skin for the three chemicals tested. Copyright © 2017 The Authors Journal of Applied Toxicology published by John Wiley & Sons Ltd.

Comparison of protocols for measuring cosmetic ingredient distribution in human and pig skin.

The Cosmetics Europe Skin Bioavailability and Metabolism Task Force aims to improve the measurement and prediction of the bioavailability of topically-exposed compounds for risk assessment. Key parameters of the experimental design of the skin penetration studies were compared. Penetration studies with frozen human and pig skin were conducted in two laboratories, according to the SCCS and OECD 428 guidelines. The disposition in skin was measured 24h after finite topical doses of caffeine, resorcinol and 7-ethoxycoumarin. The bioavailability distribution in skin layers of cold and radiolabelled chemicals were comparable. Furthermore, the distribution of each chemical was comparable in human and pig skin. The protocol was reproducible across the two laboratories. There were small differences in the amount of chemical detected in the skin layers, which were attributed to differences in washing procedures and anatomical sites of the skin used. In conclusion, these studies support the use of pig skin as an alternative source of skin should the availability of human skin become a limiting factor. If radiolabelled chemicals are not available, cold chemicals can be used, provided that the influence of chemical stability, reactivity or metabolism on the experimental design and the relevance of the data obtained is considered.

Patient characteristics associated with risk of first hospital admission and readmission for acute exacerbation of chronic obstructive pulmonary disease (COPD) following primary care COPD diagnosis: a cohort study using linked electronic patient records.

OBJECTIVES: To investigate patient characteristics of an unselected primary care population associated with risk of first hospital admission and readmission for acute exacerbation of chronic obstructive pulmonary disease (AECOPD). DESIGN: Retrospective open cohort using pseudonymised electronic primary care data linked to secondary care data. SETTING: Primary care; Lothian (population approximately 800,000), Scotland. PARTICIPANTS: Data from 7002 patients from 72 general practices with a COPD diagnosis date between 2000 and 2008 recorded in their primary care record. Patients were followed up until 2010, death or they left a participating practice. MAIN OUTCOME MEASURES: First and subsequent admissions for AECOPD (International Classification of Diseases (ICD) 10 codes J44.0, J44.1 in any diagnostic position) after COPD diagnosis in primary care. RESULTS: 1756 (25%) patients had at least 1 AECOPD admission; 794 (11%) had at least 1 readmission and the risk of readmission increased with each admission. Older age at diagnosis, more severe COPD, low body mass index (BMI), current smoking, increasing deprivation, COPD admissions and interventions for COPD prior to diagnosis in primary care, and comorbidities were associated with higher risk of first AECOPD admission in an adjusted Cox proportional hazards regression model. More severe COPD and COPD admission prior to primary care diagnosis were associated with increased risk of AECOPD readmission in an adjusted Prentice-Williams-Peterson model. High BMI was associated with a lower risk of first AECOPD admission and readmission. CONCLUSIONS: Several patient characteristics were associated with first AECOPD admission in a primary care cohort of people with COPD but fewer were associated with readmission. Prompt diagnosis in primary care may reduce the risk of AECOPD admission and readmission. The study highlights the important role of primary care in preventing or delaying a first AECOPD admission.

Vehicle and positive control values from the in vivo rodent comet assay and biomonitoring studies using human lymphocytes: historical database and influence of technical aspects.

There is increased interest in the in vivo comet assay in rodents as a follow-up approach for determining the biological relevance of chemicals that are genotoxic in in vitro assays. This is partly because, unlike other assays, DNA damage can be assessed in this assay in virtually any tissue. Since background levels of DNA damage can vary with the species, tissue, and cell processing method, a robust historical control database covering multiple tissues is essential. We describe extensive vehicle and positive control data for multiple tissues from rats and mice. In addition, we report historical data from control and genotoxin-treated human blood. Technical issues impacting comet results are described, including the method of cell preparation and freezing. Cell preparation by scraping (stomach and other GI tract organs) resulted in higher % tail DNA than mincing (liver, spleen, kidney etc) or direct collection (blood or bone marrow). Treatment with the positive control genotoxicant, ethyl methanesulfonate (EMS) in rats and methyl methanesulfonate in mice, resulted in statistically significant increases in % tail DNA. Background DNA damage was not markedly increased when cell suspensions were stored frozen prior to preparing slides, and the outcome of the assay was unchanged (EMS was always positive). In conclusion, historical data from our laboratory for the in vivo comet assay for multiple tissues from rats and mice, as well as human blood show very good reproducibility. These data and recommendations provided are aimed at contributing to the design and proper interpretation of results from comet assays.

The Cosmetics Europe strategy for animal-free genotoxicity testing: project status up-date.

The Cosmetics Europe (formerly COLIPA) Genotoxicity Task Force has driven and funded three projects to help address the high rate of misleading positives in in vitro genotoxicity tests: The completed "False Positives" project optimized current mammalian cell assays and showed that the predictive capacity of the in vitro micronucleus assay was improved dramatically by selecting more relevant cells and more sensitive toxicity measures. The on-going "3D skin model" project has been developed and is now validating the use of human reconstructed skin (RS) models in combination with the micronucleus (MN) and Comet assays. These models better reflect the in use conditions of dermally applied products, such as cosmetics. Both assays have demonstrated good inter- and intra-laboratory reproducibility and are entering validation stages. The completed "Metabolism" project investigated enzyme capacities of human skin and RS models. The RS models were shown to have comparable metabolic capacity to native human skin, confirming their usefulness for testing of compounds with dermal exposure. The program has already helped to improve the initial test battery predictivity and the RS projects have provided sound support for their use as a follow-up test in the assessment of the genotoxic hazard of cosmetic ingredients in the absence of in vivo data.

Report from the EPAA workshop: in vitro ADME in safety testing used by EPAA industry sectors.

There are now numerous in vitro and in silico ADME alternatives to in vivo assays but how do different industries incorporate them into their decision tree approaches for risk assessment, bearing in mind that the chemicals tested are intended for widely varying purposes? The extent of the use of animal tests is mainly driven by regulations or by the lack of a suitable in vitro model. Therefore, what considerations are needed for alternative models and how can they be improved so that they can be used as part of the risk assessment process? To address these issues, the European Partnership for Alternative Approaches to Animal Testing (EPAA) working group on prioritization, promotion and implementation of the 3Rs research held a workshop in November, 2008 in Duesseldorf, Germany. Participants included different industry sectors such as pharmaceuticals, cosmetics, industrial- and agro-chemicals. This report describes the outcome of the discussions and recommendations (a) to reduce the number of animals used for determining the ADME properties of chemicals and (b) for considerations and actions regarding in vitro and in silico assays. These included: standardisation and promotion of in vitro assays so that they may become accepted by regulators; increased availability of industry in vivo kinetic data for a central database to increase the power of in silico predictions; expansion of the applicability domains of in vitro and in silico tools (which are not necessarily more applicable or even exclusive to one particular sector) and continued collaborations between regulators, academia and industry. A recommended immediate course of action was to establish an expert panel of users, developers and regulators to define the testing scope of models for different chemical classes. It was agreed by all participants that improvement and harmonization of alternative approaches is needed for all sectors and this will most effectively be achieved by stakeholders from different sectors sharing data.

Metabolite screening of aromatic amine hair dyes using in vitro hepatic models.

Aromatic amines and heterocyclic amines are widely used ingredients in permanent hair dyes. However, little has been published on their potential for oxidation via hepatic cytochrome P450s. Therefore, the authors screened nine such compounds for their potential to undergo oxidative metabolism in human liver microsomes. Toluene-2,5-diamine (TDA), p-aminophenol, m-aminophenol, p-methylaminophenol, N,N'-bis(2-hydroxyethyl)-p-phenylenediamine, and 1-hydroxyethyl-4,5-diaminopyrazole showed no evidence of oxidative metabolism. Oxidized metabolites of 4-amino-2-hydroxytoluene (AHT), 2-methyl-5- hydroxyethylaminophenol (MHEAP), and phenyl methyl pyrazolone (PMP) were detected, but there was no evidence of beta-nicotinamide adenine dinucleotide phosphate (NADPH)-dependent covalent binding to microsomal protein, suggesting that these are not reactive metabolites. Metabolism of AHT, MHEAP, PMP, and TDA was further studied in human hepatocytes. All these compounds underwent conjugation, but no oxidative metabolites were found. The results suggest that none of the hair dye ingredients tested showed evidence of hepatic metabolism to potentially biologically reactive oxidized metabolites.

Comparison of intrinsic metabolic clearance in fresh and cryopreserved human hepatocytes.

1. We compared the intrinsic clearance (CL(int)) of a number of substrates in suspensions of fresh and cryopreserved human hepatocytes from seven donors. 2. CL(int) values for a cocktail incubation of phenacetin, diclofenac, diazepam, bufuralol, midazolam, and hydroxycoumarin were 4.9 +/- 3.4, 18 +/- 7.2, 5.1 +/- 4.9, 6.3 +/- 3.3, 9.8 +/- 5.8 and 22 +/- 14 microl min(-1)/10(6) cells, respectively, and they correlated well with corresponding CL(int) values using cryopreserved hepatocytes from 25 different donors. 3. CL(int) values of each cocktail substrate and 20 AstraZeneca new chemical entities were compared in fresh and cryopreserved hepatocytes from the same three donors. There was a statistically significant correlation between CL(int) in fresh and cryopreserved hepatocytes for each of the three livers (p < 0.002) and the geometric mean of the ratio of fresh to cryopreserved CL(int) values was 1.03. 4. In conclusion, the results add further support to the use of cryopreserved human hepatocytes as a screening model for the intrinsic clearance of new chemical entities.

Skin metabolism of aminophenols: human keratinocytes as a suitable in vitro model to qualitatively predict the dermal transformation of 4-amino-2-hydroxytoluene in vivo.

4-Amino-2-hydroxytolune (AHT) is an aromatic amine ingredient in oxidative hair colouring products. As skin contact occurs during hair dyeing, characterisation of dermal metabolism is important for the safety assessment of this chemical class. We have compared the metabolism of AHT in the human keratinocyte cell line HaCaT with that observed ex-vivo in human skin and in vivo (topical application versus oral (p.o.) and intravenous (i.v.) route). Three major metabolites of AHT were excreted, i.e. N-acetyl-AHT, AHT-sulfate and AHT-glucuronide. When 12.5 mg/kg AHT was applied topically, the relative amounts of each metabolite were altered such that N-acetyl-AHT product was the major metabolite (66% of the dose in comparison with 37% and 32% of the same applied dose after i.v. and p.o. administration, respectively). N-acetylated products were the only metabolites detected in HaCaT cells and ex-vivo whole human skin discs for AHT and p-aminophenol (PAP), an aromatic amine known to undergo N-acetylation in vivo. Since N-acetyltransferase 1 (NAT1) is the responsible enzyme, kinetics of AHT was further compared to the standard NAT1 substrate p-aminobenzoic acid (PABA) in the HaCaT model revealing similar values for K(m) and V(max). In conclusion NAT1 dependent dermal N-acetylation of AHT represents a 'first-pass' metabolism effect in the skin prior to entering the systemic circulation. Since the HaCaT cell model represents a suitable in vitro assay for addressing the qualitative contribution of the skin to the metabolism of topically-applied aromatic amines it may contribute to a reduction in animal testing.

Induction of hepatic cytochrome P450 enzymes: methods, mechanisms, recommendations, and in vitro-in vivo correlations.

Induction of drug-clearance pathways (Phase 1 and 2 enzymes and transporters) can have important clinical consequences. Inducers can (1) increase the clearance of other drugs, resulting in a decreased therapeutic effect, (2) increase the activation of pro-drugs, causing an alteration in their efficacy and pharmacokinetics, and (3) increase the bioactivation of drugs that contribute to hepatotoxicity via reactive intermediates. Nuclear receptors are key mediators of drug-induced changes in the expression of drug-clearance pathways. However, species differences in nuclear receptor activation make the prediction of cytochrome P450 (CYP) induction in humans from data derived from animal models problematic. Thus, in vitro human-relevant model systems are increasingly used to evaluate enzyme induction. In this review, the authors' current understanding of the mechanisms of enzyme induction and the in vitro methods for assessing the induction potential of new drugs will be discussed. Relevant issues and considerations surrounding proper study design and the interpretation of in vitro results will be discussed in light of the current US Food and Drug Administration (FDA) recommendations.

Theory of planned behaviour predictors of intention to use condoms among Xhosa adolescents in South Africa.

HIV/AIDS is taking a heavy toll on South African youth. Reducing their risk for HIV requires an understanding of the determinants of their HIV risk behaviours that are amendable to intervention-induced change. This study draws upon the theory of planned behaviour to identify the modifiable determinants of the intention to use condoms among Xhosa-speaking South African adolescents. The participants were 390 Xhosa-speaking 6th grade students (mean age = 12.1 years) in public schools in the township of Mdantsane, South Africa who completed an anonymous questionnaire. Multiple regression revealed that attitude and perceived behavioural control were significantly related to the intention to use condoms, whereas subjective norm was not, controlling for sexual experience, gender, and language preference. Consistent with this were additional analyses using beliefs as predictors: Hedonistic behavioural beliefs and control beliefs about condom-use negotiation and technical skills predicted intention, whereas normative beliefs did not. The theory of planned behaviour may be a useful model of condom use among Xhosa-speaking South African adolescents. An emphasis on beliefs about the adverse effects of condom use on sexual enjoyment, the ability to negotiate condom use, and the ability to use condoms correctly might improve the efficacy of HIV/STD interventions for such adolescents.

Phase I and II enzyme characterization of two sources of HepG2 cell lines.

1: The metabolism by HepG2 cell from two sources (M1, M2) of 12 substrates is reported: ethoxyresorufin, ethoxycoumarin, testosterone, tolbutamide, chlorzoxazone, dextromethorphan, phenacetin, midazolam, acetaminophen, hydroxycoumarin, p-nitrophenol and 1-chloro-2,4-dinitrobenzene (CDNB), and a pharmaceutical compound, EMD68843. 2: Activities varied markedly. Some were present in M1 (CYP1A, CYP2C9, CYP2E1) but absent in M2. M1 had a more complete set of Phase I enzymes than M2. CYP1A2, CYP2C9, CYP2D6, CYP2E1 and CYP3A activities were present at levels similar to human hepatocytes. Phase II metabolism differed between M1 and M2. M1 conjugated hydroxycoumarin and p-nitrophenol to glucuronides only, whereas M2 produced sulfates. Glutathione conjugation of CDNB metabolism was 10-fold higher in M1 than in M2, but was still much lower than in human hepatocytes. CYP2E, CYP2C, CYP2B6 and CYP3A (but not CYP1A, glucuronyl S-transferase or S-transferase) were inducible in M1. Metabolites of EMD68843, produced by induced (but not uninduced) M1 were the same as those produced in human hepatocytes. 3: In conclusion, HepG2 cells have both Phase I and II enzymes, which activities and at what levels depend on the source and culture conditions. Therefore, HepG2 cells routinely used in in vitro assays should be characterized for their drug-metabolizing capabilities before any results can be fully interpreted.

Metabolism of propafenone and verapamil by cryopreserved human, rat, mouse and dog hepatocytes: comparison with metabolism in vivo.

In the present study we examined the metabolism of [(14)C]propafenone (P) and [(14)C]verapamil (V) using cryopreserved human, dog (Beagle), rat (Sprague-Dawley) and mouse (NMRI) hepatocytes. The percentage ratios of the metabolites were identified after extraction by HPLC with UV and radioactivity detection. Phase-II metabolites were cleaved using beta-glucuronidase. Metabolism of the drugs by cryopreserved hepatocytes was compared with that in the respective species in vivo. All phase-I and -II metabolites known from in vivo experiments: 5-hydroxy-P (5-OH-P); 4'-hydroxy-P (4'-OH-P); N-despropyl-P (NdesP) and the respective glucuronides, were identified after incubation with cryopreserved hepatocytes. Interspecies differences were observed concerning the preferential position of propafenone hydroxylation: 5-OH-P made up 91, 51, 16 and 3% of the total metabolites after incubation with cryopreserved human ( n=4), dog ( n=3), rat ( n=3) and mouse ( n=4) hepatocytes respectively. These results are consistent with interspecies differences known from in vivo experiments. The metabolism of V is more complex than that of P. Nevertheless, all phase-I metabolites known from in vivo experiments and the expected glucuronides were identified after incubation with cryopreserved hepatocytes from all four species. As expected from the results of in vivo experiments, there were no major interspecies differences with respect to phase-I metabolites although the conjugation of verapamil phase-I metabolites by cryopreserved canine hepatocytes was much weaker than for the other species. In conclusion, phase-I and phase-II metabolism of P and V was evaluated using hepatocytes in vitro. All of the relevant interspecies differences known from in vivo experiments were identified after short-term incubation with cryopreserved hepatocytes in suspension.

Studies comparing in vivo:in vitro metabolism of three pharmaceutical compounds in rat, dog, monkey, and human using cryopreserved hepatocytes, microsomes, and collagen gel immobilized hepatocyte cultures.

The in vivo metabolism of three pharmaceutical compounds, EMD68843, EMD96785, and EMD128130, was compared in fresh and cryopreserved hepatocyte (CPH) suspensions and microsomes from rat, dog, monkey, and human livers and fresh human and rat hepatocyte collagen gel immobilized cultures (GICs). Half of the major in vivo metabolites was produced by phase 1 (hydroxylation, oxidation, hydrolysis, N-dealkylation) and half by phase 2 metabolism (mostly glucuronidation but also sulfation and glycine conjugation). The identity and percentage of phase 1 and 2 metabolites from each compound produced in hepatocytes compared well with that in each species in vivo. Glucuronidation was more extensive in GICs than in CPHs. In contrast, CPHs but not GICs, produced sulfate metabolites. Microsomes (supplemented with NADPH only) produced most of the phase 1 but no phase 2 metabolites. Metabolism in CPHs was the same as in fresh hepatocyte suspensions. Discrete species differences in metabolism were detected by CPHs and microsomes. Cytochrome P450 and glucuronosyl S-transferase contents of CPHs did not account for species differences in the percentage of phase 1 and 2 metabolites or the rate of disappearance of the parent compounds in these cells. These data show a good correlation between major metabolites formed in vivo and in vitro. CPHs and GICs, unlike microsomes, carried out sequential phase 1 and 2 metabolism. Each in vitro system has its own advantages, however, for short-term metabolism studies CPHs may be more useful since they are readily available, easier and quicker to prepare than GICs, and have more comprehensive enzyme systems than microsomes.

Metabolic activity of fresh and cryopreserved cynomolgus monkey (Macaca fascicularis) hepatocytes.

1. The effect of cryopreservation on the metabolic capacity of monkey hepatocytes over 4 h in suspension and 24 h in culture was determined. Hepatocytes were diluted in a buffer containing 10% DMSO and frozen in a computer-controlled chamber. 2. Initial ethoxyresorufin and ethoxycoumarin O-deethylase (ECOD) activities were the same in fresh and cryopreserved (CP) hepatocytes. ECOD activity in suspensions declined over 4 h but was the same in fresh and CP hepatocytes. 3. The formation of testosterone hydroxy (OHT) metabolites (namely 6beta-OHT, 2beta-OHT, 16beta-OHT, 16alpha-OHT, 15beta-OHT, 2alpha-OHT and 6beta-OHT) was unaffected by cryopreservation. The loss of OHT activities over 4 h in CP and fresh whole cell suspensions was attributed to a loss of cofactor. CP hepatocyte cultures had equivalent OHT activities to freshly isolated hepatocytes. 4. Initial UDP-glucuronyltransferase (UGT) activities, using the substrates 4-methylumbelliferone, ethoxycoumarin and hydroxycoumarin, were equivalent in fresh and CP whole hepatocytes. At later times, UGT activity was lower in CP than fresh hepatocytes but this was due to a loss of UDPGA. Initial sulphotransferase (SULT) activities, using the substrates 2-naphthol, ethoxycoumarin and hydroxycoumarin, were equivalent in fresh and CP hepatocytes. SULT activities were less stable than UGT activities but were the same in fresh and CP hepatocytes throughout the 4-h incubation. 5. Initial glutathione S-transferase activities (using 1-chloro-2,4-dinitrobenzene) were the same in fresh and CP hepatocytes and both did not decrease over 4 h. 6. CP monkey hepatocytes are a useful model for metabolic and cytotoxicity studies. These cells can be can be used either in suspension or in culture.

Paradoxical activation of Raf by a novel Raf inhibitor.

BACKGROUND: Raf is a proto-oncogene that is activated in response to growth factors or phorbol esters, and is thought to activate MAP kinase kinase-1 (MKK1) and hence the classical MAP kinase (MAPK) cascade. RESULTS: The compound ZM 336372 is identified as a potent and specific inhibitor of Raf isoforms in vitro. Paradoxically, exposure of cells to ZM 336372 induces > 100-fold activation of c-Raf (measured in the absence of compound), but without triggering any activation of MKK1 or p42 MAPK/ERK2. The ZM 336372-induced activation of c-Raf occurs without any increase in the GTP-loading of Ras and is not prevented by inhibition of the MAPK cascade, protein kinase C or phosphatidylinositide 3-kinase. ZM 336372 does not prevent growth factor or phorbol ester induced activation of MKK1 or p42 MAPK/ERK2, or reverse the phenotype of Ras- or Raf-transformed cell lines. The only other protein kinase inhibited by ZM 336372 out of 20 tested was SAPK2/p38. Although ZM 336372 is structurally unrelated to SB 203580, a potent inhibitor of SAPK2/p38, the mutation of Thr106-->Met made SAPK2/p38 insensitive to ZM 336372 as well as to SB 203580. CONCLUSIONS: Raf appears to suppress its own activation by a novel feedback loop, such that inhibition is always counterbalanced by reactivation. These observations imply that some agonists reported to trigger the cellular activation of c-Raf might actually be inhibitors of this enzyme, and that compounds which inhibit the kinase activity of Raf might not be useful as anticancer drugs. The binding sites for ZM 336372 and SB 203580 on Raf and SAPK2/p38 are likely to overlap.