In order to investigate the structural features responsible for the cytotoxicity of the naturally occurring lignan nordihydroguaiaretic acid, the synthesis of four structural analogues of NDGA is proposed for the purpose of studying their oxidative metabolism. One analogue in particular (1), a mono-catechol analogue, is successfully synthesized employing a double Stobbe condensation approach. Following synthesis of this compound a series of oxidation experiments is performed consisting of: incubation in rat liver microsomes with and without the trapping agent glutathione (GSH), oxidation with mushroom tyrosinase, oxidation with silver oxide, and oxidation with horseradish peroxidase. Results are analyzed via HPLC and UPLC-MS. It is found that 1 does not autoxidize at pH 7.4 as NDGA does. Two products are produced during incubation of 1 in rat liver microsomes with UPLC-ESI(-)-MS results giving m/z of 879.2 and 574.18. This is consistent with 1 plus 2 GSH and 1 plus 1 GSH respectively; confirming 1 will oxidize to an electrophilic moiety. Oxidation with mushroom tyrosinase is found to produce high levels of product two with m/z 574.2. Oxidation with horseradish peroxidase is found to produce high levels of the m/z 879.2 product. Silver Oxide produced multiple products rather than the expected one major product, but most are found to be inconsistent with the products seen during rat liver microsomal incubation, and are not pursued.

A major limitation in developing a successful cancer treatment is the need for a distinction between normal and cancerous tissue. For solid tumors, this distinction can be made on a spatial basis, and successful treatments have been developed accordingly; however, many of these treatments cause pathologies in healthy tissue, much to the detriment of patient health. To address this issue for solid tumours, a conceptual approach would be to administer the chemotherapeutic drug locally, such that the intra-tumour concentration was high, while the systemic exposure to the drug remained low, thus, minimizing side effects. The present research focuses on providing proof-of-concept for the electrochemical generation of a toxicant from a prodrug, and subsequent use to elicit cytotoxicity in cancer cells, in attempts to electrochemically mimic monooxygenase-catalyzed bioactivation. Electro-oxidation of acetaminophen and cyclophosphamide substrates at graphite and Ti/RuO2 anodes was successful in generating their respective toxicants; however, the graphite anode was superior with respect to current efficiency and toxicant yield. Electrolyses conducted in batch and flow reactors produced effluents which reduced EMT-6 cell viability to the IC99 level. This thesis provides proof-of-concept for electrochemical prodrug activation as a viable area for further cancer research.

A screening method was developed for identification of catalytically active enzymes in combinatorial cDNA libraries of mutated glutathione transferase (GST) derivatives expressed in E. coli. The method is based on spraying monochlorobimane (MCB) directly over bacterial colonies growing on agar. The substrate MCB become fluorescent under UV light, when the bacterial colony contains active GSTs catalyzing the conjugation with endogenous glutathione. Eleven out of twelve GSTs investigated where active with MCB. This method can be used to screen libraries generated from most cytosolic GSTs in the search for proteins with altered functions and structures. Azathioprine (Aza), a thiopurine that has been used clinically for 40 years was investigated with 14 GSTs. Three enzymes showed prominent catalytic activities with Aza and all of them are highly expressed in the liver. We estimated the contribution of the three enzymes GSTs A1-1, A2-2 and M1-1 bioactivation of Aza in the liver and concluded that it was about 2 orders of magnitude more effective than the uncatalyzed reaction. GST bioactivation of Aza could clarify aspects of idiosyncratic reactions observed in some individuals. Two other thiopurine prodrugs, cis-acetylvinylthiopurine (cAVTP) and trans-acetylvinylthioguanine (tAVTG), were investigated with the same 14 GSTs. The results displayed diverse catalytic activities. A mechanism of consecutive reactions was proposed. The studies contribute to knowledge under what conditions the drug should optimally be administered. A study of the same prodrugs with several mutants from the Mu class characterized by a point mutation of a hypervarible residue. We conclude that the effects of the mutations were qualitatively parallel for cAVTP and tAVTG, but they vary significantly in magnitude; steric hindrance may interfere with transition-state stabilization. From the evolutionary perspective the data show that a point mutation can alternatively enhance or attenuate the activity with a particular substrate and illustrate the functional plasticity of GSTs.

<p>Accumulating evidence is suggesting that exposure to some environmental contaminants may alter adipogenesis, resulting in accumulation of adipocytes, and often significant weight gain. Thus these types of contaminants are often referred to as obesogens. Many of these contaminants act via the activation (i.e. agonism) of the peroxisome proliferator activated receptor &#947; (PPAR&#947;) nuclear receptor. To date, very few chemicals have been identified as possible PPAR&#61543; ligands. In the thesis, our goal was to determine the PPAR&#947; ligand binding potency and activation of several groups of major semi-volatile organic compounds (SVOCs) that are ubiquitously detected in indoor environments, including flame retardants such as polybrominated diphenyl ethers (PBDEs) and Firemaster 550 (FM550), and other SVOCs such as phthalates, organotins, halogenated phenols and bisphenols. Additional attention was also given to the potential activity of the major metabolites of several of these compounds. Since the primary sink for many of these SVOCs is dust, and dust ingestion has been confirmed as an important pathway for SVOCs accumulation in humans, the potential PPAR&#61543; binding and activation in extracts from environmentally relevant dust samples was also investigated. </p><p> Previous studies have also shown that SVOCs sorbed to organic matrices (e.g., soil and sediment), were only partially bioaccessible (bioavailable), but it was unclear how bioaccessible these compounds are from indoor dust matrices. In addition, bioactivation of SVOCs (via metabolism) could exacerbate their PPAR&#61543; potency. Therefore, to adequately assess the potential risk of PPAR&#947; activation from exposure to SVOC mixtures in house dust, it is essential that one also investigates the bioaccessibility and bioactivation of these chemicals following ingestion. </p><p> In the first research aim of this thesis, the bioaccessibility and bioactivation of several important SVOCs in house dust was investigated. To accomplish this, Tenax beads (TA) encapsulated within a stainless steel insert were used as an infinite adsorption sink to estimate the dynamic absorption of a suite of flame retardants (FRs) commonly detected in indoor dust samples, and from a few polyurethane foam samples for comparison. Experimental results demonstrate that the bioaccessibility and stability of FRs following ingestion varies both by chemical and by matrix. Organophosphate flame retardants (OPFRs) had the highest estimated bioaccessibility (~80%) compared to brominated compounds (e.g. PBDEs), and values generally decreased with increasing Log Kow, with <30% bioaccessibility measured for the most hydrophobic compound tested, BDE209. In addition, the stability of the more labile SVOCs that contained ester groups (e.g. OPFRs and 2-ethylhexyl-tetrabromo-benzoate (TBB)) were examined in a simulated digestive fluid matrix. No significant changes in the OPFR concentrations were observed in this fluid; however, TBB was found to readily hydrolyze to tetrabromobenzoic acid (TBBA) in the intestinal fluid in the presence of lipases. </p><p> In research aims 2 and 3, two commercially available high-throughput bioassays, a fluorescence polarization PPAR&#61543; ligand binding assay (PolarScreenTM PPAR&#947;-competitor assay kit, Invitrogen, Aim 2) and a PPAR&#61543; reporter gene assay (GeneBLAzer PPAR&#947; non-DA Assay, Invitrogen, Aim 3) were used to investigate the binding potency and activation of several groups of SVOCs and dust extracts with human PPAR&#947; LBD; respectively. In the PPAR&#61543; binding assay (Aim 2), most of the tested compounds exhibited dose-dependent binding to PPAR&#947;. Mono(2-ethylhexyl) tetrabromophthalate (TB-MEHP), halogenated bisphenol/phenols, triphenyl phosphate and hydroxylated PBDEs were found to be potent or moderate PPAR&#947; ligands, based on the measured ligand binding dissociation constant (Kd). The most potent compound was 3-OH-BDE47, with an IC50 of 0.24 &#956;M. The extent of halogenation and the position of the hydroxyl group strongly affected binding. Of the dust samples tested, 21 of 24 samples showed significant PPAR&#61543; binding potency at a concentration of 3 mg dust equivalents (DEQ)/mL. In the PPAR&#61543; reporter assay (Aim 3), many SVOCs or their metabolites were either confirmed (based on previous reports) or for the first time were found to be potential PPAR&#947; agonists with various potency and efficacy. We also observed that 15 of 25 dust extracts examined showed an activation percentage more than 8% (calculated activation threshold) of the maximal activation induced by rosiglitazone (positive control). In some cases, activation was as high as 50% of the rosiglitazone activation for the dust extracts with the highest efficacy. Furthermore, the correlation between the reporter assay and the ligand binding assay among the house dust extracts was significant and positive (r = 0.7, p < 0.003), suggesting the binding potency was predicting activation. In research aim 2, the effect of bioactivation on the PPAR&#947; binding potency was also investigated. In vitro bioactivation of house dust extracts incubated with rat and human hepatic S9 fractions was used to investigate the role of in vivo biotransformation on PPAR gamma activity. The result showed that metabolism may lead to an increased binding affinity, as a 3-16% increase in PPAR&#947; binding activity was observed following bioactivation of the dust extracts.</p><p> In research aim 4, an effect-directed analysis (EDA) was used to identify compounds likely contributing to the observed PPAR&#61543; activity among the dust extract. Three dust extracts which showed significant PPAR&#61543; activity with approximately 25, 30, and 50% of the maximal response induced by rosiglitazone at the highest efficacy were fractionated using normal phase high-performance liquid chromatography (NP-HPLC) and each fraction was individually tested for PPAR&#61543; activity. Active fractions were then analyzed using gas-chromatography mass spectrometry (GC-MS) and possible compounds identified. Three dust extracts showed a similar PPAR&#61543; activity distribution among the NP-HPLC fractions. In the most active fractions, fatty acids (FAs) were identified as the most active chemicals. The concentrations of four FAs were measured in the house dust extracts, and the concentrations were found to be highly correlated with the observed PPAR&#61543; activity. These four FAs were also tested for PPAR&#61543; activity and found to be partial PPAR&#61543; agonists, particularly oleic and myristic acid. To tentatively identify sources of FAs, FAs in human/animal hair, dead skin cells, and two brands of cooking oil were analyzed. We found the same FAs in those samples and there concentrations were relatively abundant, ranging from 186 to 14,868 µg/g. Therefore, these results suggest that FAs are likely responsible for the observed PPAR&#61543; activity in indoor dust. Also, this is the first study reporting on the level of FAs in dust samples. The source of these FAs in dust may be either from the cooking or accumulation of human/animal cells in indoor dust.</p><p> In conclusion, this research demonstrates that many SVOCs ubiqutiously detected in house dust, and/or their metabolites, can be weak or moderate PPAR&#61543; ligands. In addition, chemical mixtures in house dust can effectively bind to and activate PPAR&#61543;. However, our results suggest FAs are probably responsible for these observations, and likely outcompeting the synthetic environmental contaminants present in the dust extract. Furthermore, bioactivation of contaminants present in house dust can potentially increase their affinity for PPAR&#61543;. And lastly, the bioaccessibility and stability of SVOCs in house dust after ingestion are likely to modulate the PPAR&#61543; activity in the environmental mixtures and should be considered in future risk assessments.</p> / Dissertation

<p>I developed a precision-cut adrenal slice culture procedure to investigate cytochrome P450 (CYP) catalysed irreversible binding and adrenocorticolytic effects in human, rodent, and fish adrenal tissue, <i>ex vivo</i>. Autoradiography and radioluminography of exposed tissue slices showed that the potent adrenal toxicant 3-methylsulphonyl-2,2´-bis(4-chlorophenyl)-1,1´-dichloroethene (MeSO₂-DDE) causes a selective metabolite binding in <i>zona fasciculata</i> (<i>ZF</i>), which is diminished by the CYP11B1 inhibitor metyrapone. MeSO₂-DDE also reduces corticosterone secretion, increases 11-deoxycorticosterone secretion and causes mitochondrial degeneration in <i>ZF</i> cells in cultured mouse adrenal slices. ACTH treatment of mice induces CYP11B1 and increases irreversible MeSO₂-DDE binding and toxicity in <i>ZF</i> cells. Metyrapone-sensitive binding of MeSO₂-DDE is also observed in human <i>zona fasciculata/reticularis</i> (<i>ZF/ZR</i>) and 11-deoxycorti- sol/corticosterone secretion increases in MeSO₂-DDE-exposed cultured human adrenal slices. The adrenocorticolytic drug 2-(2-chlorophenyl)-2-(4-chlorophenyl)-1,1-dichlorethane (o,p´- DDD, Mitotane^®) is also bound in <i>ZF/ZR </i>but does not to impair hormone secretion in human adrenal slices at equimolar concentration. A targeted, presumably CYP1B1-catalysed irreversible binding of the adrenocorticolytic carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) in <i>ZF/ZR </i>occurs in rat adrenal slices, whereas presumably CYP1A1-catalysed irreversible binding in endothelial cells is observed in CYP1-induced rats and mice. The rat-specific adrenocorticolytic activity of DMBA may rely on two independent pathological processes resulting in cell death and haemorrhage in the adrenal cortex. In Atlantic cod, selective binding of o,p´-DDD is observed in interrenal cells in cultured anterior kidney slices.</p><p>In conclusion, precision-cut adrenal slice culture is a simple <i>ex vivo</i> test system with which to investigate CYP-catalysed metabolite binding, alteredsteroid hormone secretion and target cell ultrastructure in human, experimental and wild animal tissue. The results imply that organisms under stress could be at increased risk of MeSO₂-DDE induced adrenal toxicity. MeSO₂-DDE is an expected human adrenal toxicant, which should be evaluated as a possible alternative in the therapy of adrenocortical hypersecretion and tumour growth.</p>

Die Enzymsuperfamilie der löslichen Sulfotransferasen (SULT) spielt eine wichtige Rolle in der Phase II des Fremdstoffmetabolismus. Sie katalysieren den Transfer einer Sulfonylgruppe auf nucleophile Gruppen endogener und exogener Substrate. Die Sulfokonjugation von Fremdstoffen erhöht deren Wasserlöslichkeit und behindert die passive Permeation von Zellmembranen. Dadurch wird die Ausscheidung dieser konjugierten Substanzen erleichtert. In Abhängigkeit von der Struktur des Zielmoleküls kann die Sulfokonjugation aber auch zur metabolischen Aktivierung von Fremdstoffen durch die Bildung instabiler Metabolite führen. Die SULT-vermittelte Aktivierung promutagener Substanzen ist somit von toxikologischem Interesse. Für die Detektion SULT-vermittelter Mutagenität mittels bakterieller in-vitro Testsysteme ist die heterologe Expression der fremdstoffmetabolisierenden Enzyme direkt in den Indikatorzellen notwendig. S. typhimurium exprimieren selbst keine SULT, und externe Metabolisierungssysteme sind problematisch, weil die negativ geladenen, kurzlebigen Metabolite nur schlecht die Zellmembran penetrieren können. Die Expression humaner Enyme in Bakterien ist jedoch zum Teil sehr kritisch. So zeigen z.B. sehr ähnliche Enzyme (SULT1A2*1 und *2) deutliche Unterschiede im Expressionsniveau bei exakt gleichen äußeren Bedingungen. Dies erschwert den Vergleich der enzymatischen Aktivitäten dieser Enzyme im in-vitro Testsystem. Andere Enzyme (z.B. SULT2B1b) werden unter Verwendung ihrer Wildtyp-cDNA zum Teil nicht detektierbar exprimiert. Deshalb sollte in dieser Arbeit eine Methode zur Optimierung der heterologen Expression fremdstoffmetabolisierender Enzyme für Genotoxizitätsuntersuchungen etabliert werden. <br><br> Es wurde bereits gezeigt dass synonyme Codonaustausche am 5’-Ende der humanen SULT1A2-cDNA zu einer Erhöhung der Expression des entsprechenden Enzyms in S. typhimurium führten. Dementsprechend wurden in dieser Arbeit Codonaustausche am 5’-Ende der cDNA verschiedener SULT (1A1*1, 1A2*1, 2B1b) sowie der Ratten Glutathion-S-Transferase Theta 2 (rGSTT2) und dem Reportergen Luciferase durchgeführt. Die Expression der so generierten Konstrukte wurde in verschiedenen S. typhimurium und E. coli Stämmen quantifiziert und die Aktivität der überexprimierten Enzyme im Ames-Test bzw. im Enzym-Aktivitätsassay überprüft. Durch das Einführen seltener Codons in die cDNA konnte die Proteinexpression von SULT1A1*1, SULT1A2*1 und SULT2B1b maximal 7-fach, 18-fach und 100-fach im Vergleich zur Wildtyp-cDNA gesteigert werden. Die Expression der rGSTT2 wurde ebenfalls durch das Einführen seltener Codons erhöht (maximal 5-fach). Bei dem Reportergen Luciferase jedoch führte das Austauschen häufiger Codons gegen seltene Codons zu einer Reduktion der Proteinexpression um 80 %. Die Expression von Fusionsproteinen aus 2B1b (5’-Ende) und Luciferase (3’-Ende) wurde durch das Einführen seltener Codons ebenfalls um 50 % reduziert. Die S. typhimurium Stämme mit optimierter SULT 1A1*1- bzw. 1A2*1-Expression wurden im Ames-Test eingesetzt und zeigten im Vergleich zu den geringer exprimierenden Stämmen eine höhere Sensitivität. Für SULT2B1b konnte keine Mutagenaktivierung im Ames-Test nachgewiesen werden. Allerdings zeigte ein Enzym-Aktivitätsassay mit Dehydroepiandosteron, dass das bakteriell exprimierte Enzym funktionell war. Da in der Literatur der Effekt seltener Codons auf die Expression in Bakterien bisher fast ausschließlich als inhibitorisch beschrieben wurde, sollte die Wirkungsweise der hier beobachteten Expressionserhöhung durch seltene Codons genauer untersucht werden. Dazu wurden verschiedene Konstrukte der SULT1A2*1 und der SULT2B1b, die unterschiedlich viele seltene Codons in verschiedenen Kombinationen besaßen, hergestellt. Es konnten jedoch keine einzelnen Codons, die für die Expressionssteigerung allein verantwortlich waren, identifiziert werden. Die Plasmidkopienzahl in den verschiedenen SULT2B1b-Klonen war konstant und die SULT2B1b-mRNA-Konzentration zeigte nur moderate Schwankungen, die nicht als Ursache für die dramatische Erhöhung der SULT2B1b-Expression in Frage kommen. Die berechnete Stabilität der potentiellen mRNA-Sekundärstrukturen wurde durch die seltenen Codons häufig stark gesenkt und ist als eine mögliche Ursache für die Expressionssteigerung anzusehen. Zusätzlich erhöhten die seltenen Codons den Consensus der Downstream Box zur 16S rRNA, was ebenfalls eine Ursache für die Expressionssteigerung sein kann. <br><br> In dieser Arbeit konnte somit die Expression der humanen SULT1A1*1, 1A2*1 und der 2B1b sowie der rGSTT2 erfolgreich mittels synonymer Codonaustausche erhöht werden. Die so optimierten S. typhimurium Stämme zeigten im Ames-Test eine erhöhte Sensitivität gegenüber SULT aktivierten Promutagenen bzw. erhöhte Aktivität in spezifischen Enymaktivitätsassays. / The enzyme super familiy of human sulfotransferases (SULT) plays an important role in phase II metabolism of xenobiotics. They catalyze the transfer of a sulfonyl moiety to nucleophilic groups of endogenous and exogenous substrates. Sulfoconjugation of xenobiotics facilitates their excretion by increasing the water solubility and inhibiting passive permeation of cell membranes. Depending on the molecular structure of the substance, sulfonation can also lead to metabolic activation. Highly reactive resonance-stabilized carbenium- and nitrenium-ions that are able to covalently bind to cellular nucleophiles, e.g. DNA, can be formed. Thus, SULT-mediated activation of promutagenic compounds is of toxicological interest. The detection of SULT-mediated mutagenicity in bacterial in-vitro testsystems (e.g. S. typhimurium) requires the heterologous expression of xenobiotic-metabolizing enzymes directly in these indicator cells. S. typhimurium do not express endogenous SULT and external metabolic systems are problematic as penetration of cell membranes is hampered for charged and short-lived metabolites. But the expression of human enzymes in bacteria can be problematic too. SULT1A2*1 and *2 for instance are allelic variants that differ only in two amino acids. However, using the same experimental conditions strong differences in their expression level have been observed. This complicates the comparison of the mutagenic activities of the polymorphic enzymes in the in-vitro test system. Other enzymes (e.g. SULT2B1b) show no detectable expression in bacteria when their genuine cDNA obtained from human tissues is used. Therefore, the aim of this study was to to optimize protein levels of heterologously expressed xenobiotic-metabolizing enzymes in indicator cells for mutagenicity testing. <br><br> So far it has been shown that synonymous codon-exchanges at the 5’end of human SULT1A2-cDNA led to an enhanced expression of the corresponding enzyme in S. typhimurium. Accordingly, codon-exchanges at the 5’end of SULT1A1*1, -1A2*1, -2B1b, rat glutathione-S-transferase theta 2 (rGSTT2) and the reportergene luciferase were conducted. The expression of the resulting constructs was quantified in S. typhimurium and E. coli using specific antibodies and activity of the overexpressed enzymes was proved by Ames test and enzyme activity assays. The introduction of low-usage codons at the 5’end of SULT1A1*1, -1A2*1 and -2B1b cDNA led to a 7-, 18- and 100-fold increase of expression level, respectively. Expression of rGSTT2 was 5-fold enhanced after the introduction of low-usage codons. In contrast, the introduction of low-usage codons into the luciferase cDNA resulted in a decrease of protein expression up to 80 %. Fusionproteins of SULT2B1b (5’end) and luciferase (3’end) showed a reduction of protein expression about 50 % after the introduction of low-usage codons. S. typhimurium strains with optimized SULT1A1*1 and -1A2*1 expression were used in the Ames test and showed a higher sensitivity compared to the lower expressing strains. For SULT2B1b no mutagen-activation could be detected in the Ames test, but enzyme activity was proved through Dehydroepiandosterone sulfation in vitro. <br><br> Since an inhibitory effect of low-usage codons on expression in bacteria was described in literature, the enhancement of expression after the introduction of low-usage codons observed in this study was analyzed more in detail. Various constructs of SULT1A2*1 and -2B1b cDNAs containing different numbers and combinations of synonymous low-usage codons were generated. No single codon that was responsible for the enhanced expression could be identified. Plasmid copy number of different SULT2B1b constructs was unchanged and SULT2B1b-mRNA showed only moderate variations that could not explain the strong enhancement of SULT2B1b expression. Calculations suggested that the stability of potential mRNA secondary structures was reduced due to the introduction of low-usage codons. Moreover, the consensus of the downstream box and the 16S rRNA was increased. Both effects probably improved the efficiency of translation and thereby increased the yield of protein expression. <br><br> In this study the heterologous expression of SULT1A1*1, -1A2*1, -2B1b and rGSTT2 could be enhanced by the introduction of synonymous low-usage codons. The optimized S. typhimurium strains showed higher activities in enzyme assays with specific substrates and an increased sensitivity towards SULT-activated promutagens.

I developed a precision-cut adrenal slice culture procedure to investigate cytochrome P450 (CYP) catalysed irreversible binding and adrenocorticolytic effects in human, rodent, and fish adrenal tissue, ex vivo. Autoradiography and radioluminography of exposed tissue slices showed that the potent adrenal toxicant 3-methylsulphonyl-2,2´-bis(4-chlorophenyl)-1,1´-dichloroethene (MeSO2-DDE) causes a selective metabolite binding in zona fasciculata (ZF), which is diminished by the CYP11B1 inhibitor metyrapone. MeSO2-DDE also reduces corticosterone secretion, increases 11-deoxycorticosterone secretion and causes mitochondrial degeneration in ZF cells in cultured mouse adrenal slices. ACTH treatment of mice induces CYP11B1 and increases irreversible MeSO2-DDE binding and toxicity in ZF cells. Metyrapone-sensitive binding of MeSO2-DDE is also observed in human zona fasciculata/reticularis (ZF/ZR) and 11-deoxycorti- sol/corticosterone secretion increases in MeSO2-DDE-exposed cultured human adrenal slices. The adrenocorticolytic drug 2-(2-chlorophenyl)-2-(4-chlorophenyl)-1,1-dichlorethane (o,p´- DDD, Mitotane®) is also bound in ZF/ZR but does not to impair hormone secretion in human adrenal slices at equimolar concentration. A targeted, presumably CYP1B1-catalysed irreversible binding of the adrenocorticolytic carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) in ZF/ZR occurs in rat adrenal slices, whereas presumably CYP1A1-catalysed irreversible binding in endothelial cells is observed in CYP1-induced rats and mice. The rat-specific adrenocorticolytic activity of DMBA may rely on two independent pathological processes resulting in cell death and haemorrhage in the adrenal cortex. In Atlantic cod, selective binding of o,p´-DDD is observed in interrenal cells in cultured anterior kidney slices. In conclusion, precision-cut adrenal slice culture is a simple ex vivo test system with which to investigate CYP-catalysed metabolite binding, alteredsteroid hormone secretion and target cell ultrastructure in human, experimental and wild animal tissue. The results imply that organisms under stress could be at increased risk of MeSO2-DDE induced adrenal toxicity. MeSO2-DDE is an expected human adrenal toxicant, which should be evaluated as a possible alternative in the therapy of adrenocortical hypersecretion and tumour growth.

<p>The nasal passages are constantly exposed to both air- and bloodborne foreign compounds. In particular, the olfactory mucosa is demonstrated to be susceptible to a variety of drugs and chemicals. In this thesis, mechanisms involved in tissue-selective toxicity in the olfactory mucosa of rodents have been investigated using the olfactory toxicant 2,6-dichlorophenyl methylsulphone (2,6-diClPh-MeSO₂) as a model compound. Comparative studies were performed with the non-toxic 2,5-dichlorophenyl methylsulphone (2,5-diClPh-MeSO₂) and the reasons for the strikingly different toxicity were investigated. </p><p>A strong bioactivation and protein adduction of 2,6-diClPh-MeSO₂ in olfactory microsomes and S9-fractions of rodents was demonstrated. In contrast, no significant metabolic activation of 2,5-diClPh-MeSO₂ was observed and the bioactivation in the liver for both chlorinated isomers was negligible. <i>In vitro</i> studies with recombinant yeast cell microsomes expressing mouse cytochrome P450 2A5 (CYP2A5) demonstrated a metabolic activation of 2,6-diClPh-MeSO₂. The 2,6-diClPh-MeSO₂-induced lesions and CYP2A5 expression preferentially occurred in Bowman’s glands and sustentacular cells of the olfactory mucosa. A significant depletion of glutathione (GSH) in the olfactory mucosa was demonstrated <i>in vivo</i>, while no changes were observed in the liver. There was a rapid induction of the endoplasmic reticulum (ER)-specific chaperone Grp78, activation of the ER-specific caspase-12 and the downstream caspase-3 in the Bowman’s glands. Electron microscopy revealed swelling of ER and mitochondria and a lost integrity of the Bowman’s glands. </p><p>Based on these results, the proposed mechanism for 2,6-diClPh-MeSO₂-induced toxicity in the olfactory mucosa is bioactivation by CYP2A5 into a reactive intermediate causing protein adduction and GSH-depletion. This is initiating a sequence of downstream events of ER-stress, changes in ion homeostasis, ultrastructural organelle disruption and apoptotic signalling. In spite of the initial apoptotic signals, the terminal phase of apoptosis seemed to be blocked and necrotic features occurred. The predominant expression of CYP2A5 in the olfactory mucosa is proposed to play a key role for the tissue- and cell-specific toxicity induced by 2,6-diClPh-MeSO₂.</p>

The nasal passages are constantly exposed to both air- and bloodborne foreign compounds. In particular, the olfactory mucosa is demonstrated to be susceptible to a variety of drugs and chemicals. In this thesis, mechanisms involved in tissue-selective toxicity in the olfactory mucosa of rodents have been investigated using the olfactory toxicant 2,6-dichlorophenyl methylsulphone (2,6-diClPh-MeSO2) as a model compound. Comparative studies were performed with the non-toxic 2,5-dichlorophenyl methylsulphone (2,5-diClPh-MeSO2) and the reasons for the strikingly different toxicity were investigated. A strong bioactivation and protein adduction of 2,6-diClPh-MeSO2 in olfactory microsomes and S9-fractions of rodents was demonstrated. In contrast, no significant metabolic activation of 2,5-diClPh-MeSO2 was observed and the bioactivation in the liver for both chlorinated isomers was negligible. In vitro studies with recombinant yeast cell microsomes expressing mouse cytochrome P450 2A5 (CYP2A5) demonstrated a metabolic activation of 2,6-diClPh-MeSO2. The 2,6-diClPh-MeSO2-induced lesions and CYP2A5 expression preferentially occurred in Bowman’s glands and sustentacular cells of the olfactory mucosa. A significant depletion of glutathione (GSH) in the olfactory mucosa was demonstrated in vivo, while no changes were observed in the liver. There was a rapid induction of the endoplasmic reticulum (ER)-specific chaperone Grp78, activation of the ER-specific caspase-12 and the downstream caspase-3 in the Bowman’s glands. Electron microscopy revealed swelling of ER and mitochondria and a lost integrity of the Bowman’s glands. Based on these results, the proposed mechanism for 2,6-diClPh-MeSO2-induced toxicity in the olfactory mucosa is bioactivation by CYP2A5 into a reactive intermediate causing protein adduction and GSH-depletion. This is initiating a sequence of downstream events of ER-stress, changes in ion homeostasis, ultrastructural organelle disruption and apoptotic signalling. In spite of the initial apoptotic signals, the terminal phase of apoptosis seemed to be blocked and necrotic features occurred. The predominant expression of CYP2A5 in the olfactory mucosa is proposed to play a key role for the tissue- and cell-specific toxicity induced by 2,6-diClPh-MeSO2.

The DDT metabolite 3-methylsulphonyl-DDE (3-MeSO2-DDE) induces cell death specifically in the adrenal cortex of mice after a cytochrome P45011B1 (CYP11B1)-catalysed bioactivation. This substance is not only an environmental pollutant, but also a suggested lead compound for an improved chemotherapy of adrenocortical carcinoma (ACC). The aim of the thesis was to further investigate this compound in terms of kinetics, cell death mechanisms and species differences. The pharmacokinetics of 3-MeSO2-DDE and the current drug for ACC, o,p’-DDD, was studied during 6 months following a single dose in minipigs. The elimination was slower for 3-MeSO2-DDE than for o,p’-DDD, indicated by a lower clearance and longer t½ in plasma and subcutaneous fat. Both substances remained in fat tissue during the whole study period. Unlike o,p’-DDD, 3-MeSO2-DDE was retained also in liver. The adequacy of the murine adrenocortical cell line Y-1 was evaluated for studies of adrenotoxic compounds. The Y-1 cells proved to be an appropriate test system for future mechanism studies, since CYP-catalysed irreversible binding, inhibited corticosterone production induced by 3-MeSO2-DDE and o,p’-DDD were successfully demonstrated. Cell death of 3-MeSO2-DDE in the mouse adrenal cortex was implied to be necrotic. Early apoptotic signalling (i.e. up-regulation of caspase-9) was observed, although it seemed to be interrupted by ATP-depletion and anti-apoptotic actions by heat shock protein 70, resulting in lack of activation of caspase-3. Using cultured adrenal tissue slices, two not previously studied species were examined ex vivo regarding adrenal binding of 3-MeSO2-[14C]DDE. Binding was found in the hamster adrenal cortex and in assumed cortical cells in the medulla, while the guinea pig adrenal was devoid of binding. This emphasises the species specificity in bioactivation of 3-MeSO2-DDE. The thesis forms a basis for further investigations in the human adrenocortical cell line H295R and provides new knowledge of importance for toxicological risk assessment of 3-MeSO2-DDE.