Dissertation (Ph.D.)--University of Michigan.

<p>Drug toxicity may cause liver injury, resulting in damage to cells and tissues. This damage can lead to cytotoxic events that may result in an activation of poly (ADP-ribose) polymerase (PARP). A study was conducted to determine if cocaine and acetaminophen toxicity lead to DNA damage and to the activation of the repair protein, PARP in the liver using the hepatotoxicants: cocaine and acetaminophen (APAP). A dose-response analysis for cocaine concluded that a dose as low as 20 mg/kg resulted in elevated ALT levels. A higher dose of 60 mg/kg was tested for analyses but resulted in severe hemorrhaging. The dose-response analyses for APAP resulted in no elevated liver enzyme levels for a 75 mg/kg and 150 mg/kg dose. A dose of 50 mg/kg for cocaine, and a dose of 300 mg/kg for APAP were used to analyze temporal trends for both toxicants. Both cocaine and APAP produced incremental increases in ALT at the 2 hour, 6 hour, 18 hour, and 24 hour time points, respectively. PARP activity analysis for cocaine measured the highest activity at the 2hr and 6hr time points. PARP analysis for acetaminophen measured gradual increases until the 18 hour time point where the highest level of PARP activity was measured. A PARP inhibition analysis was conducted with cocaine and (APAP) to understand the impact of a PARP inhibitor, 1,5-dihydroxyisoquinoline (DIQ), on PARP activity in the liver. A 50 mg/kg dose of cocaine or a 300 mg/kg dose of APAP was administered, followed by a 10 mg/kg dose of DIQ at 1) the time of initial toxicant dose (0 hour), or 2) 1 hour after initial toxicant dose (1hr). The PARP inhibition analysis for cocaine and APAP was conducted at 6 and 18 hours post initial dose, respectively, when the highest levels of PARP were observed. Inhibition analyses determined that ALT declined significantly when DIQ was administered immediately following the initial toxicant dose for both toxicants. DIQ administered 1 hr after initial toxicant dose resulted in slightly higher ALT than the 0 hr time point. Decreases in PARP activity were observed at the 0 hr time point, with slightly higher PARP levels observed at the 1 hr time point. Decreased PARP activity was observed following DIQ treatment with both, a concurrent drug treatment and treatment following drug administration. Cocaine and APAP treatment did not cause DNA fragmentation. A liver glutathione (GSH) analysis conducted for cocaine and APAP did not correlate with DIQ alteration of PARP activity. The mechanism of DIQ effects on drug-induced hepatotoxicity appears to be GSH independent. DIQ was effective in reducing drug-induced hepatotoxicity and preserving organ function.

Amphetamine-type stimulants are widely abused due to their ability to stimulate the central nervous system and elicit feelings of confidence, wakefulness, mood elevation, and euphoria. After cannabis, amphetamines were the most abused group of illicit substances in 2016 according to the National Forensic Laboratory Information System Annual Report. Included in this group are ephedrine, amphetamine, methamphetamine, 3,4-methylenedioxyamphetamine (MDA), and 3,4-metheylenedioxymethamphetamine (MDMA). Structurally, each of these compounds contain a chiral center, causing them to have an R-(-) and S-(+)- enantiomer (also called levo- and dextro-, respectively). Despite their similarity, the R- and S- enantiomers display differing pharmacological effects, with the S-enantiomer producing a stronger, longer-lasting effect than the R-enantiomer. Because of this, R-methamphetamine, for example, has therapeutic uses and is the active ingredient in some over-the-counter nasal decongestant products (e.g. Vicks® vapor inhaler). S-methamphetamine, on the other hand, is generally found in illicit sources. As a result of these chiral centers, these compounds have differing legal statuses. The aim of this research was to develop and validate a method for the separation and analysis of eleven amphetamine-class compounds in blood for forensic casework. This was accomplished using a liquid-liquid extraction and analysis on a liquid chromatography-tandem mass spectrometer (LC-MS/MS) (Agilent Technologies, Santa Clara, CA, USA). Chromatographic separation was achieved using a Phenomenex Lux® AMP chiral column (Phenomenex, Torrance, CA, USA), under gradient aqueous and organic mobile phase conditions, with a total run time of just over 17 minutes. The target analytes included 1S,2R-ephedrine, 1R,2S-ephedrine, R-amphetamine, S-amphetamine, R-methamphetamine, S-methamphetamine, phentermine, R-MDMA, S-MDMA, R-MDA, and S-MDA with 1S,2R-ephedrine-d3 and MDMA-d5 as the internal standards (Cerilliant, Round Rock, TX, USA). The analytical method was validated according to the Scientific Working Group for Forensic Toxicology guidelines (now a subcommittee of the Organization of Scientific Area Committees for Forensic Science), including the assessment of its linearity, limits of detection and quantitation, bias, precision, interferences, matrix effects, carryover, and processed sample stability [2]. The limit of detection (LOD) was 2 µg/L for all compounds except MDMA and MDA, which had LODs of 10 µg/L. The lower limit of quantitation (LLOQ) and upper limit of quantitation (ULOQ) was 10 µg/L and 1000 µg/L for all compounds, respectively. The precision was within 15% for all analytes, with the bias extending outside the ±20% range for at least one set of samples for all analytes except 1S,2R-ephedrine and both MDA enantiomers. Matrix effect studies showed average ion enhancement (140%-361%), extraction efficiencies (60%-123%), and process efficiencies (105%-432%) across all analytes. No interferences were detected from isotope internal standards, postmortem blood, antemortem blood, or 85 commonly seen drugs in forensic casework. No carryover was observed following injections of analytes at the ULOQ (1000 µg/L). To demonstrate applicability in authentic casework, the method was applied to 28 cases that had previously been analyzed using a non-chiral method. By selectively identifying R- and S- enantiomers, this method may be used in forensics laboratories where the question of the licit or illicit use of amphetamines is of importance.

This dissertation describes an innovative methodological approach to assess hormetic (biphasic) dose-responses in a single high-throughput study. This approach was applied to data from a National Cancer Institute database that evaluated the capacity of 2,189 candidate antineoplastic drugs to inhibit the growth of 13 yeast strains with and without genetic mutations typical of cancer cells. Evidence of hormesis in these data was assessed in several unique evaluations. Described first, is a method to quantitatively evaluate the concentration-response curves on features pertaining to the magnitude of stimulation, the measure of variation between replications, and the consistency of response across concentrations in the below threshold zone using a novel "Hormetic Metric“. These results are compared to an evaluation of the database using a previously described Hormesis Scoring System. Described next is a quantitative evaluation of the features of the hormetic concentration-responses where the width of the hormetic zone, the width from the maximum stimulation to the toxic threshold (i.e., the zero equivalence point), and the maximum stimulation of the hormetic concentration-responses are all assessed. In a summary evaluation, the evidence of hormesis that was quantified using the previous methodologies is discussed relative to (a) the number of concentrations below the toxic threshold; (b) the chemical compounds toxicity; and (c) the differential response of the various genotypes.

The epidermal growth factor receptor (EGFR) is a tyrosine kinase cell surface receptor, belonging to the ErbB family of receptors, which functions to initiate downstream signaling pathways resulting in cellular proliferation, differentiation, migration, adhesion, and metastasis. EGFR is over expressed in numerous types of cancers including head and neck, breast, colon, lung, renal, and ovarian, and this overexpression of EGFR has been correlated with poor prognosis and survival rates specifically in head and neck squamous cell carcinoma (HNSCC). Due to the pro-oncogenic activity of EGFR (e.g. cell growth, angiogenesis, and metastasis), ameliorating the function of EGFR makes it an attractive target for chemotherapy. While clinical therapy with Erlotinib is quite successful, resistance to the chemotherapy agent is highly prevalent. To date, the molecular mechanism by which cancers become resistant to Erlotinib remains unclear. Here we show a novel pathway in head and neck carcinoma cell lines involving the up-regulation of NADPH oxidase enzymes (NOX) after Erlotinib treatment, which leads to an increased production of hydrogen peroxide and finally increased IL-6 production. The IL-6 axis has been proposed to play a definitive role in the long-term proliferation and survival of various cancers, and perturbation of this axis by Erlotinib may initiate pro-survival signals which render the cancers resistant to Erlotinib. Therefore, we put forth a model of multidrug therapy targeting NOX enzymes and/or IL-6 in combination with Erlotinib to counteract the potential drug-resistance mechanisms often observed in malignancy.

Pulmonary fibrosis is an aberrant transformation of injured lung tissue. It is characterized by irreversible accumulation of extracellular matrix produced by fibroblasts and myofibroblasts during tissue remodeling, resulting in destruction and dysfunction of the lung. Asbestos is an important cause of pulmonary fibrosis. In response to asbestos exposure, alveolar macrophages and recruited monocytes generate reactive oxygen species (ROS), especially hydrogen peroxide (H2O2), pro-inflammatory cytokines, such as TNF-á and IL-1â, and induce subsequent collagen deposition in the lung. We have found that increased H2O2 levels are linked to the development of pulmonary fibrosis. Catalase converts H2O2 into water and oxygen, so we hypothesized that catalase may attenuate the development of pulmonary fibrosis. Interestingly, previous studies from our lab have demonstrated that decreased H2O2 levels are associated not only with a decrease in fibrosis but also an increase in pro-inflammatory cytokines. In these current studies, we demonstrate that in the presence of asbestos, catalase increases TNF-á and IL-1â in macrophages while decreasing collagen production in fibroblasts. This is reversed when TNF-á receptor-1 is knocked down, suggesting that TNF-á may play a role in fibrosis development. To investigate these finding in vivo, catalase-treated mice showed decreased fibrosis histologically, decreased in collagen levels in BAL, and decreased hydroxyproline in lung tissue. The major finding of my study is that catalase attenuates asbestos-induced fibrosis while increasing pro-inflammatory cytokines. This is contrary to the typical thought that pro-inflammatory states are associated with the fibrotic phenotype. The studies in this thesis may uncover a therapeutic target to attenuate the progression and/or development of pulmonary fibrosis.

Polyhalogenated aromatic hydrocarbon (PHAH) belongs to a group of ubiquitous environmental contaminants. Among them, 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD) is one of the most toxic synthetic chemicals known. It accumulates in adipose tissues of animals and has a biological half-life estimated at 6-12 years in human. As an endocrine disruptor, TCDD affects plasma levels of pituitary hormones including gonadotropins (FSH/LH), PRL (PRL), adrenocorticotropic hormone (ACTH, corticotropin), and thyroid-stimulating hormone (TSH). Most, if not all, of these effects may be related to the ability of TCDD to exert antiestrogenic effects. Therefore the goal of this work was to first use microarrays to identify gene targets of TCDD in the pituitary gland in vivo and then use a pituitary-derived cell line to determine the mechanisms underlying TCDD effects on these genes. The work described in my dissertation is the first to demonstrate that TCDD acts directly on rat pituitary lactotropes and gonadotropes to interfere with E2 regulation of genes encoding PRL and glycoprotein hormones. It is also the first to show that TCDD exerts antiestrogenic activity on the regulation of the PRL gene both in vivo and in vitro, and that this interference occurs at the transcriptional level. My findings suggest that neither activation of a putative inhibitory dioxin response element, metabolism of E2 by CYP 1A1 and CYP 1B1 enzymes nor the down-regulation of the AhR is likely responsible for the antiestrogenicity of TCDD. Instead, I found that limitation of ERα, but not ERβ availability, after TCDD co-treatment with E2 is responsible for the underlying antiestrogenic effects of TCDD on PRL gene expression. My findings suggest that it is likely that several pathways are involved in the ability of TCDD to limit ERα availability for PRL gene transcription. They are decreased ERα levels through down-regulation of ERα gene transcription; degradation of ERα through the ubiquitin–proteasome pathway, and direct interference with E2-induced ERα binding to an ERE enhancer in the PRL gene promoter. Interestingly, I found that at the same time TCDD interfered with ERα binding and PRL promoter activation, liganded ER enhanced gene targets of the AhR, CYP1A1 and CYP1B1. Based on these findings, I propose a novel model of cross-talk between the AhR and ER that may explain the antiestrogenic activity of TCDD in the regulation of PRL gene transcription. In this model, ligand-activated AhR/ARNT complex may sequester liganded ERα and recruit it to DREs in AhR target genes, thereby limiting availability of liganded ERα for recruitment to ERE sequences in the PRL gene. In addition to providing a novel model to explain antiestrogenic effects of TCDD, my findings that TCDD interferes with PRL and glycoprotein hormone synthesis may explain the broad range of reproductive dysfunctions caused by TCDD exposure during development and adulthood.^

As a primary colonizer of carrion, Phormia regina plays a vital role in both nutrient recycling and ecosystems via carrion decomposition. They are currently used by forensic scientists in establishing minimum post mortem interval estimation during criminal cases but recent research has that larvae can be further used for detection of drugs and toxins when traditional matrices are unavailable. In addition, the presence of toxins has been shown to increase or delay development of entomological specimens, which can be critical in a criminal investigation. It has become critical to understand the detection and effects of fentanyl on different matrices due to its increased use and high contribution to overdosing. The purpose of this study was to further analyze the question currently poised for forensic entomologists/toxicologists: if there is any correlation between concentrations of toxins in human postmortem tissue to those concentrations detected in blowfly larvae and can they be used reliably as samples for toxicological investigations. Using Phormia regina, this study evaluated LC-MSMS to quantify and qualify fentanyl accurately as well as obtain some preliminary information pertaining to the effects of fentanyl on larvae development. A maximum of 100 individual pupae received from the University of New Haven (West Haven, Connecticut, USA) and University College of St. Mary (Omaha, Nebraska, USA) were placed inside a mesh enclosure with an ambient temperature of 26-28 °C. Chunks of beef liver were provided to the adults to facilitate oviposition. Eggs were collected and placed on homogenized ground liver pre-dosed with fentanyl. The doses for trial 1 were low end therapeutic levels (1.5 ng/mL and 3.0 ng/mL). The doses for trial 2 were a high end therapeutic level (50 ng/mL), a toxic level (125 ng/mL), a lethal level (250 ng/mL) and beyond lethal level (500 ng/mL). The larvae were monitored, and sample larvae were taken every day to confirm the lifecycle stage. Triplicate larvae samples were collected at the 2nd instar, 3rd instar, post-feeding, pupae and adult life stages. The larvae were homogenized, extracted and analyzed using liquid chromatograph tandem mass spectrometry (Shimadzu, Kyoto; Japan). Fifteen random specimens were selected and used as a representative set every day for each dose to monitor life stage duration. These larvae were also measured to obtain a growth curve. During trial 1 where larvae were fed liver substrate with low-level amounts of fentanyl, the detection of the drug during 2nd/3rd instar, pupae and adult was either minimal or nonexistent. Most of the results did not meet the criteria of acceptability for LOD and LOQ (0.5 ng/mL) for the method used for the LC-MSMS except for one pupae from the 3.0 ng/mL dosed liver. False positives were seen in trial 1 of the adult control sample set. As preliminary results remain inconclusive, recovery involving chitinized empty puparial cases should be re-analyzed utilizing a separate validated method for the empty puparial casings for optimum recovery. During trial 1 the larvae went through life stages but only the control group successfully emerged into adults. The development and survival rates of the larvae were noted but ultimately deemed inconclusive due to inconsistent temperatures forcing the larvae into diapause. During the trial 2 fentanyl was detected in 2nd and 3rd instar larvae fed on toxic and lethal level dosed meat while detection remained minimal or nonexistent in the post-feeding larvae of any dose. There was no correlation found between the feeding substrate and the larvae samples of any stage of the life cycle. The development rate to 2nd instar was delayed and correlated with the increase of fentanyl thus prolonging the total time for development from 2 to 4 days. This study demonstrated that fentanyl was detectable in 2nd and 3rd instar larvae in toxic and lethal doses. Further, it demonstrated that fentanyl delayed development in Phormia regina by up to 4 days when exposed to high levels of lethal dosing, effecting accurate post mortem interval estimation. It was also concluded that there is no correlation between spiked food substrates and the concentration found in the analyzed insect. Factors such as the considerable variation in toxins found in a corpse, decomposition chemicals, and different carrion insects during different times of the year, as well as more research are needed in the field before correlation can be confirmed. As for development and survival rates, this remains to be the most important to note when studying toxins in insects.

Metabolic homeostasis is controlled, in part, by a family of proteins called nuclear receptors through which lipophilic hormones and hormone-like molecules regulate gene expression. One such nuclear receptor is peroxisome proliferator activated receptor γ (PPARγ). Its activation is essential for white, brite (brown-in-white) and brown adipogenesis, adipocyte function, mature adipocyte maintenance, and insulin sensitivity. PPARγ activation regulates energy homeostasis by both promoting storage of excess energy as lipids in white adipocytes and stimulating energy dissipation in brite and brown adipocytes. Accumulation of white adipocytes significantly increases the risk of obesity and metabolic syndrome. On the other hand, brown and brite adipocytes potentially counteract metabolic disease-related symptoms. The adipocyte differentiation and function as well as insulin sensitizing activities of PPARγ are regulated separately through differential post-translational modifications and/or co-regulator recruitment, with ligands having distinct abilities to activate each of PPARγ’s functions. These provide mechanisms by which a ligand could induce adipogenesis without stimulating PPARγ’s health promoting functions (i.e. insulin sensitivity, energy dissipation). The central hypothesis of this dissertation is that compared to therapeutic PPARγ ligands (i.e. rosiglitazone), environmental PPARγ ligands will activate a distinct PPARγ transcriptional program that disrupts adipose and metabolic homeostasis. Two study aims were developed to test and refine this central hypothesis. The first aim identified genes and pathways that differentiate environmental PPARγ ligands from endogenous and therapeutic chemicals. In primary mouse bone marrow multipotent stromal cells and 3T3-L1 cells, the environmental PPARγ ligands tributyltin (TBT, an antifouling agent and plasticizer) and triphenyl phosphate (TPhP, an organophosphate flame retardant) induced transcriptomic profiles that were distinct from rosiglitazone. All ligands induced adipogenesis; yet, only rosiglitazone strongly enriched pathways related to brown fat differentiation and mitochondrial processes and induced brite adipocyte gene markers (Cidea, Elovl3, Ucp1). Using the transcriptional profiles from 3T3-L1 adipocytes differentiated in the presence of 76 different chemicals, a taxonomy was built to identify environmental chemicals as PPARγ-modifying chemicals distinct from known PPARγ-modifying therapeutics. The second aim investigated the role of phosphorylation of PPARγ in defining environmental ligand-induced changes in adipocyte differentiation and function. In differentiated 3T3-L1 cells, rosiglitazone and TPhP both induced adipogenesis through PPARγ, but only rosiglitazone enhanced mitochondrial biogenesis and mitochondrial respiration, which contribute to healthy energy expenditure. Rosiglitazone, but not TPhP, protected PPARγ from phosphorylation at Ser-273. However, in 3T3-L1 cells in which PPARγ cannot be phosphorylated, TPhP was able to induce mRNA expression of a suite of brite adipocyte genes. In male C57BL/6J mice fed either a low or high fat diet, TPhP caused a significant decrease in brite adipocyte gene expression (Elovl3, Ucp1) in mature adipocytes from inguinal adipose tissue. Together, these studies support our hypothesis that environmental PPARγ ligands (i.e. TBT and TPhP) skew adipocyte differentiation toward white adipogenesis at the expense of brite adipogenesis, potentially because of differential post-translational modification of PPARγ.

Effects of alpha-chaconine were examined in the central nervous system on the basis of electrophysiological tests and by assaying for changes in the levels of several known neurotransmitters of the brain. The possibility that brain acetylcholine levels might increase corresponding to acetylcholinesterase inhibition in treated rats was investigated where an increase would suggest that the anti-acetylcholinesterase activity of alpha-chaconine was physiologically significant. Various doses of alpha-chaconine were given to rats prepared for physiological tests. Some symptoms were observed at relatively low doses (8 to 10 milligrams per kilogram). These included constriction of abdominal muscles~ respiratory impairment~ and sedation. At the same dosage the electroencephalogram pattern was altered by a notable increase in the low-frequency component. Tachycardia was observed at both low (10 milligrams per kilogram) and at high doses (40 milligrams per kilogram)~ whereas intermediate doses (20 to 30 milligrams per kilogram) were associated with bradycardia. Although other physiological tests implicated the brain as the target organ, analyses of brain neurotransmitter chemicals failed to show significant trends. Compounds assayed were acetylcholine, norepinephrine, dopamine, serotonin, and the serotonin metabolite, 5-hydroxyindoleacetic acid. The observation that acetylcholine levels were unchanged after alpha-chaconine administration did not support the hypothesis that the major toxic effect of the test compound was caused by alterations in the levels of this neurotransmitter.