abstract: The need for rapid, specific and sensitive assays that provide a detection of bacterial indicators are important for monitoring water quality. Rapid detection using biosensor is a novel approach for microbiological testing applications. Besides, validation of rapid methods is an obstacle in adoption of such new bio-sensing technologies. In this study, the strategy developed is based on using the compound 4-methylumbelliferyl glucuronide (MUG), which is hydrolyzed rapidly by the action of E. coli β-D-glucuronidase (GUD) enzyme to yield a fluorogenic product that can be quantified and directly related to the number of E. coli cells present in water samples. The detection time required for the biosensor response ranged from 30 to 120 minutes, depending on the number of bacteria. The specificity of the MUG based biosensor platform assay for the detection of E. coli was examined by pure cultures of non-target bacterial genera and also non-target substrates. GUD activity was found to be specific for E. coli and no such enzymatic activity was detected in other species. Moreover, the sensitivity of rapid enzymatic assays was investigated and repeatedly determined to be less than 10 E. coli cells per reaction vial concentrated from 100 mL of water samples. The applicability of the method was tested by performing fluorescence assays under pure and mixed bacterial flora in environmental samples. In addition, the procedural QA/QC for routine monitoring of drinking water samples have been validated by comparing the performance of the biosensor platform for the detection of E. coli and culture-based standard techniques such as Membrane Filtration (MF). The results of this study indicated that the fluorescence signals generated in samples using specific substrate molecules can be utilized to develop a bio-sensing platform for the detection of E. coli in drinking water. The procedural QA/QC of the biosensor will provide both industry and regulatory authorities a useful tool for near real-time monitoring of E. coli in drinking water samples. Furthermore, this system can be applied independently or in conjunction with other methods as a part of an array of biochemical assays in order to reliably detect E. coli in water. / Dissertation/Thesis / Doctoral Dissertation Civil and Environmental Engineering 2015

Abstract Reactive metabolites are believed to be responsible for rare but serious idiosyncratic adverse drug reactions (IADRs) that have led to the withdrawal of numerous drugs from the market. This has resulted in major harm to patients, economic losses for the pharmaceutical companies and represents a serious problem in drug development. Reactive metabolites can be studied by trapping them with suitable nucleophiles, most commonly with glutathione. The glutathione conjugates formed in these reactions can be analyzed with liquid chromatography mass spectrometry (LC/MS) techniques. In this study, new in vitro methods for the detection and analysis of reactive metabolites were developed. The suitability for reactive metabolite screening of different enzyme sources commonly used in vitro were compared. It was found that sub-cellular fractions yielded significantly larger amounts of glutathione-trapped reactive metabolites as compared to the amounts obtained from intact hepatocytes. Additionally, different metabolites were detected in some cases. Biomimetic metalloporphyrin catalysts were tested for their ability to produce larger amounts of glutathione-trapped metabolites relative to liver S9 fraction incubations. An increase in reactive metabolite production was observed with biomimetic models, but not all of the metabolites produced by liver S9 were observed. The glutathione conjugates of pulegone and of its metabolite menthofuran were analyzed with LC/MS/MS, and the fragmentation spectra of N- and S-/N- di-linked glutathione conjugate were interpreted in detail for the first time. These results will enable more efficient screening of reactive metabolites of furan-containing compounds. Acyl glucuronides are metabolites produced from carboxylic acid-containing compounds and can be reactive. A good correlation was found between the acyl migration half-life and the tendency of a drug to cause IADRs. The carboxylic moiety can also be metabolized to yield acyl coenzyme A (CoA) conjugates that may be more reactive than their corresponding acyl glucuronides. The formation of CoA conjugates and additional conjugates formed from them was found to be more likely with drugs that cause IADRs. / Tiivistelmä Reaktiivisten metaboliittien uskotaan olevan syypää tietyntyyppisiin harvinaisiin, mutta vakaviin idiosynkraattisiin lääkehaittavaikutuksiin, jotka ovat johtaneet useiden lääkeaineiden poistamiseen markkinoilta. Ne ovat aiheuttaneet merkittäviä haittoja potilaille, tappioita lääkeyhtiöille ja ovat vakava ongelma lääkekehityksessä. Reaktiivisia metaboliitteja voidaan tutkia vangitsemalla niitä sopivilla nukleofiileillä, yleisimmin glutationilla. Muodostuneet glutationikonjugaatit voidaan sitten analysoida nestekromatografia / massaspektrometrisin tekniikoin. Tässä tutkimuksessa kehitettiin uusia in vitro tapoja havaita ja analysoida reaktiivisia metaboliitteja. Tavallisimmin käytettyjen entsyymilähteiden soveltuvuutta testattiin reaktiivisten metaboliittien seulontaan. Solufraktioiden havaittiin tuottavan huomattavasti suurempia määriä glutationi-vangittuja reaktiivisia metaboliitteja kuin elävät solut. Lisäksi eri metaboliitteja havaittiin joillekin aineille eri entsyymilähteissä. Biomimeettisen metalliporfyriinikatalyytin kykyä tuottaa suurempia määriä glutationilla vangittuja reaktiivisia metaboliitteja testattiin vertaamalla sitä maksan S9 fraktioon. Vaikka katalyytillä pystyi tuottamaan suurempia määriä reaktiivisia metaboliitteja, kaikkia S9 fraktiossa havaittuja metaboliitteja se ei tuottanut. Pulegonin ja menthofuraanin glutationikonjugaatteja analysoitiin LC/MS/MS tekniikalla ja N- sekä S-/N- sitoutuneiden glutationikonjugaattien pilkkoutumisspektrit tulkittiin tarkasti ensimmäistä kertaa. Tulokset mahdollistavat furaanirenkaan sisältävistä yhdisteistä syntyvien reaktiivisten metaboliittien tehokkaamman seulonnan. Asyyliglukuronit ovat karboksyylihapporyhmän sisältämien yhdisteiden metaboliitteja, jotka voivat olla reaktiivisia. Asyyliglukuronien vaeltamisen puoliintumisajan ja idiosynkraattisten lääkehaittavaikutusten välillä havaittiin selvä korrelaatio. Karboksyylihapporyhmän kanssa voi muodostua myös asyyli koentsyymi A konjugaatteja, jotka voivat olla reaktiivisempia kuin vastaavat asyyliglukuronit. Koentsyymi A ja siitä edelleen syntyviä muita konjugaatteja havaittiin pääasiassa lääkeaineille, joiden todennäköisyys aiheuttaa idiosynkraattisia lääkehaittavaikutuksia oli suurempi.

Ethyl glucuronide (EtG) is a minor ethanol metabolite that confirms the absorption and metabolism of ethanol after oral or dermal exposure. Human data suggest that maximum blood EtG (BEtG) concentrations are reached between 3.5 and 5.5. h after ethanol administration. This study was undertaken to determine if the Sprague-Dawley (SD) rat biotransforms ethanol to EtG after a single high oral dose of ethanol. SD rats (male, n=6) were gavaged with a single ethanol dose (4g/kg), and urine was collected for 3. h in metabolic cages, followed by euthanization and collection of heart blood. Blood and urine were analyzed for ethanol and EtG by gas chromatography and enzyme immunoassay. Blood and urine ethanol concentrations were 195 ± 23 and 218 ± 19. mg/dL, whereas BEtG and urine EtG (UEtG) concentrations were 1,363 ± 98. ng equivalents/mL and 210 ± 0.29. mg equivalents/dL (X̄±standarderrorofthemean[S.E.M.]).Sixty-six male SD rats were gavaged ethanol (4. g/kg) and placed in metabolic cages to determine the extent and duration of ethanol to EtG biotransformation and urinary excretion. Blood and urine were collected up to 24. h after administration for ethanol and EtG analysis. Maximum blood ethanol, urine ethanol, and UEtG were reached within 4. h, whereas maximum BEtG was reached 6. h after administration. Maximum concentrations were blood ethanol, 213 ± 20. mg/dL; urine ethanol, 308 ± 34. mg/dL; BEtG, 2,683 ± 145. ng equivalents/mL; UEtG, 1.2 ± 0.06. mg equivalents/mL (X̄±S.E.M.). Areas under the concentration-time curve were blood ethanol, 1,578. h*mg/dL; urine ethanol, 3,096. h*mg/dL; BEtG, 18,284. h*ng equivalents/mL; and UEtG, 850. h*mg equivalents/dL. Blood ethanol and BEtG levels were reduced to below limits of detection (LODs) within 12 and 18. h after ethanol administration. Urine ethanols were below LOD at 18. h, but UEtG was still detectable at 24. h after administration. Our data prove that the SD rat biotransforms ethanol to EtG and excretes both in the urine and suggest that it is similar to that of the human.

Alcohol is the most abused drug in the United States and most frequently performed assay in forensic laboratories. Alcohol is routinely present in biological specimens from fatal residential fires and forensic toxicologists must interpret if these individuals are impaired by determination of their blood alcohol concentrations on post-incineration blood collected at autopsy. There is no known data available to confirm or refute blood alcohol concentrations and impairment in fire-related deaths. Ethyl glucuronide (EtG), a non-volatile minor ethanol metabolite, may provide a better biomarker for ethanol consumption prior to burn injury. The literature does not address the possibility that ethanol or EtG concentrations are altered in fire deaths. A Sprague Dawley rat model was employed to determine if ethanol and EtG concentrations in blood, liver, heart, and kidney were altered after burn injuries using two incineration models with varying durations and temperatures. Blood and tissues were analyzed for ethanol by gas chromatography and EtG by enzyme immunoassay. Other measurements including organ weights, lower hindquarter weights, and blood glucose concentrations were chosen for analysis to determine the mechanism by which the blood and organ ethanol and EtG concentrations are altered in burnt corpses. The rodent provided an excellent model for studying the biotransformation of ethanol to EtG and the effects of burn injury on ethanol and EtG concentrations. Our study revealed that blood ethanol concentrations were not significantly altered by burn injury but tissue ethanol concentrations were altered by burn injury. EtG concentrations were found to be altered in blood and tissue specimens in both incineration models. Our data suggest that the change in ethanol and EtG concentrations may be correlated to higher core body temperatures from burn injury and not changes in organ weight. Determining if blood ethanol concentrations are altered in burnt corpses is important for forensic toxicologists to conclude if victims were impaired at the time of death. The knowledge gained from these experiments will help forensic toxicologists by confirming the current interpretation that blood ethanol concentrations are not altered in fire deaths and provide a better understanding for the interpretation of impairment in burn deaths.

Resveratrol (RES) has been associated with numerous pharmacological effects. Yet its pharmacokinetics is not clearly understood. It is known to get extensively metabolized into its sulfated and glucuronidated metabolites and has very low circulating RES concentrations in plasma. Although the concentrations of conjugated metabolites of RES have been reported to be much greater than that of RES, they have not been evaluated. This also becomes important in light of positive biological activities reported for sulfated metabolites of RES. Conjugation is a complex process which can sometimes be a reversible process and needs comprehensive evaluation to better understand RES and its metabolites' disposition. There has been a debate among the researchers regarding the differences in kinetics of preformed versus in vivo formed metabolites in the light of guidelines issued by regulatory bodies regarding metabolites in safety testing (MIST). We have addressed the above questions in this work, in addition to evaluating brain permeability of a potent RES analog, trimethoxy-trans-stilbene (TMS). Chapter 1 presents a detailed introduction, hypothesis and significance of my work. Chapter 2 includes the development and validation of a bioanalytical method for quantitation of RES and its metabolites on LC/MS/MS. We were able to develop and validate a robust bioanalytical method to quantitate RES and its four major metabolites resveratrol-4'-glucuronide (R4'G), resveratrol-3-glucuronide (R3G), resveratrol-4'-sulfate (R4'S) and resveratrol-3-sulfate (R3S). In Chapter 3, lung as a possible metabolizing organ for RES was evaluated. This study was performed in vivo in mouse model using multiple site of administration and single site of sampling approach. In vitro studies were also performed to confirm the in vivo results. Inter species differences in RES pulmonary metabolism were also studied. We observed lungs to be the major metabolizing organs for RES with inter species differences in its metabolism. Chapter 4 provides detailed pharmacokinetics of sulfated metabolites of RES, i.e. resveratrol-3-sulfate (R3S) and resveratrol-4'-sulfate (R4'S) in mouse model by both systemic and oral routes. In vitro studies were also conducted to test the desulfation in liver. Although we did not observe any significant RES in plasma, we observed from our in vitro studies that sulfated metabolites were desulfated in liver. Chapter 5 explains the detailed pharmacokinetics of glucuronidated metabolites of RES i.e. resveratrol-3-glucuornide (R3G) after both systemic and oral route. R3G was observed to undergo enterohepatic circulation. Explanation of R3G disposition in hepatocytes and enterocytes were proposed based on our own and reported results. In Chapter 6 we compared the differences in the kinetics of preformed versus in vivo formed metabolites using modeling and simulation approach. Individual models for disposition of RES, R3S and R3G were developed. These models were combined to give a global model for RES metabolism into R3S and R3G. Simulations were performed under two assumptions; preformed versus in vivo formed metabolite kinetics a) are same and b) they are not same. Our results reported that preformed and in vivo formed metabolite kinetics are not same at least for hydrophilic phase II metabolites. Chapter 7 includes method development and validation for quantitation of TMS in plasma and brain of mouse. Chapter 8 includes a steady state study to characterize the pharmacokinetic parameters of TMS, which was used to evaluate brain permeability of TMS. In summary we developed a robust bioanalytical method for direct quantitation of RES and its metabolites, found the lung to be a major metabolizing organ for RES, delineated complex kinetics of conjugated metabolites of RES, and showed differences in preformed versus in vivo formed metabolite kinetics and better brain permeability of TMS. / Pharmaceutical Sciences

This thesis describes the in vitro production of drug metabolites using fungi of the Cunninghamella species. The metabolites were characterized with mainly liquid chromatography-mass spectrometry using ion-trap and quadrupole-time-of-flight instruments. A fungal in vitro model has several advantages e.g., it is easily up-scaled and ethical problems associated with animal-based models are avoided. The metabolism of bupivacaine and the selective androgen receptor modulators (SARMs) S1, S4 and S24 by the fungi Cunninghamella elegans and Cunninghamella blakesleeana was investigated. The detected metabolites were compared with those formed in vitro and in vivo by human and horse and most phase I metabolites formed by mammals were also formed by the fungi. The higher levels of bupivacaine metabolites in the fungal samples allowed an extensive mass spectrometric structural characterization which shows that the fungi are relevant metabolic models. Glucuronides are important drug metabolites but they are difficult to synthesize. The discovery that the fungus Cunninghamella elegans formed large amounts of glucosides led to the idea that they could be used to form glucuronides. A new concept was developed where a fungal incubate containing a SARM S1 glucoside was mixed with the free radical tetramethylpiperidinyl-1-oxy (TEMPO), sodium bromide and sodium hypochlorite which produced a glucuronide. Isolation and characterization by nuclear magnetic resonance spectroscopy proved that the new method could produce glucuronides for use as reference material. An investigation of reactive metabolite formation of the drugs paracetamol, mefenamic acid and diclofenac by the fungus Cunninghamella elegans was performed. It was demonstrated for the first time that the fungus could produce glutathione, glutathione ethyl-ester, cysteine and N-acetylcysteine conjugates that are indicative of a preceding formation of reactive intermediates. A comparison with conjugates formed by human liver microsomes showed that both models formed identical metabolites. The presented investigations prove that Cunninghamella fungi are relevant drug metabolism models. They show that the fungi to a large extent forms the same metabolites as mammals and that they can produce metabolites for use as reference material in, e.g. doping control. It was also demonstrated that the fungal model can be used in the important assessment of drug toxicity.

The hip-flask defence is a common claim in drunkdrinking cases. In Sweden and Norway two different models are used to determinethese cases. In Sweden one blood and two urine samples taken 60 minutes apartare used for analysis. In Norway two blood samples taken 30 minutes apart areused. Sweden focuses on the rise or fall of alcohol concentration in urine(UAC), and the ratio between UAC and blood alcohol concentrations (BAC). Norwayfocuses on the rise or fall of the alcohol metabolite ethylglucuronide (EtG) and the ratio between BAC and EtG. The aim of this study wasto test the models for multiple intakes and with different alcoholic beverages.Thirtyfive participants ingested two doses, first0.51 g/kg of beer and later either 0.25, 0.51 or 0.85 g/kg of beer, wine orvodka. Blood and urine samples were obtained before and after alcoholingestion. Alcohol was measured by GC-HS, and the alcohol metabolite byUPLC-MS/MS. The results showed that there are kineticdifferences between single and repeated intakes, that there are no significantdifferences in kinetics from different alcoholic beverages and thatthe Norwegian model appears to be the stronger one in hip-flask determination.

Modifying substrate specificity of β-glucuronidase (GUS) would be helpful in various enzyme prodrug systems in delivering drug dose to the site of action in the cancer treatment. Due to the presence of endogenous enzyme in human tissues, GUS-based Antibody-Directed Enzyme Prodrug Therapy (ADEPT) requires a novel substrate to avoid undesirable systemic activation. GUS is a glycosyl hydrolase, highly specific towards the glucuronide derivatives. It catalyzes the glycosidic cleavage of β-D-glucuronides to β-D-glucuronic acid and aglycone moiety. In order to gain insight on the substrate specificity of GUS, C6 carboxyl group of glucuronic acid was modified to C6 carboxamide (amide derivative). We have examined amide derivatized substrates with a variety of different aglycone groups including p-nitrophenyl, phenyl and 4-methylumbelliferone to further probe the activity profile of GUS. In an effort to optimize GUS activity, docking studies have been performed which indicated that amino acid point mutations near C6 carboxyl group of glucuronic acid could improve binding of the derivatized substrates. As a result point mutations to Arg-562 and Lys-568 which make the active site less positively charged either by glutamine or glutamate lead to an enzyme with much lower native substrate activity but abolished activity for the amide-derivatized substrate. This research study showed that there is still a further need of finding appropriate mutations required to make glucuronamide a better substrate for the mutated version of GUS.

17β-nandrolone (17βN) is illegally used in sport and in livestock as a growth promoter.17βN and some of their metabolites have been described as endogenous compounds in some animals, such as adult boars and entire male horses. The aim of this thesis was to identify markers of exogenous administration of 17βN in boars and horses. A method to quantify and identify metabolites of 17βN in the different metabolic fractions has been optimized and validated in boar and horse urine. The method has been applied to urine samples collected after administration of 17βN preparations and to urine samples from non-treated animals to identify endogenous compounds. The following metabolites were identified after the administration of 17βN: 17βN, NorA, NorE, NorEpia, BAB and ABB in boars and 17βN, 17αN, NorEpiA, ABA and ABB in horses. The endogenous compounds detected were: 17βN in boars and horses, and ABA and 17αN only in entire male horses. A chromatographic method by LC/MS/MS has been evaluated to quantify and 17βN sulfate in horse urine and it has been applied to analyze samples from non-treated and treated horses. Concentrations above 30 ng/ml of 17βN sulfate would be evidences of the exogenous administration of 17βN in entire male horses. Preliminary results indicate the capability of GC/C/IRMS analysis to differentiate the endogenous or exogenous origin of 17βN and their metabolites in boars and horses. / La 17β-nandrolona (17βN) s’usa il•legalment en l’esport i com a promotor del creixement en el bestiar. Tant la 17βN com alguns dels seus metabòlits s’han descrit com a compostos endògens en alguns animals, com els porcs adults i els cavalls mascles. L’objectiu de la tesi va ser identificar els marcadors de l’administració exògena de 17βN a porcs i cavalls. Es va optimitzar i validar un mètode basat en la CG/EM per quantificar i identificar els metabòlits de la 17βN en orina de porcs i cavalls en les diferents fraccions metabòliques i es va aplicar a l’anàlisis de mostres procedents d’estudis d’excreció de 17βN i de mostres procedents d’animals no tractats per identificar els compostos endògens. Els metabolits identificats després l’aministració de 17βN van ser: 17βN, NorA, NorE, NorEpiA, BAB i ABB en porcs i 17βN, 17αN, NorEpiA, ABA i ABB en cavalls. Els compostos endògens detectats van ser: 17βN per porcs i cavalls mascles i ABA i 17αN únicament pels cavalls. S’ha avaluat un mètode per CL/EM/EM per quantificar 17βN sulfat en orina de cavall que ha estat aplicat a l’anàlisi de mostres procedents de cavalls no tractats i tractats amb 17βN. Concentracions superiors a 30 ng/ml de 17βN sulfat en orina, seria indicació d’una administració exògena de 17βN. Resultats preliminars obtinguts pel mètode de GC/C/IRMS permeten indicar que mitjançant aquesta tècnica és possible diferenciar l’origen endogen o exogen de 17βN i els seus metabòlits en porcs i en cavalls.

Les glioblastomes (GBMs) sont des tumeurs cérébrales au pronostic défavorable. La résistance aux thérapies actuelles et la rechute des GBMs pourraient être due à l'existence de cellules aux propriétés souches. L'objectif de ma thèse a été la caractérisation des cellules souches de GBMs (CSGs) isolées à partir de tumeurs. L'analyse du profil souche et de pluripotence des CSGs a montré qu'elles sont maintenues dans un état souche par SOX2 et que COL1A1 et IFITM1 peuvent être des cibles thérapeutiques potentielles. L'étude de la radiosensibilité des CSGs à travers l'analyse des courbes de clonogénicité a mis en évidence deux groupes dont un «atypique» pouvant être composé de sous-populations de cellules aux radiosensibilités différentes qu'il conviendra de caractériser. L'étude de la réparation a mis en évidence deux autres groupes dont un ayant un fort potentiel de réparation qui exprime plus fortement le gène RAD51 après irradiations. Le traitement par un inhibiteur spécifique de RAD51 ralentirait la capacité de réparation de ces cellules. Malgré cette hétérogénéité, l'inhibition de la voie Hedgehog (HH) par un vecteur glucuronylé de la cyclopamine, activé par le microenvironnement tumoral, inhibe la prolifération et l'auto-renouvellement des CSGs in vitro et ralentit la croissance tumorale in vivo. La voie HH semble être une cible thérapeutique intéressante commune à toutes les CSGs. Néanmoins, il est nécessaire de prendre en compte l'hétérogénéité dans les populations tumorales pour le développement de la médecine personnalisée. / Glioblastomas (GBMs) are brain tumors with a poor prognosis. Their resistance to current therapies and the occurrence of tumor relapse may be related to the existence of cells bearing stem cell characteristics. The aim of this PhD research was to characterize glioblastoma stem cells (GSCs) having been isolated from tumors. Analysis of the stemness and pluripotency profiles of GSCs indicated that their stemness states are maintained by SOX2 and that COL1A1 and IFITM1 may be potential therapeutic targets. Clonogenic studies of GSC radiosensitivity underscored the presence of two groups, one of them composed of sub-populations of cells with different degrees of radiosensitivity that have yet to be fully characterized. Study of DNA repair capacity highlighted two additional groups including one with high repair potential overexpressing the RAD51 gene after 4Gy. However, treatment with RAD51 inhibitor is likely to slow down repair of GSC lesions. Notwithstanding GSC heterogeneity, in our study inhibition of the Hedgehog pathway (HH) by a cyclopamine glucuronid prodrug, activated by the tumor microenvironment, inhibited in vitro proliferation and self-renewal in all the GSCs tested and slowed down tumor growth in vivo. Hence, HH pathway appears to be conserved among GSCs and constitutes an interesting potential therapeutic target. With regard to the development of personalized medicine, it is nevertheless highly advisable to take into account the pronounded heterogeneity of tumor populations.