O-Acetyl-L-serine sulfhydrylase-A (OASS-A) forms acetate and L-cysteine from O-acetyl-L-serine (OAS) and sulfide. One molecule of the cofactor pyridoxal 5'- phosphate (PLP) is bound in each holoenzyme protomer.

This dissertation involves two different studies. The first concerns the real-time detection of microbial contamination in drinking water using intrinsic fluorescence of the microorganisms. The prototype, “Blinky”, uses LEDs that emit light at 365nm, 590nm, and 635nm for ultraviolet, amber, and red light, respectively. At 365 nm, the cellular components excited include reduced pyridine nucleotides (RPNs), flavins, and cytochromes to distinguish viable bacteria; at 590 nm, the cellular components excited include cytochromes for non-viable bacteria; at 635 nm, the cellular components excited include calcium dipicolinic acid (DPA) for spores. By using these three different wavelengths, the prototype can differentiate between viable and non-viable organisms and also has the potential to detect spores. The aim of this study was to improve the detection limit by modifying the design of the instrument and to establish the detection limit of viable and non-viable bacteria and spores. The instrument was modified by replacing existing LEDs with LEDs that had 50% more intensity. Two additional LEDs were added for amber and red light, bringing the total to four LEDs for each. The LEDs were also positioned closer to the photomultiplier tube so as to increase sensitivity. For UV, only two LEDs were used as previous. The detection limit of the viable bacteria was ~50 live bacteria/L. No change in the intrinsic fluorescence below the concentration of ~10⁸ dead bacteria/L was observed. The results for spore measurements suggested that most of the spores had germinated before or during the measurements and could not be detected. The instrument was successful in detection of viable bacteria and also differentiating viable and non-viable bacteria. The instrument was not successful in detection of spores. The second study was designed to assess the water quality of well construction in southeastern Tanzania. Three designs were tested: Msabi rope pump (lined borehole and covered), an open well converted into a closed well (uncovered well into a covered and lined well), and an open well (uncovered and may or may not be lined). The study looked at the microbial and chemical water quality, as well as turbidity. The survey included 97 water collection points, 94 wells and three rivers. For microbial analysis, heterotrophic plate count (HPC), total coliforms and E. coli tests were performed. Fifteen of these wells were further analyzed for microflora and diversity for wells comparison purposes, using culture methods, followed by polymerase chain reaction (PCR) and genome sequencing. Ten wells out of the fifteen were analyzed for calcium (water hardiness), potassium, nitrates, nitrites, chloride, fluoride, bromide, sulfate, iron, and arsenic. Two water collection points were also selected for organic compound analysis (gasoline components). All samples tested positive for coliforms. Two samples tested positive for Escherichia coli for the lined borehole (Msabi rope pump) and four samples from closed wells. All open wells tested positive for E. coli. There was more microbial diversity in open wells than the closed wells and Msabi rope pumps. Potential bacterial pathogens were detected in seven wells out of the fifteen examined. The wells that tested positive were one Msabi rope pump, one closed well; the rest were from open water sources. Open wells had high turbidity followed by closed wells. Msabi rope pumps had low turbidity comparing to the two wells designs. No traces of gasoline components were detected in any of the water sources. One well out of ten had high amounts of nitrates-nitrogen (> 10 mg/L). The results of this study showed that the Msabi rope pumps performed comparably to the closed wells in terms of microbial quality but performed better with regard to turbidity. The open wells performed poorly in terms of microbial water quality as well and turbidity. There was a statistical difference in HPC, total coliforms, E.coli numbers and turbidity between open wells, closed wells and the Msabi rope pumps. However, there was no statistical difference in HPC, total coliforms and E.coli numbers between the closed wells and Msabi rope pumps. Msabi rope pumps performed better in turbidity

Development of the advanced, real-time methods for monitoring of bioaerosols is becoming increasingly important. At present, the Ultraviolet Aerodynamic Particle Sizer (UVAPS, Model 3312, TSI, St. Paul., MN) is the only commercially available method for in-situ, continuous measurements of viable airborne microorganisms. Research included in this thesis aimed towards comprehensive evaluation of the method over a wide range of operating conditions, linking the experimental results to the theoretical basis of its design and operation, and to developing a scientific basis for its application to real-time monitoring of bioaerosols. Specifically, due to a growing concern in the general community about the environmental and health aspects of biological aerosols originated from various types of agricultural operations including animal farming, this research was focussed on developing a research methodology/strategy for applying the method to the investigation of bioaerosols in the swine confinement buildings (SCB). Investigations under controlled laboratory conditions were primarily concerned with selectivity, sensitivity, counting efficiency, and detection limits of the spectrometer. This study also examined the effect of physiological state (metabolic activity) of bacteria on the performance characteristics of the method. The practical implications of the research findings are discussed in this thesis. Further field investigations undertaken on a pig farm advanced understanding of the UVAPS performance in the real-life environmental settings. The research also provided a new insight on the particle size distribution and the effect of on-farm-activities on aerosol load inside the SCBs, for both biological and non-biological aerosols. This study has proved that the UVAPS is a powerful tool for investigation of viable bioaerosols in the environment. However, this method is limited to detection of active metabolising bacteria that excludes dormant bacterial spores. In addition, the method is very sensitive to physiological state of bacteria and to the effect of adverse environmental conditions on metabolic activity of airborne bacteria, which may decrease the amount of the intrinsic fluorophores in the cells below sensitivity level iv of the monitor. Possible limitations of this technology include also the lack of selectivity and thus interferences from the non-microbial organic components of airborne particles. In addition, the sensitivity of the method is insufficient for monitoring viable bacteria in the environments with relatively low concentrations of bioaerosols. In order to increase sensitivity of the method, it would be desirable to concentrate the bioaerosols into a smaller volume with the aim of high-volume virtual impactors (aerosol concentrators) prior to the monitoring. Therefore, in the indoor environments where an application of the concentrator is not feasible, the utilisation of the UVAPS may be problematic. Due to the intrinsic limitations, the method is not recommended for the direct measurements of viable bioaerosols and should be used in conjunction with the conventional biosamplers for obtaining more realistic insights into the microbial air quality. Nevertheless, the UVAPS has been found to be an adequate method for the investigation of the dynamics of biological aerosols in real-time. Overall, this thesis contributes to the advancing of the understanding of the method and may assist in developing new, more advanced technologies for the real-time monitoring of viable bioaerosols, as well as in developing sampling strategies for the application of the method to various bioaerosol studies.

Multi-view datasets arise naturally in statistical genetics when the genetic and trait profile of an individual is portrayed by two feature vectors. A motivating problem concerning the Skin Intrinsic Fluorescence (SIF) study on the Diabetes Control and Complications Trial (DCCT) subjects is presented. A widely applied quantitative method to explore the correlation structure between two domains of a multi-view dataset is the Canonical Correlation Analysis (CCA), which seeks the canonical loading vectors such that the transformed canonical covariates are maximally correlated. In the high dimensional case, regularization of the dataset is required before CCA can be applied. Furthermore, the nature of genetic research suggests that sparse output is more desirable. In this thesis, two regularized CCA (rCCA) methods and a sparse CCA (sCCA) method are presented. When correlation sub-structure exists, stand-alone CCA method will not perform well. To tackle this limitation, a mixture of local CCA models can be employed. In this thesis, I review a correlation clustering algorithm proposed by Fern, Brodley and Friedl (2005), which seeks to group subjects into clusters such that features are identically correlated within each cluster. An evaluation study is performed to assess the effectiveness of CCA and correlation clustering algorithms using artificial multi-view datasets. Both sCCA and sCCA-based correlation clustering exhibited superior performance compare to the rCCA and rCCA-based correlation clustering. The sCCA and the sCCA-clustering are applied to the multi-view dataset consisted of PrediXcan imputed gene expression and SIF measurements of DCCT subjects. The stand-alone sparse CCA method identified 193 among 11538 genes being correlated with SIF#7. Further investigation of these 193 genes with simple linear regression and t-test revealed that only two genes, ENSG00000100281.9 and ENSG00000112787.8, were significance in association with SIF#7. No plausible clustering scheme was detected by the sCCA based correlation clustering method. / Thesis / Master of Science (MSc)

La BSP1, principale protéine du plasma séminal bovin, interagit avec les membranes des spermatozoïdes et joue un rôle crucial dans les événements qui conduisent à la fécondité des spermatozoïdes, lors du phénomène de la capacitation. Le but de cette recherche est d’investiguer la nature de ces interactions. Ce travail vise à démontrer l’influence des lipides qui composent les membranes sur l’action de la protéine BSP1. À l’aide de la fluorescence intrinsèque de la protéine, l’affinité de la protéine a été caractérisée pour quatre systèmes lipidiques. Les résultats montrent que la composition lipidique affecte significativement l'affinité de la protéine pour les membranes. Nous avons observé l'ordre suivant : 1-palmitoyl-2-oléoyl-sn-glycéro-3-phosphocholine (POPC) > POPC/1-palmitoyl-2-oléoyl-sn-glycéro-3-phosphoéthanolamine (POPE) ≈ POPC/1-palmitoyl-2-oléoyl-sn-glycéro-3-phospho-L-sérine (POPS) > POPC/cholestérol. La protéine interagit préférentiellement avec POPC. La présence de POPE, POPS, ou cholestérol dans la membrane diminue systématiquement l’affinité. Il est connu que la présence de POPE ou cholestérol augmente l’empilement des lipides dans les membranes. Cet effet de condensation des chaînes pourrait être défavorable à l’insertion de la partie hydrophobe de la protéine dans les membranes et réduire ainsi l'affinité. La diminution de l’affinité de la protéine induite par la présence de POPS, un lipide chargé négativement, pourrait être associée aux interactions électrostatiques répulsives car la protéine porte une charge globale négative. La littérature mentionne que la BSP1 extrait sélectivement les phospholipides de type choline et le cholestérol lors de son association avec les membranes de spermatozoïdes. Un efflux lipidique est aussi observé avec des membranes modèles. Nous avons désiré caractériser la « solubilisation » des membranes par la BSP1, par diffusion dynamique de la lumière. Comme étape préliminaire, nous avons étudié comment le détergent Triton X-100 solubilise les membranes en utilisant cette technique. Les mesures démontrent que la composition lipidique des membranes (POPC, POPC/POPE, POPC/1-palmitoyl-2-oléoyl-sn-glycéro-3- [phospho-rac-(1-glycérol)] (POPG)) n’affecte pas le mécanisme général de solubilisation/reconstitution des membranes modèles. Il a été montré qu'il existe trois régions lors des processus de solubilisation pour les différents systèmes lipidiques : i) le détergent se distribue dans les membranes, ii) une coexistence de membranes saturées en détergents et de micelles mixtes de phospholipides/Triton X-100 et iii) exclusivement des micelles mixtes de phospholipides/Triton X-100. Nos résultats montrent que la forme conique de POPE augmente la résistance des membranes à la solubilisation. La présence de POPG, apportant une charge négative à l’interface des membranes, n’induit aucun changement aux processus de solubilisation/reconstitution des membranes par Triton X-100. La diffusion dynamique de la lumière a également permis d’observer si la protéine BSP1 induit des modifications morphologiques des membranes suite à son interaction avec les membranes de POPC. Nos observations n'ont montré aucune variation significative de la taille des particules lors du titrage des vésicules de POPC par la protéine, sur une gamme de rapport molaire de POPC/BSP1 variant de 20 à 0.6. Avec des compositions aussi différentes, on suppose une transition des vésicules saturées en protéine à des complexes de protéine avec un peu de lipides. Cependant, il semble impossible avec la diffusion dynamique de la lumière de différencier ces particules. / BSP1, the main protein in bovine seminal plasma, interacts with sperm membranes and plays a crucial role in events that lead to sperm fertility, during the capacitation. The purpose of this research is to investigate the nature of these interactions. This work aims to demonstrate the influence of the lipids that compose membranes on the action of the BSP1 protein. Using the intrinsic fluorescence of the protein, the affinity of the protein was characterized for four lipid systems. The results show that the lipid composition significantly affects the affinity of the protein for membranes. We observed the following order: 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) > POPC/1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) ≈ POPC/1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoserine (POPS) > POPC/cholesterol. The protein interacts preferentially with POPC. The presence of POPE, POPS, or cholesterol in membranes decreases systematically the affinity. It is established that the presence of POPE or cholesterol increases the packing of lipids in membranes. This condensation effect could be detrimental to the insertion of the hydrophobic part of the protein into the membranes and reduces, as a consequence, the affinity. The decrease in protein affinity induced by the presence of POPS, a negatively charged lipid, could be associated with repulsive electrostatic interactions as the protein global charge is negative. The literature mentions that BSP1 selectively extracts choline phospholipids and cholesterol when combined with sperm membranes. A lipid efflux is also observed with model membranes. We characterized membrane "solubilisation" by BSP1, using dynamic light scattering. As a preliminary step, we studied how Triton X-100 detergent solubilizes membranes using this technique. The measurements showed that the lipid composition of the membranes does not affect the general solubilization/reconstitution mechanism of the model membranes (POPC, POPC/POPE, POPC/1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1-rac-glycerol) (POPG)). It is known that three different regions exist during the solubilization process for the different lipid systems: i) the detergent is distributed in the membranes, ii) a coexistence of membranes saturated with detergents and mixed phospholipid/Triton X-100 micelles and iii) exclusively mixed phospholipid/Triton X-100 micelles. Our results show that the conical shape of POPE increases the resistance of the membranes to solubilization. The presence of POPG, bringing a negative charge at the membrane interface, does not induce any change in solubilization/reconstitution processes. Dynamic light scattering also made it possible to observe if the BSP1 protein induces morphological changes in the membranes following its interaction with POPC membranes. Our observations showed no significant variation in particle size during the titration of POPC vesicles by the protein, over a molar ratio range of POPC/BSP1 from 20 to 0.6. Considering such different compositions, a transition from vesicles saturated with protein to protein complexes with some lipids is assumed. However, it appeared impossible with dynamic light scattering to differentiate these particles.