The stochasticity of cellular processes and the small number of molecules in a cell make deterministic models inappropriate for modelling chemical reactions at the single cell level. The Chemical Master Equation (CME) is widely used to describe the evolution of biochemical reactions inside cells stochastically but is computationally expensive. The Linear Noise Approximation (LNA) is a popular method for approximating the CME in order to carry out inference and parameter estimation in stochastic models. Data from stochastic systems is often aggregated over time. One such example is in luminescence bioimaging, where a luciferase reporter gene allows us to quantify the activity of proteins inside a cell. The luminescence intensity emitted from the luciferase experiments is collected from single cells and is integrated over a time period (usually 15 to 30 minutes), which is then collected as a single data point. In this work we consider stochastic systems that we approximate using the Linear Noise Approximation (LNA). We demonstrate our method by learning the parameters of three different models from which aggregated data was simulated, an Ornstein-Uhlenbeck model, a Lotka-Voltera model and a gene transcription model. We have additionally compared our approach to the existing approach and find that our method is outperforming the existing one. Finally, we apply our method in microscopy data from a translation inhibition experiment.

Submitted by Maria Medeiros (maria.dilva1@ufcg.edu.br) on 2018-08-23T12:14:13Z No. of bitstreams: 1 MARCOS HENRIQUE SILVA ARAGÃO - DISSERTAÇÃO (PPGECA) 2017.pdf: 4540927 bytes, checksum: d25472de237d384e285a54b061998c1e (MD5) / Made available in DSpace on 2018-08-23T12:14:13Z (GMT). No. of bitstreams: 1 MARCOS HENRIQUE SILVA ARAGÃO - DISSERTAÇÃO (PPGECA) 2017.pdf: 4540927 bytes, checksum: d25472de237d384e285a54b061998c1e (MD5) Previous issue date: 2017-08 / Capes / No processo de produção do álcool são gerados até 15 litros de vinhoto por litro de álcool produzido. O vinhoto é um rejeito caracterizado por apresentar elevadas concentrações de material orgânico e nutrientes. Essas características fazem do vinhoto um substrato adequado para a produção de proteína unicelular (PUC) em sistemas de lodos ativados mediante a produção de excessiva de biomassa bacteriana, que é constituída por cerca de 60 -70 % de proteínas, podendo ser empregada como ração animal. O lodo do tipo PAO, produzido em sistemas EBPR, é caracterizado por ter uma taxa de decaimento bacteriano notadamente inferior em comparação aos microrganismos ordinários de lodos ativados. Isso representa uma vantagem para a produção de PUC, tanto pela manutenção de uma quantidade maior de biomassa, quanto pelos menores requisitos de oxigênio para a oxidação das células mortas. O desenvolvimento desses microrganismos no lodo exige uma mudança operacional no regime de aeração, introduzindo uma fase anaeróbia em cada batelada, na qual é realizada a alimentação com o substrato. Este trabalho teve como objetivos: avaliar e comparar a produção máxima de PUC a partir do vinhoto em sistemas de lodos ativados em função da idade de lodo tanto para os organismos heterotróficos ordinários (OHO) como para organismos acumuladores de polifosfato (PAO), determinar a eficiência de remoção de material orgânico e de nutrientes, a determinação das composições dos lodos e a determinação das principais constantes cinéticas e estequiométricas de cada um deles. Foi aplicada a respirometria como ferramenta para a caracterização cinética do consumo do material orgânico e adotada a teoria de lodos ativados proposta por van Haandel e Marais (1999). Foram operados e monitorados dois reatores em bateladas sequenciais com idade de lodo de 8 dias, sendo um deles submetido a uma interrupção no regime de aeração, introduzindo uma fase anaeróbia no início de cada batelada visando ao desenvolvimento de um lodo do tipo PAO. Os resultados obtidos demonstraram que houve uma produção satisfatória de lodo em ambos os reatores, bem como uma elevada remoção de material orgânico no efluente. Não se observou uma remoção em excesso de fósforo ou uma elevada fração de fósforo incorporada ao lodo produzido no sistema submetido à modificação no regime de aeração, além disso não foi observada uma produção mais elevada de lodo nesse sistema, indicando que o lodo PAO não foi produzido. A determinação do comportamento cinético dos lodos demonstrou que existe um elevado potencial de produção de PUC em sistemas lodos ativados tratando vinhoto, embora essa produção seja limitada pela capacidade de oxigenação dos aeradores e esteja condicionada ao avanço desses equipamentos. / In the process of producing the alcohol, up to 15 liters of vinasse per liter of alcohol produced are generated. The vinasse is a reject characterized by high concentrations of organic material and nutrients. These characteristics make vinasse a suitable substrate for the production of single cell protein (SCP) in activated sludge systems through the production of excessive bacterial biomass, which consists of about 60-70% of proteins, can be used as animal feed. The PAO-type sludge, produced in EBPR systems, is characterized by having a markedly lower bacterial decay rate compared to ordinary activated sludge microorganisms. This represents an advantage for the production of PUC, both by maintaining a higher amount of biomass and by lower oxygen requirements for the oxidation of dead cells. The development of these microorganisms in the sludge requires an operational change in the aeration regime, introducing an anaerobic phase in each batch, in which the feeding with the substrate is carried out. The objective of this work was to evaluate and compare the maximum production of SCP from vinasse as a function of sludge age in activated sludge systems for both ordinary heterotrophic organisms (OHO) and polyphosphate accumulating organisms (PAO). Efficiency of removal of organic matter and nutrients, determination of the sludge compositions and determination of the main kinetic and stoichiometric constants of each of them. It was applied the respirometry as a tool for the kinetic characterization of organic material consumption and adopted the theory of activated sludge proposed by van Haandel and Marais (1999). Two reactors were operated and monitored in sequential batches with a sludge age of 8 days, one of them being submitted to an interruption in the aeration regime, introducing an anaerobic phase at the beginning of each batch, aiming at the development of a PAO type sludge. The results showed that there was a satisfactory production of sludge in both reactors, as well as a high removal of organic material in the effluent. There was no excess removal of phosphorus or a high fraction of phosphorus incorporated into the sludge produced in the system submitted to the modification in the aeration regime, in addition no higher sludge production was observed in this system, indicating that the PAO sludge was not produced. The determination of the kinetic behavior of the sludge showed that there is a high potential in the PUC production in activated sludge treatment systems, although this production is limited by the oxygenation capacity of the aerators and is conditioned to the advance of these equipment.

The epicardium, a cell layer enveloping the heart muscle, drives embryonic heart development and heart repair in the adult zebrafish. Previous studies found the epicardium to consist of multiple cell populations with distinct phenotypes and functions. Here, I investigated epicardial heterogeneity in the developing zebrafish heart, focusing on the developmental gene program that is also reactivated during adult heart regeneration. Transcription factor 21 (Tcf21), T-box 18 (Tbx18) and Wilms' tumor suppressor 1b (Wt1b) are often used interchangeably to identify the zebrafish epicardium. Analyzing newly generated reporter lines and endogenous gene expression, I showed that the epicardial expression of tcf21, tbx18 and wt1b during development is heterogeneous. I then collected epicardial cells from newly generated reporter lines at 5 days-post-fertilization and performed single-cell RNA sequencing. I identified three distinct epicardial subpopulations with specific gene expression profiles. The first subpopulation expressed tcf21, tbx18 and wt1b and appeared to represent the main epicardial layer. The second subpopulation expressed tbx18, but not tcf21 or wt1b. Instead, it expressed smooth muscle markers and seemed restricted to the bulbus arteriosus. The third epicardial subpopulation only expressed tcf21 and resided within the epicardial layer. I compared the single-cell subpopulations with transcriptomic bulk data and visualized the expression of marker genes to investigate their spatial distribution. Using ATAC sequencing, I additionally identified open regulatory regions located in proximity to subpopulation-specific marker genes and showed subpopulation-specific activity in vivo. My results detail distinct cell populations in the developing zebrafish epicardium, likely to fulfil distinct and specific cellular functions. Future experiments will involve targeting signature genes enriched within each epicardial subpopulation, such as those encoding Adrenomedullin a (first subpopulation), Alpha Smooth Muscle Actin (second subpopulation) and Claudin 11a (third subpopulation), employing cell type-specific genome editing to test whether and how the identified heterogeneity underlies distinct epicardial cell fates and functions. Taken together, my work adds significantly to the understanding of the cellular and molecular basis of epicardial development and can offer novel insights in the context of heart regeneration.

Colonic stromal cells provide critical structural support but also regulate immunity, tolerance and inflammatory responses in the mucosa. Substantial variability and plasticity of mucosal stromal cells has been reported but a paucity of distinct marker genes exist to identify distinct cell states. Here single-cell RNA-sequencing is used to document heterogeneity and subtype specific markers of individual colonic stromal cells in human and mouse. Marker-free transcriptional clustering of fibroblast-like cells derived from healthy human tissue reveals distinct populations corresponding to myofibroblasts and three transcriptionally and functionally dissimilar populations of fibroblasts. A SOX6 high fibroblast subset occupies a position adjacent to the epithelial basement membrane and expresses multiple epithelial morphogens including WNT5A and BMP2. Additional fibroblast subtypes show specific enrichment for chemokine signalling and prostaglandin E₂ synthesis respectively. In ulcerative colitis, substantial remodelling occurs with depletion of the SOX6 high population and emergence of an immune enriched population expressing genes associated with fibroblastic reticular cells including CCL19, CCL21 and IL33. A large murine dataset comprising over 7,000 colonic mesenchymal cells from an acute colitis model and matched healthy controls reveals strong preservation of the SOX6 high and myofibroblast transcriptional signatures. Unsupervised pseudotemporal ordering is used to relate fibroblast subsets to one another producing a branched developmental hierarchy that includes a potential progenitor population with mesothelial characteristics at its origin. This work provides a molecular basis for re-classification of colonic stromal cells and identifies pathological changes in these cells underpinning inflammation in UC.

Live single-cell imaging has emerged as an advanced single-cell study tool for approaching a quantitative understanding of many biological questions in recent years. In previous cell studies using bulk cell measurements, the population averages can miss the information from cell to cell variability and mask the underlying signaling networks and mechanisms. Currently, some single cell analysis methods, including but not limited to, live single-cell imaging experiments that built around a fluorescent imaging setup and microfluidic devices enable the measurement and analysis of cell dynamics and responses of single cells across a population and across time. Furthermore, by changing the cells’ environmental conditions in well controlled ways, e.g. balanced steady growth, or temporal pulses, live single-cell imaging can record the cellular behaviors corresponding to these changes in exquisite details. An important question of current interest in both developmental, stem cell and cancer biology is the question of epigenetic differentiation. Continuous long-term live single-cell observations offer insights into the molecular control of cell fate. However, maintaining the imaged cells in a healthy state remains a major challenge. One of our aims in this work was to develop a semi-automated single-cell live imaging and analysis platform to obtain dynamic information of the cellular processes. An imaging incubator that controls and regulates the environmental conditions of the imaged cells also had to be designed and tested. In this thesis, I address the key design considerations of developing a single-cell live imaging platform and demonstrate the capability of this technology through three case studies. To test the design and fabrication of microfluidic devices and micro-valves in imaging malaria infected red blood cells (iRBCs), I recorded the flow of iRBCs through microfluidic channels and constrictions in Chapter 3. Our results illustrate the behaviors of iRBCs with different flow rates and the potential to offer dynamic control in studying the infection probability of iRBCs by implementing the micro-valve system. In order to develop a more adaptable live cell imaging platform, we further developed our semi-automated imaging software and in house built imaging incubator to explore the link between proliferation and differentiation of CD4+ T cells in Chapter 4. By using cells expressing an IL-13-GFP reporter, we distinguished between differentiating and non-differentiating CD4+ T cell population and demonstrated a positive association between cycling differentiation of CD4+ T cells. In Chapter 5, we incorporated the FUCCI cell reporter system in our single cell live imaging system to reveal the effect of different media conditions on the cell cycle progression and cell fate choices of mouse embryonic stem (mES) cells. By improving different factors such as longer pre-incubation time before imaging and exchanging media during the experiments, we maintained a healthy state of mES cells during live cell imaging for extended periods. We observed significant differences in time between divisions of mES cells cultured in 2i +LIF and serum + LIF media, and also small but significant differences in durations of sub-cell cycle phases (G1,G1/S,S/G2/M) between the two media conditions. We further applied this imaging setup to study the behaviors of differentiating mES cells in vitro, and observed lengthening of the G1 phase for both 2i-LIF and serum-LIF cells in agreement with literature. Overall, our semi-automated single cell imaging platform not only offers adjustable intervals between fluorescent imaging, but also provides a constant temperature and gas feeding devices that allows the cells to proliferate for extended microscope imaging. Commercially produced incubators that fit onto the microscope stage and satisfied all requirements in restriction of the cell movement, gas feeding, temperature regulation and optical accessibility are not easily available. Thus, there exists a significant potential for our imaging setup to provide a versatile and adaptable live cell imaging platform for both academia and industrial researchers.

abstract: Cell adhesion is an important aspect of many biological processes. The atomic force microscope (AFM) has made it possible to quantify the forces involved in cellular adhesion using a technique called single cell force spectroscopy (SCFS). AFM based SCFS offers versatile control over experimental conditions for probing directly the interaction between specific cell types and specific proteins, surfaces, or other cells. Transmembrane integrins are the primary proteins involved in cellular adhesion to the extra cellular matix (ECM). One of the chief integrins involved in the adhesion of leukocyte cells is αMβ2 (Mac-1). The experiments in this dissertation quantify the adhesion of Mac-1 expressing human embryonic kidney (HEK Mac-1), platelets, and neutrophils cells on substrates with different concentrations of fibrinogen and on fibrin gels and multi-layered fibrinogen coated fibrin gels. It was shown that multi-layered fibrinogen reduces the adhesion force of these cells considerably. A novel method was developed as part of this research combining total internal reflection microscopy (TIRFM) with SCFS allowing for optical microscopy of HEK Mac-1 cells interacting with bovine serum albumin (BSA) coated glass after interacting with multi-layered fibrinogen. HEK Mac-1 cells are able to remove fibrinogen molecules from the multi-layered fibrinogen matrix. An analysis methodology for quantifying the kinetic parameters of integrin-ligand interactions from SCFS experiments is proposed, and the kinetic parameters of the Mac-1 fibrinogen bond are quantified. Additional SCFS experiments quantify the adhesion of macrophages and HEK Mac-1 cells on functionalized glass surfaces and normal glass surfaces. Both cell types show highest adhesion on a novel functionalized glass surface that was prepared to induce macrophage fusion. These experiments demonstrate the versatility of AFM based SCFS, and how it can be applied to address many questions in cellular biology offering quantitative insights. / Dissertation/Thesis / Doctoral Dissertation Physics 2016

abstract: In the past decades, single-cell metabolic analysis has been playing a key role in understanding cellular heterogeneity, disease initiation, progression, and drug resistance. Therefore, it is critical to develop technologies for individual cellular metabolic analysis using various configurations of microfluidic devices. Compared to bulk-cell analysis which is widely used by reporting an averaged measurement, single-cell analysis is able to present the individual cellular responses to the external stimuli. Particularly, oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) are two key parameters to monitor heterogeneous metabolic profiles of cancer cells. To achieve multi-parameter metabolic measurements on single cells, several technical challenges need to be overcome: (1) low adhesion of soft materials micro-fabricated on glass surface for multiple-sensor deposition and single-cell immobilization, e.g. SU-8, KMPR, etc.; (2) high risk of using external mechanical forces to create hermetic seals between two rigid fused silica parts, even with compliance layers; (3) how to accomplish high-throughput for single-cell trapping, metabolic profiling and drug screening; (4) high process cost of micromachining on glass substrate and incapability of mass production. In this dissertation, the development of microfabrication technologies is demonstrated to design reliable configurations for analyzing multiple metabolic parameters from single cells, including (1) improved KMPR/SU-8 microfabrication protocols for fabricating microwell arrays that can be integrated and sealed to 3 × 3 tri-color sensor arrays for OCR and ECAR measurements; (2) design and characterization of a microfluidic device enabling rapid single-cell trapping and hermetic sealing single cells and tri-color sensors within 10 × 10 hermetically sealed microchamber arrays; (3) exhibition of a low-cost microfluidic device based on plastics for single-cell metabolic multi-parameter profiling. Implementation of these improved microfabrication methods should address the aforementioned challenges and provide a high throughput and multi-parameter single cell metabolic analysis platform. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2017

La microscopie par résonance de plasmon de surface (SPR) à couplage par prisme a vu le jour à la fin des années 60. Le principal avantage de cette technique d'imagerie optique réside dans sa très grande sensibilité à de faibles variations d'indice optique ou d'épaisseurs à la surface d'un métal. De ce fait, le suivi d'interactions biologiques peut se faire en temps réel sans avoir recours à l'utilisation de marqueurs fluorescents ou enzymatiques. Depuis plus de 30 ans, la microscopie SPR s'est imposée comme la technique de référence de biodétection sans marquage. Ses champs d'application vont de la détermination de constantes d'affinité à la détection de bactéries pathogènes, en passant par la biologie cellulaire. Jusqu'à présent, on pensait la résolution spatiale limitée par la longueur de propagation des plasmons de surface. Or, de nombreux exemples ne corroborent pas cette hypothèse. Dans cette thèse, nous montrons qu'à ce phénomène de propagation se rajoute des aberrations optiques induites par l'utilisation d'un prisme pour coupler la lumière et les plasmons de surface. Nous expliquons ainsi pourquoi les résolutions expérimentales étaient souvent bien moins bonnes que celles attendues. Par l'analyse de la formation des images et la quantification des aberrations, nous aboutissons à deux nouvelles configurations optiques optimisées pour la résolution. Nous analysons ensuite quel métal offre le meilleur compromis entre longueur de propagation et sensibilité. Expérimentalement, nous obtenons une résolution comprise entre 1,5 et 4 μm suivant la direction, sur des champs de vision de plusieurs mm2, et ce, avec une sensibilité standard en biodétection. Nous sommes ainsi en mesure d'observer simultanément plusieurs milliers de cellules individuelles, eucaryotes et procaryotes. Finalement, nous développons un prototype dédié au suivi en temps réel de sécrétions de protéines par des cellules immunitaires. Les limites de la microscopie SPR et les solutions qui permettraient de faire aboutir ce type d'étude sont examinées. Des études préliminaires sont aussi menées sur l'amélioration de la détection de bactéries. / Prism-based surface plasmon resonance (SPR) microscopy is an optical imaging technique invented in the late 60s'. Its main advantage lies in its high sensitivity to optical index or thickness variations at a metal surface. Therefore, the monitoring of biological reactions can be performed in real-time without labeling agent such as fluorescence or enzymes. Over the last 30 years, SPR microscopy has become the major technique in label-free biodetection. The field of application range from the determination of affinity constant in biochemistry to the detection of pathogenic bacteria via cellular biology. Until now, the propagation length of the surface plasmons has been considered as the spatial resolution limit. However, many examples do not support this statement. In this PhD thesis, we demonstrate that the resolution is also limited by optical aberrations induced by the prism used to couple light and surface plasmons. Thus, we are able to explain why the experimental resolution was usually worse than the predicted one. The analysis of the image formation and the quantification of aberrations lead us to suggest two new optical configurations optimized for resolution. We also analyze which metal exhibits the better trade-off between propagation length and sensitivity. Experimentally, we obtain a resolution between 1.5 and 4 μm depending on the direction, on field-of-view up to several mm2, and with a standard sensitivity for biodetection (monolayer of DNA). We are then able to observe simultaneously several thousands of individual eukaryote and prokaryote cells. Finally, we develop a prototype dedicated to the real-time monitoring of protein secretion by immune cells. The limits of SPR microscopy and the solutions which could allow this kind of study are discussed. Preliminary results on the improvement of bacterial detection are also presented.

abstract: This thesis research focuses on developing a single-cell gene expression analysis method for marine diatom Thalassiosira pseudonana and constructing a chip level tool to realize the single cell RT-qPCR analysis. This chip will serve as a conceptual foundation for future deployable ocean monitoring systems. T. pseudonana, which is a common surface water microorganism, was detected in the deep ocean as confirmed by phylogenetic and microbial community functional studies. Six-fold copy number differences between 23S rRNA and 23S rDNA were observed by RT-qPCR, demonstrating the moderate functional activity of detected photosynthetic microbes in the deep ocean including T. pseudonana. Because of the ubiquity of T. pseudonana, it is a good candidate for an early warning system for ocean environmental perturbation monitoring. This early warning system will depend on identifying outlier gene expression at the single-cell level. An early warning system based on single-cell analysis is expected to detect environmental perturbations earlier than population level analysis which can only be observed after a whole community has reacted. Preliminary work using tube-based, two-step RT-qPCR revealed for the first time, gene expression heterogeneity of T. pseudonana under different nutrient conditions. Heterogeneity was revealed by different gene expression activity for individual cells under the same conditions. This single cell analysis showed a skewed, lognormal distribution and helped to find outlier cells. The results indicate that the geometric average becomes more important and representative of the whole population than the arithmetic average. This is in contrast with population level analysis which is limited to arithmetic averages only and highlights the value of single cell analysis. In order to develop a deployable sensor in the ocean, a chip level device was constructed. The chip contains surface-adhering droplets, defined by hydrophilic patterning, that serve as real-time PCR reaction chambers when they are immersed in oil. The chip had demonstrated sensitivities at the single cell level for both DNA and RNA. The successful rate of these chip-based reactions was around 85%. The sensitivity of the chip was equivalent to published microfluidic devices with complicated designs and protocols, but the production process of the chip was simple and the materials were all easily accessible in conventional environmental and/or biology laboratories. On-chip tests provided heterogeneity information about the whole population and were validated by comparing with conventional tube based methods and by p-values analysis. The power of chip-based single-cell analyses were mainly between 65-90% which were acceptable and can be further increased by higher throughput devices. With this chip and single-cell analysis approaches, a new paradigm for robust early warning systems of ocean environmental perturbation is possible. / Dissertation/Thesis / Ph.D. Civil and Environmental Engineering 2013

The initiating events of ovarian cancer remain unknown, but an established risk factor is use of estrogen therapy by post-menopausal women where there is a positive correlation between duration of use and risk for disease. Mouse models of ovarian cancer have shown that exposure to exogenous 17β-estradiol (E2) accelerates tumour onset so this study aims to investigate the E2 signalling mechanisms responsible for sensitizing ovarian epithelial cells to transformation. By developing model systems that are responsive to E2 manipulation, we showed that E2 induces the formation of epithelial dysplasias both in vitro and in vivo. microRNA microarray was used to discover that E2 up-regulates microRNA-378 via the ESR1 pathway, resulting in the down-regulation of a tumour suppressor gene called Disabled-2 (Dab2). E2 suppression of Dab2 was found to result in increased proliferation, loss of contact inhibition, epithelial dysplasia, and increased sensitivity to transformation. This mechanism was also found to be active in mouse fallopian tube epithelium and human ovarian cancer cells. Single-cell RNA sequencing and trajectory analysis was subsequently used to explore additional signalling mechanisms that might contribute to the emergence of dysplastic lesions induced by E2. Multiple molecular signalling pathways dysregulated by E2 were identified and this revealed several possible biomarkers to be investigated for early detection of ovarian cancer. In the context of a current lack of strategies for ovarian cancer prevention or early detection, this work represents a significant advance in our understanding of how E2 promotes ovarian cancer initiation.