Tissue transglutaminase (TG2) is a calcium-dependent enzyme that natively catalyses the formation of isopeptidic bonds between protein- or peptide-bound glutamine and lysine residues. Physiologically, it is ubiquitously expressed in tissues, with roles in cellular differentiation, extracellular matrix stabilisation, and apoptosis, among others. However, its unregulated activity has been associated with various pathologies including fibrosis, cancer and celiac disease. Since most pathologies are associated with an increased transamidation activity, efforts have been directed towards the development of TG2 inhibitors. In this context, the work described in this thesis is centred on reversible inhibitors, building on recent work done within the Keillor group in two directions, namely localisation and potency. In a localisation-driven approach, we developed a photolabile derivative of a known reversible inhibitor, in order to form a covalent bond with the enzyme and determine the inhibitor’s binding site. In tandem, we optimised a protocol for the expression of TG2 incorporating ArgΔ10 and LysΔ8, amino acids that are 13C- and 15N-labelled to provide a mass shift of 10 and 8 Da, respectively, compared to the corresponding unlabelled amino acids. This “heavy” TG2 was developed as a tool for reference in the analysis of the tryptic digest of labelled protein. In a potency-driven approach, based on the observation that previous trans cinnamoyl inhibitor scaffolds were susceptible to nucleophilic attack by glutathione, we developed a bis(triazole) scaffold with reduced electrophilicity. The preparation of a small library of compounds showed that this scaffold demonstrates a preference for electron-withdrawing substituents, such as nitro groups. Continuing in a potency-driven approach, and inspired by work done in the identification of glutathione-resistant scaffolds, we studied a new alkynyl scaffold. While still susceptible to glutathione addition, these compounds showed a marked improvement in potency, with the lead compound having an IC50 of 930 nM and being established as a competitive inhibitor with a Ki of 420 nM, our most potent reversible inhibitor to date. Furthermore, this scaffold also produced an inhibitor lacking nitro groups (to limit eventual cellular toxicity), but maintaining good potency, with an IC50 value of 3.03 μM.

No description available.

Indiana University-Purdue University Indianapolis (IUPUI) / The human aldehyde dehydrogenase (ALDH) superfamily consists of 19 isoenzymes that are critical for normal physiology as well as the removal of toxic aldehydes. Members of the ALDH1/2 family have vital roles in cell signaling during early development, ethanol metabolism, and the removal of aldehydes derived from oxidative stress. We sought to develop selective compounds toward ALDH2 to help determine its individual contribution to biological function, as many of the ALDH1/2 family possess overlapping substrate preferences. A high-throughput screen of over 100,000 compounds uncovered a class of aromatic lactones which inhibit the ALDH1/2 enzyme family. The lactones were then characterized using a combination of enzyme kinetics, X-ray crystallography, and cell culture experiments. We found that many of the lactones are over ten times more potent toward ALDH2 than daidzin, a previously described ALDH2 inhibitor. Our ability to produce many more ALDH isoenzymes allowed us to determine that daidzin is not as selective as previously believed, inhibiting ALDH2, ALDH1B1, and ALDH1A2 with equal potency. This inhibition pattern was seen with several of the aromatic lactones as well. Structural studies show that many of the lactones bind between key aromatic residues in the ALDH1/2 enzyme substrate-binding sites. One lactone in particular mimics the position of an aldehyde substrate and alters the position of the catalytic cysteine to interfere with the productive binding of NAD+ for enzyme catalysis. Further characterization of related compounds led to the realization that the mechanism of inhibition, potency, and selectivity differs amongst the lactones based off the substituents on the aromatic scaffold and its precise binding location. Two of these compounds were found to be selective for one of the ALDH1/2 family members, BUC22, selective for ALDH1A1, and BUC27, selective for ALDH2. BUC22 demonstrates ten-fold selectivity for ALDH1A1 over ALDH1A2 and does not inhibit the remaining ALDH1/2 enzymes. Additionally, treatment with BUC22 led to decreased growth of triple-negative breast cancer cells in culture. BUC27 inhibits ALDH2 with the same potency as daidzin. Both BUC22 and BUC27 could be further developed to use as chemical tools to better understand the functional roles of ALDH1A1 and ALDH2 in biological systems.

The enzyme soluble epoxide hydrolase (sEH) plays a major role in the pathogenesis and pathophysiology of neurodegenerative diseases like depression by catalyzing the hydrolysis of epoxyeicosatrienoic acids (EETs) into dihydroxyicosatrienoicacids (DHETs), its less biologically active form, influencing the anti-inflammatory system and promoting inflammation. Therefore, inhibiting sEH leads to increased levels of EETs, reducing inflammation, especially in the brain and can help mitigate neurodegenerative diseases. This study investigated sEH inhibition by a phenolic carotenoid compound, astaxanthin and its inhibitory mechanism of action. Enzyme inhibitory activity and kinetics demonstrated that astaxanthin had a half-maximal inhibitory concentration (IC50) of 26 ± 0.92 μM and is a mixed-non-competitive inhibitor of sEH. In silico ADME/tox analysis showed that astaxanthin is bioavailable, biostable, and non-toxic when taken orally. Molecular docking study demonstrated that astaxanthin binds to an allosteric site of sEH and formed a contact and clashing-only interaction with the ASP333 residue of the hydrolase pocket of sEH. In this study, we highlight the potential therapeutic application of astaxanthin as a natural sEH inhibitor in the treatment of inflammation-related diseases, particularly neurodegenerative diseases.

Fatty acid amide hydrolase (FAAH) is an enzyme that terminates the signaling role played by the lipid mediators, N- acylethanolamines (NAEs), present both in plants and animals. FAAH is responsible for NAE hydrolysis and has been extensively studied in mammalian systems and the model plant Arabidopsis thaliana; it has been reported in various organisms as well as some crop plants such as rice and Medicago truncatula. To understand the role of FAAH in diverse organisms, here we report the identification and biochemical characterization of a FAAH homolog in tomato. Previously identified and cloned candidate FAAH from tomato was expressed in Escherichia coli as a fused protein with 6X his-tag for identification. Supernatant containing recombinant FAAH showed the ability to hydrolyze NAE substrates. The optimal reaction conditions for enzyme assay and kinetic parameters for tomato FAAH were determined and effect of inhibitor on enzyme was determined. Characterization of FAAH in tomato will contribute to further understanding of NAE metabolic pathway and its implications.

Polyelectrolyte complex coacervates (PECCs) result from liquid-liquid phase separation (LLPS) in solutions containing oppositely charged polymers 1. Multiphase polyelectrolyte complex coacervates (MPECCs) result from the combination of multiple, specific PECCs 2. The encapsulation of proteins in PECCs can serve as promising vehicles for the effective delivery of protein-based therapeutics, which are notoriously difficult to deliver. The encapsulation of model proteins, such as Bovine Serum Albumin (BSA) 3 or Human Hemoglobin (Hb) 4 have illustrated the protein-encapsulating capabilities of these PECC systems. The encapsulation of proteins in MPECCs is a topic that has yet to be explored; however, it can serve to mimic the structure and function of multiphase membraneless organelles, which are abundantly available in cells. This project sought to understand and quantify the encapsulation of enzymes in both PECC and MPECC models; as well as evaluate their efficiency upon encapsulation, as enzymes are simply proteins with catalytic functions 5. A synthesized library of charged, heterochiral polypeptides were used to form both PECC and MPECC systems. Glucose oxidase (GOx) and horseradish peroxidase (HRP) were the enzymes chosen to be assessed in both PECC and MPECC systems. Turbidity measurements, in terms of percent mole of polycation, were used to determine the optimal stoichiometric ratio between the polyanion and polyanion, in the presence of a given concentration of both or either enzyme, in which maximum complex formation occurred. Here we report that a 1:1 stoichiometric ratio of polycation to polyanion in either a solution with 25ug/mL HRP and 25ug/mL GOx, a solution with 50ug/mL GOx, or a solution with 50ug/mL HRP leads to the highest level of complexation. Enzyme encapsulation efficiency of individual PECCs for both enzymes was assessed using the Bradford assay, in which the supernatant was used to determine the concentration of enzyme left in the PECC post-centrifugation. Here we report that all PECC systems were able to encapsulate both GOx and HRP. Higher encapsulation efficiencies were seen with GOx samples compared to HRP samples. Enzymatic activity and efficiency were assessed using the 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) assay in the presence of ß-D-glucose. The chromogenic change in intensity over time of each sample was assessed using optical microscopy. Michaelis-Menten graphs were made from the data collected. The resulting data was used to evaluate the Km and Vmax of the enzyme cascade in PECC and MPECC systems compared to a control. Here we report that enzyme cascade efficiency varied among PECC and MPECC samples, with some being more efficient than others. We find that both PECC and MPECC systems generally have lower enzyme-substrate affinity (higher Km) compared to performing the reactions in water. However, this may be related to the need for the substrate to diffuse into a different phase or phases. Interestingly, many of the PECC and MPECC systems have lower Vmax values compared to the water control, indicating a faster enzyme saturation. The enzyme kinetics and efficiency could also be controlled by varying the location of the enzymes in each phase within the MPECC systems. Overall, we show that using MPECC systems allows one to select advantageous properties of individual PECCs and combine them together.

Propionyl carboxylase purified from bovine liver mitochondria catalyzes the carboxylation of 992 micromoles of propionyl-CoA per hour per milligram of protein. Relative carboxylation rates for acetyl-, propionyl-, butyryl-, and valeryl-CoA remain constant during purification. The carboxylase is inhibited by PCMB, N-ethylmaleimide, and iodoacetamide; and the inhibition by PCMB can be almost completely reversed by GSH. The K_m values for acetyl-CoA, propionyl-CoA, butyryl-CoA, valeryh-CoA, propionyl pantetheine, ATP, and HCOj were determined. The K_m values for the aeyl-CoA derivatives are approximately the same while there is a 200-fold difference between the V_m values for propionyl-CoA and valeryl-CoA. Coenzyme A and valeryl-CoA, but not propionyl pantetheine were found to be competitive inhibitors of propionyl carboxylase. The apparent equilibrium constant for the enzymatic propionyl-CoA carboxylation reaction at pH 8.15 and 37°c is 8.1 x 10^-3 and the Δ F°₃₁₀ calculated from this constant is 2970 calories per mole. / Master of Science

Thesis (MSc)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: Progress in computational systems biology depends crucially on the availability of generic rate equations that accurately describe the behaviour and regulation of catalysed processes over a wide range of conditions. Such equations for ordinary enzyme-catalysed reactions have been developed in our group and have proved extremely useful in modelling metabolic networks. However, these networks link to growth and reproduction processes through template-directed synthesis of macromolecules such as polynucleotides and polypeptides. Lack of an equation that captures such a relationship led us to derive a generic rate equation that describes catalysed, template-directed polymerisation reactions with varying monomer stoichiometry and varying chain length. A model describing the mechanism of a generic template-directed polymerisation process in terms of elementary reactions with mass action kinetics was developed. Maxima, a computational algebraic solver, was used to determine analytical expressions for the steady-state concentrations of the species in the equation system from which a steady-state rate equation could be derived. Using PySCeS, a numerical simulation platform developed in our group, we calculated the time-dependent evolution and the steadystates of the species in the catalytic mechanisms used in the derivation of the rate equations. The rate equation was robust in terms of being accurately derived, and in comparison with the rates determined with PySCeS. Addition of more elongation steps to the mechanism allowed the generalisation of the rate equation to an arbitrary number of elongations steps and an arbitrary number of monomer types. To test the regulatory design of the system we incorporated the generic rate equation in a computational model describing a metabolic system consisting of multiple monomer supplies linked by a template-directed demand reaction. Rate characteristics were chosen to demonstrate the utility of the simplified generic rate equation. The rate characteristics provided a visual representation of the control and regulation profile of the system and showed how this profile changes under varying conditions. / AFRIKAANSE OPSOMMING: Die beskikbaarheid van generiese snelheidsvergelykings wat die gedrag en regulering van gekataliseerde prosesse akkuraat oor ’n wye reeks omstandighede beskryf is van kardinale belang vir vooruitgang in rekenaarmatige sisteembiologie. Sulke vergelykings is in ons groep ontwikkel vir gewone ensiem-gekataliseerde reaksies en blyk uiters nuttig te wees vir die modellering van metaboliese netwerke. Hierdie netwerke skakel egter deur templaat-gerigte sintese van makromolekule soos polinukleotiede en polipeptiede aan groei- en voorplantingsprosesse. Die gebrek aan vergelykings wat sulke verwantskappe beskryf het ons genoop om ’n generiese snelheidsvergelyking af te lei wat gekataliseerde, templaatgerigte polimerisasie-reaksies met wisselende monomeerstoigiometrie en kettinglengte beskryf. ’n Model wat die meganisme van ’n generiese templaat-gerigte polimerisasie-proses in terme van elementêre reaksies met massa-aksiekinetika beskryf is ontwikkel. Maxima, ’n rekenaarmatige algebraïese oplosser, is gebruik om analitiese uitdrukkings vir die bestendige- toestand konsentrasies van die spesies in die vergelyking-stelsel te vind. Hierdie uitdrukkings is gebruik om ’n bestendige-toestand snelheidsvergelyking af te lei. Ons het die tyd-afhanklike progressie en die bestendige toestande bereken van die spesies in die katalitiese meganismes wat gebruik is in die afleiding van die snelheidsvergelykings. Die rekenaarprogram PySCeS is ’n numeriese simulasieplatform wat in ons groep ontwikkel is. Die snelheidsvergelyking blyk akkuraat afgelei te wees en is in ooreenstemming met snelhede deur PySCeS bereken. Die toevoeging van verdere verlengingstappe tot die meganisme het dit moontlik gemaak om die snelheidsvergelyking te veralgemeen tot ’n arbitrêre hoeveelheid verlengingstappe en monomeertipes. Om die regulatoriese ontwerp van die sisteem te toets het ons die generiese snelheidsvergelyking in ’n rekenaarmatige model geïnkorporeer wat ’n metaboliese sisteem bestaande uit verskeie monomeer-aanbodblokke en ’n templaatgerigte aanvraagblok beskryf. Snelheidskenmerkanalise is gekies om die nut van die vereenvoudigde generiese snelheidsvergelyking te demonstreer. Met hierdie snelheidskenmerke kon ons die kontrole- en reguleringsprofiel van die stelsel visualiseer en wys hoe hierdie profiel verander onder wisselende omstandighede.

Thesis (MSc)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Metabolic networks of cellular systems are complex, in that there are numerous components with multiple non-linear interactions. To understand how these networks work they are often split into manageable pieces and studied individually. However, an individual part is unable to account for the complex properties of systems. In order to study these interactions the eld of systems biology has developed. Systems biology makes use of computers to construct models as a method to describe aspects of living systems. Using cellular pathways, kinetic models of metabolic pathways can be constructed and used as a tool to study the biological systems and provide a quantitative description. This thesis describes the quantitative analysis of a bacterium using a systems biology approach. Zymomonas mobilis is a rod shaped, Gram-negative, non-mobile facultative anaerobe and has one of the fastest observed fermentations, yet least energy e cient extractions found in nature. Furthermore it is the only known micro-organism to use the Entner-Doudoro (2-keto-3-deoxy-6- phosphogluconate) pathway anaerobically. The low energy yield of fermentation in Z. mobilis is a result of the usage of the Entner-Doudoro glycolytic pathway, which has half the energy yield per mol substrate compared to the well known Embden-Meyerhof-Parnas glycolytic pathway. The work presented in this thesis forms part of a larger project to compare glycolytic regulation in di erent micro-organisms Z. mobilis, Escherichia coli, Saccharomyces cerevisiae and Lactococcus lactis. These organisms were chosen based on their usage of di erent glycolytic mechanisms. Kinetic models are suitable tools to draw a comparison between these organisms. The emphasis here is on the construction of a kinetic model of the Entner-Doudoroff glycolytic pathway as it occurs in Z. mobilis. The aim of this thesis was to characterise as many of the Entner-Doudoro pathway enzymes as possible, under standard conditions. This was done using enzyme assays, to obtain the kinetic parameters of each of the enzymes. Microtitre plate assays were used to characterise most of the enzymes of the Entner-Doudoro pathway. However, not all characterisations could be done using plate assay methods, as some intermediates were not commercially available to perform coupled assays. Nuclear magnetic resonance (NMR) spectroscopy was used to characterise these enzymes. These experimentally obtained parameters were then incorporated in a mathematical framework. Time simulations on the initial model were unable to reach a steady-state, with a build up of metabolic intermediates. A secondary model was constructed (using calculated maximal activities) which allowed us to identify discrepancies in the initial model. This showed that the experimentally determined maximal activities of three enzymes in lower glycolysis were unrealistically low, which might be due to protein denaturation by sonication. A nal model was constructed which incorporated a correction factor for these three enzymes. The models' predicted output (steady-state concentrations and ux) was compared to that of either literature or experimentally determined values, as a method to validate the model. The model output compared well to literature values. The constructed and partially validated kinetic model was then used as an analytical tool to identify points of control and regulation of glycolysis in Z. mobilis. The model presented in this work was also compared to published models. Our model relies much less on literature obtained values, and uses kinetic parameters experimentally determined under the same conditions. The parameters of the published models were obtained from the literature and in many instances, the assay conditions for these parameters were set-up to yield the maximum activity under non-physiological conditions. Furthermore, the number of excluded or assumed parameters is much less in our model. However, introduction of a milder, more predictable extraction technique for preparing cell lysates, should be considered for future work, to obtain the parameters that was not determined during this study. The published models do include reactions not included in our model (e.g ATP metabolism), which should be considered for inclusion, as we strive to construct a detailed kinetic model of glycolysis in Z. mobilis in the future. / AFRIKAANSE OPSOMMING: Sellul^ere metaboliese netwerke is komplekse stelsels, omdat hulle bestaan uit talle komponente met verskeie nie-lineêre interaksies. Om die funksionering van hierdie netwerke te verstaan, word hulle dikwels in hanteerbare stukke verdeel en individueel bestudeer. 'n Enkele komponent is egter nie in staat om die komplekse eienskappe van sulke stelsels te verklaar nie. Die veld van sisteembiologie het ontwikkel met die doel om sulke stelsels te bestudeer. Sisteembiologie maak gebruik van rekenaarmodelle as 'n metode om aspekte van lewende sisteme te beskryf. Kinetiese modelle van metaboliese paaie word gebou en gebruik as gereedskap om die biologiese stelsels te bestudeer en 'n kwantitatiewe beskrywing te bekom. Hierdie tesis beskryf die kwantitatiewe ontleding van 'n bakterie deur middel van 'n sisteembiologiese benadering. Zymomonas Mobilis is 'n staafvormige, Gram-negatiewe, nie-mobiele fakultatiewe ana erobe, en het een van die vinnigste waargenome fermentasies, maar met die minste energie-doeltre ende ekstraksie wat in die natuur aangetref word. Verder is dit die enigste bekende mikro-organisme wat die Entner-Doudoro (2-keto-3-dioksi-6-fosfoglukonaat) pad ana erobies gebruik. Die lae-energieopbrengs van fermentasie in Z. mobilis is 'n gevolg van die gebruik van die Entner-Doudoro metaboliese pad, wat die helfte van die energie-opbrengs per mol substraat lewer, in vergelyking met die bekende Embden-Meyerhof-Parnas pad. Die werk wat in hierdie tesis aangebied word, vorm deel van 'n groter projek om glikolitiese regulering in verskillende mikro-organismes te vergelyk, naamlik Z. mobilis, Escherichia coli, Sac- charomyces en Lactococcus lactis. Hierdie organismes is gekies op grond van hul gebruik van verskillende glikolitiese meganismes. Kinetiese modellering is 'n handige metode om 'n vergelyking tussen hierdie organismes te trek. Hierdie werk fokus op die bou van 'n kinetiese model van die Entner-Doudoro glikolitiese metaboliese pad soos dit in Z. mobilis voorkom. Die doel van hierdie tesis was om so veel moontlik van die Entner-Doudoro ensieme onder standaard-toestande te karakteriseer. Die kinetiese parameters van elk van die ensieme is met behulp van ensimatiese essai's bepaal. Vir die meeste essai's is 96-put mikrotiterplate gebruik, maar nie al die karakteriserings kon met behulp van hierdie metode gedoen word nie, omdat sommige intermediate nie kommersieel beskikbaar was om gekoppelde essai's mee uit te voer nie. Kernmagnetiese resonansie (KMR) spektroskopie is gebruik om hierdie ensieme te karakteriseer. Die eksperimenteel bepaalde parameters is opgeneem in 'n wiskundige raamwerk. Tydsimulasies op die aanvanklike model was nie in staat om 'n bestendige toestand te bereik nie, omdat metaboliete opgebou het. 'n Sekond^ere model is gebou (met behulp van berekende maksimale aktiwiteite) wat ons toegelaat om teenstrydighede in die aanvanklike model te identi seer. Dit het getoon dat die eksperimenteel bepaalde maksimale aktiwiteite van drie ensieme in die laer gedeelte van glikolise te laag was, waarskynlik as gevolg van prote en denaturering tydens die ultrasoniese disintegrasieproses. 'n Finale model is gebou waarin 'n korreksiefaktor vir hierdie drie ensieme opgeneem is. Die modelle se voorspelde uitset (bestendige toestand konsentrasies en uksie) is vergelyk met waardes uit die literatuur of wat ons self bepaal het, as 'n metode om die model te valideer. Die model uitset was in goeie ooreenstemming met hierdie waardes. Die gedeeltelik gevalideerde kinetiese model is voorts gebruik as 'n analitiese instrument om beheer en regulering van glikolise in Z. mobilis te ondersoek. Die model wat in hierdie werk ontwikkel is, is ook vergelyk met die vorige gepubliseerde modelle. Ons model berus baie minder op waardes uit die wetenskaplike literatuur, en maak gebruik van parameters wat eksperimenteel bepaal is, onder identiese toestande. Die parameters van die gepubliseerde modelle is meesal verkry uit die literatuur, en in baie gevalle was die eksperimentele kondisies vir hierdie analises opgestel om die maksimale aktiwiteit te lewer onder nie- siologiese toestande. Verder bevat ons model minder parameters wat of uitgesluit is of wie se waardes aangeneem moes word. In toekomstige werk sal daar egter klem gel^e moet word op 'n minder wisselvallige ekstraksietegniek vir die verkryging van selekstrakte, om sodoende parameters te identi seer wat nie in hierdie werk bepaal kon word nie. Die gepubliseerde modelle sluit ook reaksies in wat nie ingesluit is in ons model nie (bv. ATP metabolisme). Hierdie sou in ag geneem moet word vir insluiting in 'n toekomstige uitgebreide model, om daarna te streef om 'n gedetailleerde kinetiese model van glikolise in Z. mobilis te bou.

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: This dissertation explores the behaviour and regulation of central carbon metabolism in Escherichia coli K12 W3110 under fermentative microaerobic conditions. To achieve this, an integrative systems modelling approach was adopted, which is introduced in Chapter 1 along with a review of metabolism in E. coli. An open-source software suite NMRPy, developed using the Python programming language, is presented in Chapter 2. NMRPy provides a host functions for basic processing, analysis and visualisation of Nuclear Magnetic Resonance (NMR) spectroscopy data. In addition to this, NMRPy offers specialised functions for the deconvolution of arrayed reaction time series, which proved indispensable to the research presented in this dissertation. NMRPy presents an easy to use, extensible tool for both routine and advanced use. In Chapter 3, a novel methodology is presented which was developed for the effective and comprehensive determination of enzyme kinetic parameters for systems biology using NMR. In contrast to traditional enzyme kinetic assay methods, this new methodology is less labour-intensive and yields significantly more information per experiment. By fitting kinetic equations to real time NMR data, dynamic changes in substrates, products and allosteric modifiers are quantified and allowed to inform the parameter fitting procedure. These data contain information on cooperative substrate binding, reversibility, product inhibition and allosteric effects. The proposed methodology is applied to the study of the first two enzymes of the glycolytic pathway. In Chapter 4, the construction, parameterisation and validation of a number of kinetic models of glycolysis in E. coli under microaerobic conditions is detailed. To model the lower half of glycolysis, a similar technique was adopted as in Chapter 3, in which models representing the reactions from triosephosphate isomerase to pyruvate kinase were parameterised by fitting them to a collection of 31P NMR reaction time series. This approach extends the methodology to enzyme sub-networks, yielding data that encompass the full complexity of the network regulatory interactions. The verified kinetic models were subjected to scrutiny, the results of which are presented in Chapter 5. The value of the modelling approach is demonstrated by the ease with which cumbersome in vivo experiments can be performed in silico. A structural analysis of the model topology was conducted, elucidating the elementary flux modes of fermentative glycolysis in E. coli, and identifying a futile cycle around PEP carboxylase and PEP carboxykinase. Model steady-state behaviour and control properties were explored in silico under various degrees of ATP demand and oxygen availability and a number of hypotheses are presented, explaining the regulation of free energy in E. coli, and the metabolic responses of E. coli to changing redox demands. Amongst other things, the results demonstrated that the glucose importing phosphoenolpyruvate: phosphotransferase pathway controlled glycolytic flux, and that under microaerobic conditions E. coli is able to regulate redox balance not only by balancing flux between acetate and ethanol, but also by altering the balance of flux between acetate and lactate at the pyruvate formate lyase/lactate dehydrogenase branch point. This study demonstrates the value of an integrated computational and experimental systems approach to exploring biological phenomena. / AFRIKAANSE OPSOMMING: In hierdie proefskrif word die gedrag en regulering van die sentrale koolstofmetabolisme in Escherichia coli K12 W3110 onder fermenterende mikro-a¨erobiese toestande ondersoek. Dit is moontlik gemaak deur ’n ge¨ıntegreerde stelsel-modelleringsbenadering, wat in Hoofstuk 1 bekendgestel word. D´ıe hoofstuk verskaf ook ’n oorsig van die metabolisme in E. coli. ’n Oopbron-kodepakket NMRPy, wat in die programmeringstaal Python ontwikkel is, word in Hoofstuk 2 beskryf. NMRPy verskaf ’n aantal funksies vir die basiese verwerking, analise en visualisering van Kern-Magnetiese Resonansie (KMR) spektroskopiese data, sowel as gespesialiseerde funksies vir die dekonvolusie van opeenvolgende reaksie-tydreekse. Hierdie funksionaliteit was onontbeerlik vir die verdere navorsing in hierdie proefskrif. Hoofstuk 3 beskryf die ontwikkeling van ’n nuwe metodiek vir die omvangryke bepaling van ensiem-kinetiese parameters vir sisteembiologie, deur van KMR gebruik te maak. In teenstelling tot tradisionele ensiem-kinetiese essai-metodes, is hierdie nuwe metodologie minder arbeidsintensief en lewer dit beduidend meer inligting per eksperiment. Deur die kinetiese vergelykings op tydsafhanklike KMR data te pas, word dinamiese veranderinge in substrate, produkte en allosteriese effektors gekwantifiseer en hierdie inligting gebruik in die passingsprosedure. Die data bevat inligting oor ko¨operatiewe substraatbinding, omkeerbaarheid, produkinhibisie en allosteriese effekte. Die voorgestelde metodologie word toegepas op die karakterisering van die eerste twee glikolitiese ensieme. In Hoofstuk 4 word die konstruksie, parameterisering en validering van ’n aantal kinetiese modelle van glikolise in E. coli onder mikro-a¨erobiese toestande uiteengesit. Die waarde van die modelleringsbenadering lˆe in die gemak waarmee omslagtige in vivo eksperimente in silico uitgevoer kan word. Om die onderste helfte van die glikolitiese pad te modelleer word ’n soortgelyke tegniek as in Hoofstuk 3 gebruik. Modelle van die reaksies vanaf triosefosfaat-isomerase tot by pirovaat-kinase is geparameteriseer deur dit op ’n versameling 31P KMR-tydreekse te pas. Hierdie benadering brei bostaande metodologie uit tot ensiem-subnetwerke en genereer data wat die volle kompleksiteit van regulerende interaksies in die netwerk insluit. Die geverifieerde modelle word in Hoofstuk 5 noukeurig ondersoek. ’n Strukturele analise van die modeltopologie word onderneem om die elementˆere fluksie-modes van fermentatiewe glikolise in E. coli te verklaar, sowel as om ’n futiele siklus rondom fosfo¨enolpirovaat karboksilase en fosfo¨enolpirovaat karboksikinase te identifiseer. Die bestendige-toestandsgedrag en kontrole-eienskappe word in silico ondersoek onder toestande van verskeie ATP beladings en suurstofbeskikbaarheid. ’n Aantal hipoteses word voorgelˆe, wat die regulering van vry energie in E. coli, sowel as die metaboliese reaksies van E. coli onder veranderende redoks-vereistes kan verklaar. Onder andere dui die resultate daarop dat die fosfo¨enolpirovaat:fosfotransferase sisteem (wat verantwoordelik is vir glukose-opname in die sel) die glikolitiese fluksie beheer en dat E. coli onder mikro-a¨erobiese toestande die redoksbalans nie net tussen asetaat en etanol kan reguleer nie, maar ook die deur wysiging van die fluksie-balans tussen asetaat en laktaat rondom die pirovaat-formiaat-liase/laktaatdehidrogenase vertakkingspunt. Hierdie studie toon die waarde van ’n ge¨ıntegreerde rekenaarmatige en eksperimentele sisteembenadering om biologiese verskynsels te ondersoek.