Les latérites couvrent plus de 30% des surfaces émergées. Comprendre leur évolution future est essentiel pour prédire les changements qui affecteront le fonctionnement de la Zone Critique dans les siècles à venir. La littérature a pu montrer que les horizons indurés de ces sols comportaient plusieurs générations de minéraux secondaires qui pouvaient être reliées à différents événements paléoclimatiques. Des études isotopiques ont montré que les horizons non indurés de ces sols pouvaient aussi posséder des générations distinctes de minéraux secondaires. La datation par Résonance Paramagnétique Électronique, a été appliquée sur des kaolinites de latérites dans le but d’identifier si ce minéral possède bien plusieurs générations au sein d’un même profil et si ces dernières étaient corrélables avec des événements paléoclimatiques globaux ou régionaux. Au cours de cette thèse, deux régolithes du Plateau du Karnataka (Inde), quatre du bassin du Rio Négro (Brésil) et un de la région de Syerstone (Australie) ont été étudiés. Les résultats de ces datations ont permis de distinguer différentes générations de kaolinites au sein de chacun des profils étudiés. Ces dernières sont par ailleurs toutes corrélables avec des périodes de fortes précipitations liées aux évolutions climatiques régionales et/ou globales. Ces résultats confirment aussi l’existence du phénomène de rajeunissement profils latéritiques au cours des temps géologiques. Le travail de cette thèse corrobore l’hypothèse de l’évolution épisodique de ces sols tropicaux et souligne leur importante stabilité dans les temps géologiques. / Laterites are covering a third of the continental surfaces. Understanding their evolution through the ongoing climate change is critical to predict the future Critical Zone transformation. Laterite duricrusts can contain several generations of secondary minerals that can be correlated with distinct paleoclimatic events. Isotopic studies showed that loose laterites may also be formed by several generations of secondary minerals. The objective of this PhD was to confirm that loose laterites contain several generations of kaolinites and that they formed during regional and/or global paleoclimatic events. With this aim in view, a recent dating methodology, the Electron Paramagnetic Resonance dating of kaolinites, was used on tropical regolith profiles from three continents with distinct geodynamic histories: India (Karnataka plateau), South America (Rio Negro basin, Brazil) and Australia (Syerstone region, New South Wale). The data set of this PhD confirms that a regolith contains several generations of kaolinites. All the identified generations of the present study can be correlated with regional paleoclimatic events or with tropical weathering favourable periods initiated by global paleoclimatic trends. Kaolinite generations formed preferentially during high precipitations periods and result from the rejuvenation of laterite profiles. The results of this PhD confirm that laterites have an episodic evolution.

Les catalases, les peroxydases et les catalase-peroxydases sont des hémoprotéines, dont l'activité enzymatique essentielle est la régulation de la concentration cellulaire du peroxyde d'hydrogène. Au cours de ce travail de thèse, nous avons mis en évidence et caractérisé, dans ces protéines, des intermédiaires réactionnels issus de l'oxydation de tyrosines et/ou de tryptophanes, alternatifs au composé I, [Fe(IV)=O Por•+]. La prise en compte de la formation de tels radicaux protéiques nous a amené à réexaminer le mécanisme des catalases et des peroxydases, jusque là centré sur l'oxydo-réduction de l'hème. Pour déterminer les structures électroniques et la réactivité des intermédiaires réactionnels, nous avons utilisé une approche multidisciplinaire combinant les spectroscopies de Résonance Paramagnétique Electronique (RPE) multifréquences (9-285 GHz), d'absorption électronique UV-visible par stopped-flow, ainsi que la mutagenèse dirigée et le marquage isotopique. L'étude comparative de huit catalases, provenant de souches bactériennes et de mammifères, nous a permis de démontrer que la température de réaction, le pH et l'excès d'oxydant favorisent la formation de l'intermédiaire [Fe(IV)=O Tyr•], résultant du transfert d'électron intramoléculaire entre la tyrosine et la porphyrine dans les catalases de foie de boeuf, d'érythrocytes humains, de B. abortus et de M. lysodeikticus. Nos études ont montré que l'intermédiaire [Fe(IV)=O Por•+] est prédominant à T ≥ 20°C. Ceci explique pourquoi les études cinétiques de la littérature, pour lesquelles seule l'absorption UV-visible à température ambiante a été utilisée, n'avaient pas permis de mettre en évidence la formation du radical tyrosyle. Nos résultats expliqueraient aussi les différences d'efficacité de dismutation du peroxyde d'hydrogène de ces catalases. En effet, le transfert spontané d'électron entre la tyrosine et la porphyrine, favorisé dans certaines catalases, serait en compétition avec la réaction entre l'intermédiaire [Fe(IV)=O Por•+] et le peroxyde d'hydrogène. Dans le cadre d'études spécifiques sur les catalase-peroxydases (KatGs) et en utilisant la catalase-peroxydase de B. pseudomallei comme cas d'étude, nous avons pu mettre en évidence la formation d'un radical tyrosyle et de deux intermédiaires tryptophanyles, dont l'un est formé sur le Trp330, équivalent au site radicalaire de la cytochrome c peroxydase. Nous avons démontré que l'intermédiaire [Fe(IV)=O Por•+] réagit avec l'ABTS et que le [Fe(IV)=O Trp330 •+] n'est pas l'intermédiaire réactif vis-à-vis de l'isoniazide (antibiotique utilisé dans le traitement de la tuberculose). Notre étude comparative de sept catalaseperoxydases a démontré que le site de formation du radical tryptophanyle est unique pour chaque enzyme, mais peut différer entre deux protéines, ce qui expliquerait en partie leurs différences de réactivité vis-à-vis de l'ABTS et de l'isoniazide. Nous avons montré que dans les catalase-peroxydases, les radicaux protéiques peuvent jouer le rôle de cofacteurs dans l'oxydation des substrats et que le mécanisme de la réaction de dismutation du peroxyde d'hydrogène diffère de celui des catalases monofonctionnelles. Afin de comprendre les propriétés physico-chimiques permettant et favorisant la formation des radicaux protéiques, en tant qu'intermédiaires alternatifs à l'intermédiaire [Fe(IV)=O Por•+], nous avons caractérisé une peroxydase de raifort reconstituée, dans laquelle l'hème à été remplacé par un hème lié par liaison covalente à une tyrosine ou à un tryptophane.

M.S. / Biochemistry / The EPR spectrum of the Tyr-Cys· free radical in oxidized apoGAOX has been investigated, using a combination of approaches. Power saturation analysis has been used to resolve two unique spectra through Evolving Factor Analysis (EFA) global fitting, indicating the presence of two distinct free radical species in the sample. The component that dominates at low microwave power arises from the Tyr-Cys· side chain, while the high power component has not yet been assigned. The experimental results show that the EPR spectrum collected at low power includes approximately 7% of the high power component. EPR spectra have been collected for ten different isotope derivatives of GAOX, including ²H-labeled, ¹³C-labeled, 17[superscript]O-labeled, and ³³S-labeled forms. XSophe simulation of the EPR spectra has been performed for the isotopically labeled samples in order to determine the spectroscopic parameters - g-values, hyperfine coupling constants, and linewidths. The g-values and the methylene proton hyperfine coupling constants obtained for the isotopically labeled samples are consistent with the literature values. The magnitude of the hyperfine coupling constants associated with each of the nuclei confirms that significant electron spin density is found on the methylene protons, the alternating carbon atoms within the aromatic π system and the 2p[subscript]z orbital of both sulfur and oxygen. Moreover, the rotation angle of the methylene protons to the phenoxyl ring around the C1-C7 bond has been evaluated based on the experimentally defined hyperfine coupling constants of the two methylene protons.

The structure and function of RNA molecules are dependent on RNA-metal ion interactions in both diffusive and direct ways. Structural information for RNA has been obtained using various biophysical and biochemical methods. In this study, using site-directed spin labeling (SDSL) and EPR spectroscopy, distances in RNA duplexes, TAR RNA, and the hammerhead ribozyme have been measured to investigate RNA structures. Kinetic measurements have been performed in the extended hammerhead ribozyme to correlate the catalytic function with metal dependent ribozyme folding. As a basic model system for distance measurements, inter-spin distances in RNA duplexes with spin labels at various positions are measured using SDSL with continuous EPR and a Fourier deconvolution method. Divalent metal-ion dependent TAR RNA folding from bent to extended conformers is monitored by measuring inter-spin distances near the bulge region. In order to investigate a proposed loop-loop interaction in the extended hammerhead ribozyme which significantly enhances the ribozyme activity, distance measurements, dynamics studies, and kinetics measurements have been performed. We have introduced PELDOR long-distance measurements in order to investigate metal dependent folding of the hammerhead ribozyme. The dynamics of the spin labels attached to the hammerhead ribozyme with increasing mono- and divalent metal ion concentrations are monitored using CW EPR spectroscopy at room temperature. EPR data show that a loop-loop interaction occurs near the U1.6 nucleotide, and that in 0.1 M NaCl the docking occurs at submillimolar Mg2+ concentrations ([Mg2+]1/2, docking = ~ 0.7 mM). Kinetics measurements show that the hammerhead ribozyme requires high concentration of Mg2+ for the maximum cleavage activity ([Mg2+]1/2, cleavage = ~ 90 mM).

Photosystem II is unique! It remains the only enzyme that can oxidize water using light as energy input. Water oxidation in photosystem II is catalyzed by the CaMn4 cluster. The electrons extracted from the CaMn4 cluster are transferred to P680+ via the tyrosine residue D1-Tyr161 (YZ). Favorable oxidation of YZ is coupled to a proton transfer along a hydrogen bond to the nearby D1-His190 residue, resulting in the neutral radical YZ•. By illuminating photosystem II at cryogenic temperatures, YZ• can be trapped in a stable state. Magnetic interaction between this radical and the CaMn4 cluster gives rise to a split electron paramagnetic resonance (EPR) signal with characteristics that depend on the oxidation state (S state) of the cluster. The mechanism by which the split EPR signals are formed is different depending on the S state. In the S0 and S1 states, split signal induction proceeds via a P680+-centered mechanism, whereas in the S2 and S3 states, our results show that split induction stems from a Mn-centered mechanism. This S state-dependent pattern of split EPR signal induction can be correlated to the charge of the CaMn4 cluster in the S state in question and has prompted us to propose a general model for the induction mechanism across the different S states. At the heart of this model is the stability or otherwise of the YZ•–(D1-His190)+ pair during cryogenic illumination. The model is closely related to the sequence of electron and proton transfers from the cluster during the S cycle. Furthermore, the important hydrogen bond between YZ and D1-His190 has been investigated by following the split EPR signal formation in the different S states as a function of pH. All split EPR signals investigated decrease in intensity with a pKa of ~4-5. This pKa can be correlated to a titration event that disrupts the essential hydrogen bond, possibly by a direct protonation of D1-His190.  This has important consequences for the function of the CaMn4 cluster as this critical YZ–D1-His190 hydrogen bond steers a multitude of reactions at the cluster.

Electron paramagnetic resonance (EPR) dosimetry using the well established dosimeter material alanine is a generally accepted dosimetric method for measurements of high absorbed doses. Alanine EPR dosimetry is however not sensitive enough for high precision measurements of low (< 5 Gy) absorbed doses using reasonably measurement times and small dosimeters. It has therefore not been possible to fully exploit the benefits of EPR dosimetry for applications in radiation therapy. The aim of this thesis was to show that sensitive EPR dosimetry is a competitive method for applications in radiation therapy fulfilling the requirements of measurement precision. Our strategy for reaching this goal was to search for new, more sensitive, EPR dosimeter materials fulfilling the criteria of being tissue equivalent, having a high radical yield and having a narrow EPR spectrum suitable for dosimetry. The best materials were found among formates and dithionates. Doping with small amounts of metal ions and recrystallisation in D2O were tested to further increase the sensitivity. Four promising candidate materials were tested regarding radical stability and dose response and among them lithium formate was chosen for dosimetry in radiation therapy applications. A high precision EPR dosimetry method was developed using lithium formate. The method included the development of a production method for EPR dosimeters with very homogenous shape, mass and composition. A read-out process was developed with maximal measurement precision for reasonably short measurement times. The method also included a dosimeter quality control before actual dose measurements. Measurement accuracy was controlled for every new dosimeter batch. This high precision lithium formate EPR dosimetry method was evaluated for pretreatment verifications of intensity modulated radiation therapy (IMRT) treatment plans. The precision and accuracy was shown to be sufficient (< 5 %) for measurements of doses above 1.5 Gy using one single dosimeter and a measurement time of 15 minutes. The described evaluation is therefore a demonstration of the improved precision at low dose determinations that is available with our sensitive EPR dosimeter materials. While the EPR signal intensity is proportional to absorbed dose, the signal shape is in some cases dependent on the radiation quality. A new method is presented for simultaneous measurements of beam LET (linear energy transfer) and absorbed dose in heavy charged particle beams using potassium dithionate EPR dosimetry. The study shows that when irradiating a dosimeter with 35 MeV carbon ions, the ratio of the signal amplitudes from two radicals in potassium dithionate vary along the track indicating a dependence on linear energy transfer, LET. Potassium dithionate may therefore be a promising EPR dosimeter material for simultaneous measurements of absorbed dose and LET in heavy charged particle radiation fields.

Les métallatranes (M=Si, Ge, Sn) ont fait l'objet d'un intérêt important de la communauté scientifique durant les 50 dernières années et ont des applications dans de nombreux domaines, de la physique à la biologie en passant par la médecine, etc... Cependant, leurs propriétés électrochimiques n'ont été étudiées que de façon très incomplète à ce jour. Dans le travail présenté ici, nous avons réalisé la synthèse de nouveaux métallatranes et les avons caractérisés, et ce par plusieurs approches et techniques différentes (synthèse organique, voltammétrie cyclique, électrolyse préparative, EPR, spectroélectrochimie, voltammétrie cyclique couplée à un EPR, RMN, GC-MS). Plusieurs nouveaux métallatranes ont été obtenus, et leurs structures validées par diffraction RX. Les mesures électrochimiques montrent que l'électrooxidation des métallatranes a lieu via des transferts électroniques réversibles, et les cations-radicaux (CRs) qui résultent de leur oxydation ont été détectés par voltammétrie cyclique, spectroscopie UV-visible, et spectroscopie EPR en temps réel. La délocalisation de spin dans ces espèces paramagnétiques a été caractérisée et soutenue par des calculs de DFT. Nous avons concentré nos efforts sur les propriétés électroniques des métallatranes et la nature des diverses liaisons entre l'atome de silicium penta-coordonné des silatranes d'une part, l'atome d'azote de la structure atrane, et les atomes de différents substituants d'autre part. Nous avons également étudié le problème posé par les sels du fond pour l'électrochimie, qui doivent être non-coordinants et non-nucléophiles, et doivent fournir une bonne conductivité ionique dans des solvants peu polaires. Ils permettent ainsi de mener des expériences de voltammétrie et de la RPE sur des espèces électrogénérées instables, notamment électrophiles ou des radicaux libres.

Iron is an essential element for life. It is involved in a number of biological processes, including iron sulfur (Fe/S) cluster assembly and heme biosynthesis. However it is also potentially toxic due to its ability to induce formation of reactive oxygen species (ROS) via Fenton chemistry. Therefore its uptake, trafficking and utilization must be regulated to avoid its toxicological effect. It has been recently discovered that Fe/S cluster biosynthesis machinery plays a key role in the cellular iron regulation and its disruption leads to impaired iron regulation and iron accumulation within mitochondria. The iron accumulation resulted from impaired Fe/S cluster assembly in the eukaryotic model organism Saccharomyces cerevisiae (baker’s yeast) was studied. Various biophysical (e.g. Mössbauer, EPR, UV-vis spectroscopy) and biochemical (e.g. Western blots, PCR, enzyme activity assay, etc.) techniques were used to characterize the iron content in yeast mitochondria isolated from several mutants strains. In these mutants one of the proteins involved in Fe/S cluster biosynthesis (Yah1p and Atm1p) is mutated and iron regulation and metabolism are disrupted. By integrating the results obtained from these different methods, it was determined that excess iron accumulates in the mutant mitochondria as inorganic phosphate Fe(III) nano-particles exhibiting superparamagnetic behaviors. Oxygen is required for iron accumulation and nanoparticle formation. The Fe(III) nano-particles can be chemically reduced to Fe(II) then largely exported from the mitochondria. These biophysical and biochemical methods were also used to examine the iron distribution in whole yeast cells of the Aft1-1up strain in which iron regulon genes are constitutively activated and compared to that of Yah1p-depleted and wild type yeast. Constitutive activation of iron regulon genes does not alter the cellular iron distribution significantly. However disruption of Fe/S cluster assembly by Yah1p depletion causes dramatic cellular iron redistribution: the vacuolar iron is largely evacuated and most of the cellular iron probably precipitates in mitochondria as Fe(III) nanoparticles. The results provide novel insights into iron trafficking and possible signal communications between organelles within cells.

Poly(N-isopropylacrylamide) (PNIPAM) is a smart polymer exhibiting an inverse temperature-solubility relationship with a sharp transition at 32&deg / C in its aqueous solution. Due to its reversible thermo-responsive phase transition behavior at around body temperature, PNIPAM promise a potential for a variety of novel applications especially in biotechnology and medicine. PNIPAM can be produced by conventional polymerization methods, as well as by use of ionizing radiation, primarily by gamma which leads mainly to a residual-free crosslinked polymer. In this study, RF plasma (glow discharge) technique is used as a novel synthesis method in solid state leading to higher proportions of linear polymer. Since plasma method is an additive-/initiator-free process, a residual-free polymer is expected. To obtain a better understanding of the plasma induced solid state polymerization mechanism of NIPAM, X-ray data are used. It is found that crystalline structures of Acrylamide (AAm) and NIPAM are isomorphous. Plasma and post plasma aging effects on crystalline structure of NIPAM are followed. From the Electron Paramagnetic Resonance (EPR) investigations it is observed that post plasma polymerization of NIPAM in solid state proceed by radicalic mechanism. After determination of temperature range in which the radical formed by plasma treatment of NIPAM is highly stable, decay kinetics of the propagating radical in solid state after plasma treatment has been studied in detail.

This dissertation focuses on the spectroscopic and thermodynamic characterization of the unusual metal-RNA cluster found in the P5abc subdomain of the Tetrahymena group I intron. The P5abc subdomain is a part of the P4-P6 domain found in the Tetrahymena thermophila group I intron selfsplicing RNA. From both X-ray crystal structures of the P4-P6 domain, a remarkable cluster of Mg2+ or Mn2+ ions was found in the P5abc subdomain (Cate et al. 1996; Juneau et al. 2001). It is believed that the metal ion core in the P5abc subdomain stabilizes the active conformation of the RNA (Cate et al. 1996). An understanding of the role of these metal ions in facilitating the correct structure of the P5abc subdomain provides insight into how metal ions help overcome the folding barriers of complex RNA structures. Under solution conditions, the properties of this uncommon metal ion core and its influence on the truncated P5abc subdomain structure have been investigated. Both EPR spectroscopy and thermal denaturation experiments have been employed to search for a spectroscopic signature of metal ion core formation and also determine the thermodynamic contribution of the metal ion core on the stability of the folded P5abc structure. A spectroscopic signature of metal ion core formation was assigned for the P5abc subdomain by EPR microwave power saturation studies. Power saturation studies of the P5abc subdomain, P4-P6 domain and corresponding mutants reveal that the addition of 5 equivalents of Mn2+ are required for the wild type P5abc subdomain to form the metal ion core under solution conditions in 0.1 M NaCl. Results from both domain and subdomain microwave power saturation studies suggest that this technique can be applied for detecting clustering of Mn2+ ions in other RNA structures. The thermodynamic consequence of this metal ion core was probed by thermal denaturation techniques including UV-Vis spectroscopy and differential scanning calorimetry (DSC). DSC experiments were utilized to directly determine the thermodynamic contribution of the metal ion core. This value was determined to be an average of ∆∆G of -5.3 kcal/mol and is consistent with ∆∆G values obtained for other RNA tertiary structures.