The emergence of CTX-M class-A extended-spectrum β-lactamases, which confer resistance to second and third-generation cephalosporins, poses a serious health threat to the public. CTX-M β-lactamases use a catalytic serine to hydrolyze the β-lactam ring. Specifically, the hydrolysis reaction catalyzed by CTX-M β-lactamase proceeds through a pre-covalent complex, a high-energy tetrahedral acylation intermediate, a low-energy acyl-enzyme complex, a high-energy tetrahedral deacylation intermediate after attack via a catalytic water, and lastly, the hydrolyzed β-lactam ring product which is released from the enzyme complex. The crystallographic structure of CTX-M at sub-angstrom resolution has enabled us to study enzyme catalysis as well as perform computational molecular docking in our efforts to develop new inhibitors against CTX-M. The goal of this project was to determine the hydrogen bonding network and proton transfer process at different stages of the reaction pathway as well as develop novel inhibitors against CTX-M β-lactamases. The results I have obtained from the project have elucidated the catalytic mechanism of CTX-M β-lactamase in unprecedented detail and facilitated the development of novel inhibitors for antibiotic drug discovery. The first aim of the project focused on developing high affinity inhibitors against class A β-lactamase using a structure-based drug discovery approach, which ultimately led to the identification of CTX-M9 inhibitors with nanomolar affinity. Compound design was based on the initial use of computational molecular docking results along with x-ray crystal structures with known inhibitors bound in the active site. In addition, chemical synthesis was used to build and extend the existing inhibitor scaffold to improve affinity to CTX-M9 and related serine β-lactamases. Through a fragment-based screening approach, we recently identified a novel non-covalent tetrazole-containing inhibitor of CTX-M. Structure-based design was used to improve the potency of the original tetrazole lead compound more than 200-fold with the use of small, targeted structural modifications. A series of compounds were used to probe specific binding hotspots present in CTX-M. The designed compounds represent the first nM-affinity non-covalent inhibitors of a class A β-lactamase. The complex structures of these potent compounds have been solved using high resolution x-ray crystallography at ~ 1.2-1.4 Å, which provides valuable insight about ligand binding and future inhibitor design against class A β-lactamases. Specifically, the first aim of the project was to use ultra-high resolution x-ray crystallography to study β-lactamase catalysis. Through the use of ultra-high resolution x-ray crystallography with non-covalent and covalent inhibitors, I was able to structurally characterize the critical stages of the enzyme mechanism. Here we report a series of ultra-high resolution x-ray crystallographic structures that reveal the proton transfer process for the early stages of the class A β-lactamase catalytic mechanism. The structures obtained include an a 0.89 Å crystal structure of CTX-M β-lactamase in complex with a recently-developed 89 nM non-covalent inhibitor, and a 0.80 Å structure in complex with an acylation transition state boronic acid inhibitor. Nearly all the hydrogen atoms in the active site, including those on the ligand, polar protein side chains and catalytic water, can be identified in the unbiased difference electron density map. Most surprisingly, compared with a previously determined 0.88 Å apo structure determined under the same conditions, the hydrogen-bonding network has undergone a series of reshuffling upon the binding of the non-covalent ligand. Two key catalytic residues, Lys73 and Glu166, appear to have both changed from a charged state to being neutral. Interestingly, structural evidence suggests the presence of a low barrier hydrogen bond (LBHB) shared between Lys73 and Ser70. These unprecedented detailed snapshots offer direct evidence that ligand binding can alter the pKa's of polar protein side chains and their affinities for protons. Such effects can be a common mechanism utilized by enzymes to facilitate the proton transfer process of a reaction pathway. They also have important implications for computational modeling of protein-ligand interactions. Ultra-high resolution x-ray crystallography allowed us to determine the hydrogen atom positions for key active site residues involved in catalysis. As a result, the ability to characterize the hydrogen bonding network led to the determination of the specific proton transfer process that occurs during the reaction stages of the CTX-M β-lactamase mechanism. Overall, the results from this project demonstrate the effectiveness of using ultra high resolution x-ray crystallography as a useful tool to study enzyme catalysis as well as develop and discover novel inhibitors.

In Gram-negative bacteria, the three-layered cell envelope, including the cell wall, outer and inner membranes, is essential for cell survival in the changing, and often hostile environments. Conserved in all prokaryotes, the cell wall is incredibly thin, yet it functions to prevent osmotic lysis in diluted conditions. Based on observations obtained by genetic and chemical perturbations, time-lapse live cell imaging, quantitative imaging and statistical analysis, Part I of this dissertation explores the molecular and physical events leading to cell lysis induced by division-specific beta-lactams. We found that such lysis requires the complete assembly of all essential components of the cell division apparatus and the subsequent recruitment of hydrolytic amidases. We propose that division-specific beta-lactams lyze cells by inhibiting FtsI (PBP3) without perturbing the normal assembly of the cell division machinery and the consequent activation of cell wall hydrolases. On the other hand, we demonstrated that cell lysis by beta-lactams proceeds through four physical phases: elongation, bulge formation, bulge stagnation and lysis. Bulge formation dynamics is determined by the specific perturbation of the cell wall and outer membrane plays an independent role in stabilizing the bulge once it is formed. The stabilized bulge delays lysis, and allows escape and recovery upon drug removal. Asymmetrical in structure and unique to Gram-negative bacteria, outer membrane prevents the passage of many hydrophobic, toxic compounds. Together with inner membrane and the cell wall, three layers of the Gram-negative cell envelope must be well coordinated throughout the cell cycle to allow elongation and division. Part II of this dissertation explores the essentiality of the LPS layer, the outer leaflet of the outer membrane. Using a conditional mutant severely defective in LPS transport, we found that mutations in the initiation phase of fatty acid synthesis suppress cells defective in LPS transport. The suppressor cells are remarkably small with a 70% reduction in cell volume and a 50 % reduction in growth rate. They are also blind to nutrient excess with respect to cell size control. We propose a model where fatty acid synthesis regulates cell size in response to nutrient availability, thereby influencing growth rate. / Chemistry and Chemical Biology

In critically ill patients, the pathophysiology of sepsis can affect the interactions between the antibiotic, the bacteria and the patient, leading to potential therapeutic failure and the development of antibiotic resistance. It is well acknowledged that research that optimises antibiotic exposure will assist improvement of outcomes in this patient group. Although beta-lactam antibiotics, such as piperacillin and meropenem, are commonly selected for empiric therapy of sepsis, dosing is unlikely to be optimal. In patients without renal dysfunction, data suggests that disease-induced alterations to pharmacokinetic parameters result in low trough concentrations for significant periods within a dosing interval. Administration of these time-dependent antibiotics by continuous infusion has been suggested to improve the pharmacokinetic-pharmacodynamic profile. Knowledge of concentrations in the extracellular fluid of human tissue, which is the target site of most pathogens, is particularly instructive. Extracellular fluid concentrations can be determined using techniques such as microdialysis. Therefore, the principal aims of this Thesis were to determine the plasma and subcutaneous tissue pharmacokinetics of piperacillin and meropenem administered by bolus dosing and continuous infusion in critically ill patients with sepsis; and to use Monte Carlo simulations to compare the ability of different dosing strategies to achieve pharmacodynamic endpoints. This Thesis also sought to compare the clinical outcomes of bolus dosing and continuous infusion of a beta-lactam antibiotic, ceftriaxone, in a prospective randomised controlled trial and to perform a meta-analysis on clinical outcomes from other similar published studies. Finally, this Thesis aimed to systematically review the published literature to determine any correlation between antibiotic dosing and the development of antibiotic resistance. The results of the pharmacokinetic studies, using piperacillin and meropenem, indicate that beta-lactam distribution into subcutaneous tissue, in critically ill patients with sepsis, is less than that observed in previous studies in healthy volunteers yet superior to studies in patients with septic shock. This supports the notion that the peripheral concentration of drugs may be inversely related to the level of sickness severity. Administration by continuous infusion was found to maintain statistically significantly higher trough beta-lactam concentrations in both plasma and subcutaneous tissue. Further analysis of the plasma data using population pharmacokinetic modeling and Monte Carlo simulations described significant pharmacodynamic advantages for administering meropenem or piperacillin by continuous infusion to organisms with high minimum inhibitory concentrations. Given the documented pharmacodynamic advantages for administering beta-lactams by continuous infusion, a prospective randomized controlled clinical trial was conducted using the beta-lactam antibiotic ceftriaxone. In 57 critically ill patients, we found equivalence between continuous infusion and bolus dosing in the intention-to-treat analysis. However, our a priori analysis criteria, requiring patients receive at least 4-days antibiotic treatment, found significant clinical and bacteriological advantages for administration by continuous infusion. To further investigate any clinical differences between bolus dosing and continuous infusion of beta-lactam antibiotics, we performed a meta-analysis of all published studies. Our analysis of the 13 published prospective randomized controlled trials (846 hospitalised patients) showed equivalence of continuous infusion and bolus dosing. Possible confounders observed within, and between the studies, make interpretation of these results challenging. However, two large retrospective cohorts not included in the meta-analysis, found definitive clinical and bacteriological advantages suggesting further research may be appropriate. The possible relationship between antibiotic dosing, or exposures, on the development of resistance was investigated using a structured review of the published literature. Our analysis of relevant papers found a wealth of data describing increasing levels of resistance with sub-optimal antibiotic dosing, particularly for fluoroquinolone antibiotics, but also for other classes including beta-lactams. These results demonstrate the importance of optimizing antibiotic dosing to decrease the development of antibiotic susceptibility from sub-optimal dosing, particularly for critically ill patients who are likely to have low drug concentrations. The results of this Thesis, suggest that a large, prospective, multi-centre randomised controlled trial in critically ill patients with sepsis, is required to definitively determine the clinical utility of administration of beta-lactam antibiotics by continuous infusion.

Orientador: Francisco Carlos Groppo, Pedro Luiz Rosalen / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Odontologia de Piracicaba / Made available in DSpace on 2018-08-04T02:16:09Z (GMT). No. of bitstreams: 1 Baglie_Sinvaldo_D.pdf: 818069 bytes, checksum: 1c0d1eb4621ca27afda369d7e188a9a7 (MD5) Previous issue date: 2005 / Resumo: As concentrações de amoxicilina (AMO)em plasma e saliva e seus efeitos na microbiota oral de voluntários sadios foram analisados, em um estudo aberto, de forma aleatória, cruzado com dois períodos. AMO875 mg (Amoxicilina-EMS- medicamento teste e Amoxil@- medicamentoreferência) foi administrada para todos os voluntários com 1 semana de intervalo entre as doses. Amostras de plasma e saliva foram colhidas em um intervalo de até 12h após a administração. As concentrações plasmáticas de AMOforam obtidas por cromatografia líquida e espectrometria de massas com ionização por electrospray (LC-ESI-MS). As concentrações de amoxicilina em saliva foram analisadas por cromatografia líquida de alta eficiência (HPLC).As contagens dos microrganismos totais, anaeróbios e estreptococos foram obtidas em diferentes condições de cultura. ASC foi calculada pelo método de extrapolação dos trapezóides. Cmaxe Tmaxforam compilados dos dados de concentração-tempo. Análise de variância foi realizada usando dados logaritmicamente transformadospara ASCo-inf, ASCO-12h, Cmax e não transformados para Tmax.As médias (:1:dp) para ASCO-12h (jJg.h.mL-1),ASCo-inf (jJg.h.mL-1),Cmax (jJg.mL-1)and Tmax(h), foram respectivamente: 55,42(:1:16,85), 55,42 (:1:16,85), 18,59 (:1:6,3), 2,04(:1:0,75) para o medicamento teste e 51,11 (:1:18,9), 51,29 (:1:19,12), 17,83 (:1:5,86),2,02 (:i:O,87)para o medicamento referência. Intervalos de confiança (90%) para a razão das médias de AMO(testejreferência) para ASCO-12he Cmaxforam: 0,961-1,149 e 0,914 - 1,142, respectivamente. Tanto para as concentrações plasmáticas quanto para as salivares não foram observadas diferenças estatisticamente significantes entre os dois medicamentos (Teste t pareado, p>0.05). As contagens de microrganismos provindos da saliva não mostraram diferenças estatísticas (ANOVAde medidas repetidas, p>0.05) entre as duas formulações durante cada tempo. A partir de 60 min, uma redução estatisticamente significante (ANOVAde medidas repetidas, p<0.05) foi observada para todos os microrganismos. Os dois medicamentos foram considerados bioequivalentes baseado na razão e extensão de absorção e foram eficazes na redução dos microrganismos da microbiota oral avaliados até 12 horas / Abstract: The plasmatic and salivary concentration of amoxicillin(AMO)875mg and its effects on oral microbiota in healthy volunteers were assessed in an open, randomized, two-period crossover designo A single 875 mg oral dose of AMO (Amoxicillin-EMS- test formulation and Amoxil@- reference formulation) was administered to ali volunteers observing 1-week intervaI between doses. Blood and saliva samples were collected from pre-dose to 12 h. The concentrations of AMO, in plasma and saliva, were quantified by LC-ESI-MSand LCmethod, respectively. AUC was calculated by the trapezoidal rule extrapolation method. Cmaxand Tmax were compiled from the plasmatic concentration-time data. Streptococci, anaerobe, and total microorganisms' counts were obtained in different culture conditions. The mean values (:1:SD) for AUCO-12(lh1g.h.mL-1),AUCO-í(nlf1g.h.mL-1),Cmax(l1g.mL-1) and Tmax(h), were respectively: 55.42(:1:16.85), 55.42 (:I:16.85), 18.59 (:I:6.3), 2.04{:I:0.75) concerning the test formulation and 51.11 (:I:18.9), 51.29 (:I:1,9.12), 17.83 (:I:5.86), 2.02 (:1:0.87) concerning the reference formulation. Confidence intervals (90%) of amoxicillinmeans of AUCO-12ahnd Cmaxratios (testjreference) were: 0.961 - 1.149 and 0.914 - 1.142, respectively. No statistically significant differences were observed between the two formulations (t test, p>0.05) regarding AMOplasmatic and saliva concentrations. Saliva microorganisms counts did not show statistical differences (ANOVArepeated measures, p>0.05) between the two groups during each sampling time. Starting at 60 min, a statistically significant decrease (ANOYArepeated measures, p<0.05) was also observed for all microorganisms. Both formulations were bioequivalent based on both the rate and extent of absorption, and were effective in reduce microorganisms until 12 hours / Doutorado / Farmacologia, Anestesiologia e Terapeutica / Doutor em Odontologia

<p>Antibiotic treatment causes an ecological disturbance on the human microflora. Four commensal bacteria: E. coli, enterococci, a-streptococci and coagulase-negative staphylococci, from patients with extensive, high antibiotic usage were investigated with regard to resistance pattern and mutation frequency. Among 193 investigated strains it was found that high antibiotic usage selected for resistant bacteria and enriched for bacteria with a small but significantly increased mutation frequency. </p><p>The relative biological fitness cost of resistance in <i>Staphylococcus epidermidis</i> was assessed in a human in vivo model where the indigenous flora was present. In vitro data of the bacterial growth rate correlated well to in vivo fitness assayed in the competition experiments on skin. </p><p>An in vitro kinetic model was shown to be a useful tool to establish the pharmacokinetic and pharmacodynamic (PK/PD) indices for efficacy of antibiotics. It was confirmed that the time, when the concentration exceeds the minimal inhibitory concentration (MIC), correlates with efficacy for b-lactam antibiotics. To achieve maximal killing for penicillin-resistant pneumococci, with an MIC of 2 mg/L, the peak concentration was also of importance. </p><p>Suboptimal dosing regimen facilitates selection of resistance. Penicillin-resistant pneumococci were easily selected in a mixed population with penicillin-sensitive, -intermediate and -resistant pneumococci in an in vitro kinetic model. The selection of the resistant strain was prevented when the benzylpenicillin concentration exceeded the MIC for approximately 50% of 24 h.</p>

<p>Toxic products, such as endotoxin from the gram-negative and exotoxin from the gram-positive bacteria, are the most important initiators of the inflammatory host response in sepsis. In addition to antibacterial treatment, numerous attempts have been made to interfere with the exaggerated proinflammatory cascade initiated by the toxins. As most antitoxic and anti-inflammatory agents have shown no clear efficacy, an attractive alternative has been to prevent or minimise their release. Therefore, it was of interest to further study the antibiotic-induced release of toxins after exposure to antibiotics used for the treatment of the most severe infections, especially if protein synthesis inhibitors could reduce the release induced by PBP 3-specific β-lactam antibiotics.</p><p>There were significant reductions in endotoxin release from gram-negative bacteria when the combination of the PBP 3-specific β-lactam antibiotic, cefuroxime, and the protein synthesis inhibitor, tobramycin, was compared with cefuroxime alone. Increasing doses of tobramycin reduced endotoxin release and increased the killing rate. In a kinetic <i>in vitro</i> model the endotoxin release from <i>E.coli</i> was higher after the second dose of cefuroxime. Nevertheless, it was reduced after addition of tobramycin.</p><p>No binding of tobramycin to endotoxin was observed, either <i>in vivo</i> or <i>in vitro</i>. In a porcine sepsis model, a possible anti-inflammatory effect of ceftazidime and tobramycin, expressed as late cytokine inhibition, was seen.</p><p>The protein synthesis inhibitor, clindamycin, released less streptococcal pyrogenic exotoxin A (SpeA) from a group A streptococcus strain than penicillin, and addition of clindamycin to penicillin resulted in less toxin production than penicillin alone. The SpeA production was dependent on the bacterial number at the start of treatment. Higher doses of penicillin also led to less SpeA. </p><p>The choice of antibiotic class and dose may be important in the severely ill septic patient in whom an additional toxin release could be deleterious. A combination of a β-lactam antibiotic and a protein synthesis inhibitor seems beneficial but further investigations are needed.</p>

Toxic products, such as endotoxin from the gram-negative and exotoxin from the gram-positive bacteria, are the most important initiators of the inflammatory host response in sepsis. In addition to antibacterial treatment, numerous attempts have been made to interfere with the exaggerated proinflammatory cascade initiated by the toxins. As most antitoxic and anti-inflammatory agents have shown no clear efficacy, an attractive alternative has been to prevent or minimise their release. Therefore, it was of interest to further study the antibiotic-induced release of toxins after exposure to antibiotics used for the treatment of the most severe infections, especially if protein synthesis inhibitors could reduce the release induced by PBP 3-specific β-lactam antibiotics. There were significant reductions in endotoxin release from gram-negative bacteria when the combination of the PBP 3-specific β-lactam antibiotic, cefuroxime, and the protein synthesis inhibitor, tobramycin, was compared with cefuroxime alone. Increasing doses of tobramycin reduced endotoxin release and increased the killing rate. In a kinetic in vitro model the endotoxin release from E.coli was higher after the second dose of cefuroxime. Nevertheless, it was reduced after addition of tobramycin. No binding of tobramycin to endotoxin was observed, either in vivo or in vitro. In a porcine sepsis model, a possible anti-inflammatory effect of ceftazidime and tobramycin, expressed as late cytokine inhibition, was seen. The protein synthesis inhibitor, clindamycin, released less streptococcal pyrogenic exotoxin A (SpeA) from a group A streptococcus strain than penicillin, and addition of clindamycin to penicillin resulted in less toxin production than penicillin alone. The SpeA production was dependent on the bacterial number at the start of treatment. Higher doses of penicillin also led to less SpeA. The choice of antibiotic class and dose may be important in the severely ill septic patient in whom an additional toxin release could be deleterious. A combination of a β-lactam antibiotic and a protein synthesis inhibitor seems beneficial but further investigations are needed.

Antibiotic treatment causes an ecological disturbance on the human microflora. Four commensal bacteria: E. coli, enterococci, a-streptococci and coagulase-negative staphylococci, from patients with extensive, high antibiotic usage were investigated with regard to resistance pattern and mutation frequency. Among 193 investigated strains it was found that high antibiotic usage selected for resistant bacteria and enriched for bacteria with a small but significantly increased mutation frequency. The relative biological fitness cost of resistance in Staphylococcus epidermidis was assessed in a human in vivo model where the indigenous flora was present. In vitro data of the bacterial growth rate correlated well to in vivo fitness assayed in the competition experiments on skin. An in vitro kinetic model was shown to be a useful tool to establish the pharmacokinetic and pharmacodynamic (PK/PD) indices for efficacy of antibiotics. It was confirmed that the time, when the concentration exceeds the minimal inhibitory concentration (MIC), correlates with efficacy for b-lactam antibiotics. To achieve maximal killing for penicillin-resistant pneumococci, with an MIC of 2 mg/L, the peak concentration was also of importance. Suboptimal dosing regimen facilitates selection of resistance. Penicillin-resistant pneumococci were easily selected in a mixed population with penicillin-sensitive, -intermediate and -resistant pneumococci in an in vitro kinetic model. The selection of the resistant strain was prevented when the benzylpenicillin concentration exceeded the MIC for approximately 50% of 24 h.

The neurotoxic potential of intravenous administered benzylpenicillin (BPC) was studied in rabbits with intact blood-CNS barriers and rabbits with experimental E. coli meningitis. At onset of epileptogenic EEG activity or seizures, serum, CSF and brain tissue were collected for assay of BPC. Based on the fact that, in tissues, BPC seems to remain extracellularly, brain concentrations of BPC were expressed as brain tissue fluid (BTF) levels, calculated as lOx the concentration in whole brain tissue. Neurotoxicity could be precipitated in all rabbits. In normal rabbits BTF levels of BPC were considerably higher than those in CSF indicating a better penetration across the blood-brain barrier (BBB). BPC penetrated better to CSF and BTF in meningitic rabbits than in normal controls, suggesting some degree of damage of the BBB concomitant with meningeal inflammation. E. coli meningitis did not increase the neurotoxicity of BPC. In control rabbits the intracistemal injection of saline resulted in some degree of pleocytosis. Unmanipulated animals are therefore preferable as controls. Epileptogenic EEG-changes was the most precise of the two variables used for demonstration of neurotoxicity. EEG-changes were therefore used as neurotoxicity criterion in the following rabbit experiments. To evaluate the effect of uraemia alone and uraemia plus meningitis on the neurotoxity of BPC in rabbits, cephaloridine was used to induce uraemia. Meningitis was induced by intracistemal inoculation of a cephalosporinresistant strain of E. cloacae. Untreated  rabbits were used as controls. Uraemia resulted in increased BTF penetration of BPC, possibly explained by permeability changes in the BBB and/or decreased binding of BPC to albumin. Uraemia did not result in increased penetration of BPC into the CSF of non-meningitic rabbits. Uraemic non-meningitic rabbits had the highest BTF levels of BPC at the criterion, indicating that cephaloridine-induced renal failure increased the epileptogenic threshold in these rabbits. The combination of uraemia and meningitis increased the neurotoxicity of BPC since the criterion was reached at considerably lower BTF levels of BPC. Meningitis, either alone or together with uraemia, did not increase the neurotoxicity in comparison to control rabbits. Higher BTF levels of BPC were found in meningitic rabbits than in controls with intact blood-CNS barriers at onset of EEG-changes. In all groups of rabbits there was a pronounced variability of BPC levels in the CSF while the intra-group variations in BTF levels were much smaller. Thus, BTF and not CSF levels were decisive for the neurotoxicity of BPC. Using   the same EEG-model, the neurotoxic potential of imipenem/cilastatin (I) and a new penem derivative, FCE 22101 were compared in a cross-over study. Both I and FCE 22101 were significantly more neurotoxic than BPC. While BTF levels of the three antibiotics could be detected in all tested rabbits, detectable CSF levels were only found in one of twelve rabbits treated with I or FCE 22101, indicating that BTF concentrations rather than CSF ones are decisive for neurotoxicity of ß-lactam antibiotics. The EEG-model used was found to be a suitable model for cross-over studies of intravenously administered antibiotics. Using the "silent-second" as EEG-threshold, a CNS interaction between intraperitoneally administered BPC and intravenous thiopental was demonstrated in rats. The most probably site for this interaction is the organic acid transport system out of the CNS. Thiopental distribution in the rat brain seemed to depend not only on its lipid solubility. / <p>Diss. (sammanfattning) Umeå : Umeå universitet, 1988, härtill 5 uppsatser.</p> / digitalisering@umu

Les hypersensibilités retardées (HSR) cutanées sont hétérogènes, à la fois par la nature des mécanismes impliqués (allergiques versus non allergiques) mais aussi par les différents degrés de sévérité rencontrés. Seules les HSR allergiques sont dues à la présence de lymphocytes T (LT), mal caractérisés chez l'homme. Le but de ce travail est d'étudier la contribution des LT CD8 cytotoxiques dans le développement et la sévérité des HSR cutanées chez l'homme, à partir de deux pathologies fréquentes : les toxidermies aux béta lactamines et l'eczéma allergique de contact à la para-phénylènediamine (PPD). Tout d'abord, la présence de LT spécifiques de médicament au sein des toxidermies aux béta lactamines a été recherchée in vivo et in vitro. Nous avons montré que les HSR sévères étaient plus souvent d'origine allergique que les HSR bénignes. Nous avons ensuite caractérisé le rôle des LT CD8 dans les HSR allergiques. Dans les toxidermies bénignes à l'amoxicilline, l'étude de la cinétique de recrutement des LT au niveau cutané ainsi que l'analyse des LT spécifiques du sang circulant ont permis de mettre en évidence le rôle essentiel des LT CD8 cytotoxiques dans l'initiation de ces réactions. Ensuite, dans l'eczéma allergique de contact à la PPD, un recrutement épidermique précoce des LT CD8 associés à des marqueurs de cytotoxicité, a été retrouvé, corrélé avec la sévérité des lésions. Ces résultats ont été confortés par ceux obtenus dans un modèle pré-clinique d'HSR allergique à la PPD chez la souris. En conclusion, ce travail montre que les LT CD8 cytotoxiques pourraient être les principales cellules effectrices des HSR cutanées allergiques chez l'homme / Skin delayed hypersensitivity (DHS) are heterogeneous, by the nature of the mechanisms involved (allergic versus non allergic) and also by their different degrees of severity. Only allergic DHS is due to T cells, poorly characterized in humans. The aim of this work is to study the contribution of cytotoxic CD8 T cells in the development and severity of skin DHS in humans, induced by two common diseases: cutaneous adverse drug reactions to beta lactam antibiotics and allergic contact dermatitis to para-phenylenediamine (PPD). First, the presence of drug specific T cells in cutaneous adverse drug reactions to beta lactams was investigated in vivo and in vitro. We showed that severe DHS were more often allergic than benign DHS. Then, we characterized the role of CD8 T cells in allergic DHS. In benign cutaneous adverse drug reactions to amoxicillin, the study of the kinetics of skin T cell recruitment as well as the analysis of circulating specific T cells highlight the essential role of cytotoxic CD8 T cells in the initiation phase of these reactions. In allergic contact dermatitis to PPD, early recruitment of epidermal CD8 T cells associated with cytotoxic markers was found, correlated with the severity of lesions. These results were supported by those obtained in a mouse model of allergic contact dermatitis to PPD. In conclusion, this work showed that cytotoxic CD8 T cells could be the main effector cells of allergic skin DHS in humans