Finite Horizon Learning

Incorporating adaptive learning into macroeconomics requires assumptions about how agents incorporate their forecasts into their decision-making. We develop a theory of bounded rationality that we call finite-horizon learning. This approach generalizes the two existing benchmarks in the literature: Eulerequation learning, which assumes that consumption decisions are made to satisfy the one-step-ahead perceived Euler equation; and infinite-horizon learning, in which consumption today is determined optimally from an infinite-horizon optimization problem with given beliefs. In our approach, agents hold a finite forecasting/planning horizon. We find for the Ramsey model that the unique rational expectations equilibrium is E-stable at all horizons. However, transitional dynamics can differ significantly depending upon the horizon.

Anticipation, Learning and Welfare: the Case of Distortionary Taxation

We study the impact of anticipated fiscal policy changes in a Ramsey economy where agents form long-horizon expectations using adaptive learning. We extend the existing framework by introducing distortionary taxes as well as elastic labour supply, which makes agents. decisions non-predetermined but more realistic. We detect that the dynamic responses to anticipated tax changes under learning have oscillatory behaviour that can be interpreted as self-fulfilling waves of optimism and pessimism emerging from systematic forecast errors. Moreover, we demonstrate that these waves can have important implications for the welfare consequences of .scal reforms. (JEL: E32, E62, D84)

Unilateral Emissions Mitigation, Spillovers, and Global Learning

What's the role of unilateral measures in global climate change mitigation in a post-Durban, post 2012 global policy regime? We argue that under conditions of preference heterogeneity, unilateral emissions mitigation at a subnational level may exist even when a nation is unwilling to commit to emission cuts. As the fraction of individuals unilaterally cutting emissions in a global strongly connected network of countries evolves over time, learning the costs of cutting emissions can result in the adoption of such activities globally and we establish that this will indeed happen under certain assumptions. We analyze the features of a policy proposal that could accelerate convergence to a low carbon world in the presence of global learning.

What is the Causal Effect of Information and Learning about a Public Good on Willingness to Pay?

In this study we elicit agents’ prior information set regarding a public good, exogenously give information treatments to survey respondents and subsequently elicit willingness to pay for the good and posterior information sets. The design of this field experiment allows us to perform theoretically motivated hypothesis testing between different updating rules: non-informative updating, Bayesian updating, and incomplete updating. We find causal evidence that agents imperfectly update their information sets. We also field causal evidence that the amount of additional information provided to subjects relative to their pre-existing information levels can affect stated WTP in ways consistent overload from too much learning. This result raises important (though familiar) issues for the use of stated preference methods in policy analysis.

Benzodiazepine receptor ligand influences on learning: an endogenous modulatory mechanism mediated by benzodiazepines possibly of alimentary origin

In rats pre-but not post-training ip administration of either flumazenil, a central benzodiazepine (BSD) receptor antagonist, or of n-butyl-B-carboline-carboxylate (BCCB), an inverse agonist, enhanced retention of inhibitory avoidance learning. Flumazenil vlocked the enhancing effect of BCCB, and the inhibitory effect of the BZD agonists clonazepam and diazepam also given pre-training. Post-training administration of these drugs had no effects. The peripheral BZD receptor agonist/chloride channel blocker Ro5-4864 had no effect on the inhibitory avoidance task when given ip prior to training, buth it caused enhancement when given immediately post-training either ip or icv. This effect was blocked by PK11195, a competitive antagonist of Ro5-4864. These results suggest that ther is an endogenous mechanism mediated by BZD agonists, which is sensitive to inverse agonists and that normally down-regulates the formation of memories through a mechanism involving GABA-A receptors and the corresponding chloride channels. The most likely agonists for the endogenous mechanism suggested are the diazepam-like BZDs found in brain whose origin is possibly alimentary. Levels of these BZDs in the cortex were found to sharply decrease after inhibitory acoidance training or mere exposure to the training apparatus.

Development of Ewing's sarcoma from primary bone marrow-derived mesenchymal stem cells

SUMMARY : Ewing's sarcoma is a member of Ewing's family tumors (ESPY) and the second most common solid bone and soft tissue malignancy of children and young adults. It is associated in 85% of cases with the t(11;22)(q24:q12) chromosomal translocation that generates fusion of the 5' segment of the EWSR1 gene with the 3' segment of the ETS family gene FLI-1. The EWSR1-FLI-1 fusion protein behaves as an aberrant transcriptional activator and is believed to contribute to ESFT development. However, EWSR1-FLI-1 induces growth arrest and apoptosis in normal fibroblasts, and primary cells that are pemissive for its putative oncogenic properties have not been discovered, hampering basic understanding of ESFT biology. Here, we show that EWSR1-FLI-1 alone can transform mouse primary bone marrow-derived mesenchymal progenitor cells and generate tumors that display hallmarks of Ewing's sarcoma, including a small round cell phenotype, expression of ESFT-associated markers, insulin like growth factor-I dependence, and induction or repression of numerous EWSR1-FLI-1 target genes. Consistent with this finding, we tested the possibility that human mesenchymal stem cells (hMSC) might also provide a permissive cellular environment for EWSR1-FLI-1, and could represent the first adequate primary human cellular background for the oncogenic properties of the fusion protein. Indeed, expression of EWSR1-FLI-1 in human mesenchymal stem cells (hMSC) was not only stably maintained without inhibiting proliferation, but induced a gene expression profile bearing striking similarity to that of ESFT, including genes that are among the highest ESFT discriminators. Expression of EWSR1-FLI-1 in hMSCs may recapitulate the initial steps of Ewing's sarcoma development, allowing identification of genes that play an important role early in its pathogenesis. Among relevant candidate transcripts induced by EWSR1-FL/-1 in hMSC we found the polycomb group gene EZH2 which we show to play a critical role in Ewing's sarcoma growth. These observations provide the first identification of candidate primary cells from which ESFTs originate and suggest that EWSR1-FLI-1 expression may constitute the initiating event in ESFT pathogenesis. Le sarcome d' Ewing est un membre de la famille des tumeurs Ewing (ESFT) et représente la deuxième tumeur maligne solide de l'os et des tissus mous chez les enfants et les jeunes adultes. Cette tumeur est associée dans 85% des cas avec la translocation chromosomique t(11;22)(g24:g12), qui génère la fusion entre le segment 5' du gène EWSR1 avec le segment 3' du gène FLI-1, appartenant à la famille des facteurs de transcription ETS. La protéine de fusion EWSR1-FLI-1 qui en dérive joue le rSle d'un facteur de transcription aberrant, et est supposée contribuer de manière décisive au processus de développement des ESFTs. Néanmoins, l'expression de EWSR1-FLI-1 dans des fibroblastes normaux induit un arrêt de croissance et leur apoptose, et les cellules primaires permissives pour les propriétés oncogéniques attribuées à la translocation n'ont pas encore été identifiées, empêchant la compréhension de la biologie de base du sarcome d'Ewing. Dans ce travail on montre que l'expression de EWSR1-FLI-1 uniquement est capable de transformer des cellules souches mésenchymateuses dérivées de la moelle osseuse de la souris, pour générer des tumeurs qui présentent les caractéristiques du sarcome d' Ewing humain, et notamment une morphologie de petites cellules bleues et rondes, l'expression de marqueurs associés aux ESFTs, une dépendance du facteur de croissance IGF-1, et l'induction ou la répression de nombreux gènes cibles connus de EWSR1-FLI-1. Sur la base de ces observations, on a testé la possibilité que les cellules souches mésenchymateuses humaines (hMSCs) puissent aussi fournir un environnement cellulaire permissif pour EWSR1-FLI-1 ; et représenter le premier background cellulaire humain adéquat pour la manifestation du pouvoir oncogénique de la protéine de fusion. En effet, l'expression de EWSR1-FLI-1 dans des cellules souches mésenchymateuses humaines s'est révélée non seulement maintenue, mais elle a induit un profil d'expression génétique étonnamment similaire à celui des ESFTs humains, incluant les gènes qui ont été rapportés comme étant les plus discriminatifs pour ces tumeurs. L'expression de EWSR1-FLI-1 dans les hMSCs pourrait récapituler les étapes initiales du développement du sarcome d' Ewing, et de ce fait consentir à identifier les gènes qui jouent un rôle crucial dans sa pathogenèse précoce. Parmi les transcrits relevant indults par EWSR1-FL/-9 dans les hMSCs nous avons découvert le gène du groupe des polycomb EZH2, que nous avons par la suite démontré jouer un rôle essentiel dans la croissance du sarcome de Ewing. Ces observations apportent pour la première fois l'identification d'une cellule primaire candidate pour représenter la cellule d'origine des ESFTs, et en même temps suggèrent que l'expression de EWSR1-FLI-1 peut constituer l'événement initial dans la pathogenèse du sarcome d' Ewing.

Differentiation of rat striatal embryonic stem cells in vitro: monolayer culture vs. three-dimensional coculture with differentiated brain cells.

Several groups have demonstrated the existence of self-renewing stem cells in embryonic and adult mouse brain. In vitro, these cells proliferate in response to epidermal growth factor, forming clusters of nestin-positive cells that may be dissociated and subcultured repetitively. Here we show that, in stem cell clusters derived from rat embryonic striatum, cell proliferation decreased with increasing number of passages and in response to elevated concentrations of potassium (30 mM KCl). In monolayer culture, the appearance of microtubule-associated protein type-5-immunoreactive (MAP-5(+)) cells (presumptive neurons) in response to basic fibroblast growth factor (bFGF) was reduced at low cell density and with increasing number of passages. In the presence of bFGF, elevated potassium caused a more differentiated neuronal phenotype, characterized by an increased proportion of MAP-5(+) cells, extensive neuritic branching, and higher specific activity of glutamic acid decarboxylase. Dissociated stem cells were able to invade cultured brain cell aggregates containing different proportions of neurons and glial cells, whereas they required the presence of a considerable proportion of glial cells in the host cultures to become neurofilament H-positive. The latter observation supports the view that astrocyte-derived factors influence early differentiation of the neuronal cell lineage.

Learning to live together: an exploration and analysis of managing cultural diversity in ten early childhood development centres in South Africa

This study explores how South African Early Childhood Development (ECD) Practitioners and families meet the needs of the increasing number of children from diverse cultural backgrounds in their care. Research participants were identified through ten ECD centres located in two urban communities in the Eastern and Western Cape Provinces of South Africa. The values and attitudes held by Practitioners and families vis-à-vis cultural diversity was investigated, along with the knowledge and strategies they employ to manage cultural diversity in ECD programmes. The intercultural education model provides the necessary tools to address the challenges identified.

Learning-induced plasticity in human audition: Objects, time, and space.

The human auditory system is comprised of specialized but interacting anatomic and functional pathways encoding object, spatial, and temporal information. We review how learning-induced plasticity manifests along these pathways and to what extent there are common mechanisms subserving such plasticity. A first series of experiments establishes a temporal hierarchy along which sounds of objects are discriminated along basic to fine-grained categorical boundaries and learned representations. A widespread network of temporal and (pre)frontal brain regions contributes to object discrimination via recursive processing. Learning-induced plasticity typically manifested as repetition suppression within a common set of brain regions. A second series considered how the temporal sequence of sound sources is represented. We show that lateralized responsiveness during the initial encoding phase of pairs of auditory spatial stimuli is critical for their accurate ordered perception. Finally, we consider how spatial representations are formed and modified through training-induced learning. A population-based model of spatial processing is supported wherein temporal and parietal structures interact in the encoding of relative and absolute spatial information over the initial ∼300ms post-stimulus onset. Collectively, these data provide insights into the functional organization of human audition and open directions for new developments in targeted diagnostic and neurorehabilitation strategies.

The behavioral relevance and top-down susceptibility of early, low level multisensory interactions in humans

Abstract (English)General backgroundMultisensory stimuli are easier to recognize, can improve learning and a processed faster compared to unisensory ones. As such, the ability an organism has to extract and synthesize relevant sensory inputs across multiple sensory modalities shapes his perception of and interaction with the environment. A major question in the scientific field is how the brain extracts and fuses relevant information to create a unified perceptual representation (but also how it segregates unrelated information). This fusion between the senses has been termed "multisensory integration", a notion that derives from seminal animal single-cell studies performed in the superior colliculus, a subcortical structure shown to create a multisensory output differing from the sum of its unisensory inputs. At the cortical level, integration of multisensory information is traditionally deferred to higher classical associative cortical regions within the frontal, temporal and parietal lobes, after extensive processing within the sensory-specific and segregated pathways. However, many anatomical, electrophysiological and neuroimaging findings now speak for multisensory convergence and interactions as a distributed process beginning much earlier than previously appreciated and within the initial stages of sensory processing.The work presented in this thesis is aimed at studying the neural basis and mechanisms of how the human brain combines sensory information between the senses of hearing and touch. Early latency non-linear auditory-somatosensory neural response interactions have been repeatedly observed in humans and non-human primates. Whether these early, low-level interactions are directly influencing behavioral outcomes remains an open question as they have been observed under diverse experimental circumstances such as anesthesia, passive stimulation, as well as speeded reaction time tasks. Under laboratory settings, it has been demonstrated that simple reaction times to auditory-somatosensory stimuli are facilitated over their unisensory counterparts both when delivered to the same spatial location or not, suggesting that audi- tory-somatosensory integration must occur in cerebral regions with large-scale spatial representations. However experiments that required the spatial processing of the stimuli have observed effects limited to spatially aligned conditions or varying depending on which body part was stimulated. Whether those divergences stem from task requirements and/or the need for spatial processing has not been firmly established.Hypotheses and experimental resultsIn a first study, we hypothesized that auditory-somatosensory early non-linear multisensory neural response interactions are relevant to behavior. Performing a median split according to reaction time of a subset of behavioral and electroencephalographic data, we found that the earliest non-linear multisensory interactions measured within the EEG signal (i.e. between 40-83ms post-stimulus onset) were specific to fast reaction times indicating a direct correlation of early neural response interactions and behavior.In a second study, we hypothesized that the relevance of spatial information for task performance has an impact on behavioral measures of auditory-somatosensory integration. Across two psychophysical experiments we show that facilitated detection occurs even when attending to spatial information, with no modulation according to spatial alignment of the stimuli. On the other hand, discrimination performance with probes, quantified using sensitivity (d'), is impaired following multisensory trials in general and significantly more so following misaligned multisensory trials.In a third study, we hypothesized that behavioral improvements might vary depending which body part is stimulated. Preliminary results suggest a possible dissociation between behavioral improvements andERPs. RTs to multisensory stimuli were modulated by space only in the case when somatosensory stimuli were delivered to the neck whereas multisensory ERPs were modulated by spatial alignment for both types of somatosensory stimuli.ConclusionThis thesis provides insight into the functional role played by early, low-level multisensory interac-tions. Combining psychophysics and electrical neuroimaging techniques we demonstrate the behavioral re-levance of early and low-level interactions in the normal human system. Moreover, we show that these early interactions are hermetic to top-down influences on spatial processing suggesting their occurrence within cerebral regions having access to large-scale spatial representations. We finally highlight specific interactions between auditory space and somatosensory stimulation on different body parts. Gaining an in-depth understanding of how multisensory integration normally operates is of central importance as it will ultimately permit us to consider how the impaired brain could benefit from rehabilitation with multisensory stimula-Abstract (French)Background théoriqueDes stimuli multisensoriels sont plus faciles à reconnaître, peuvent améliorer l'apprentissage et sont traités plus rapidement comparé à des stimuli unisensoriels. Ainsi, la capacité qu'un organisme possède à extraire et à synthétiser avec ses différentes modalités sensorielles des inputs sensoriels pertinents, façonne sa perception et son interaction avec l'environnement. Une question majeure dans le domaine scientifique est comment le cerveau parvient à extraire et à fusionner des stimuli pour créer une représentation percep- tuelle cohérente (mais aussi comment il isole les stimuli sans rapport). Cette fusion entre les sens est appelée "intégration multisensorielle", une notion qui provient de travaux effectués dans le colliculus supérieur chez l'animal, une structure sous-corticale possédant des neurones produisant une sortie multisensorielle différant de la somme des entrées unisensorielles. Traditionnellement, l'intégration d'informations multisen- sorielles au niveau cortical est considérée comme se produisant tardivement dans les aires associatives supérieures dans les lobes frontaux, temporaux et pariétaux, suite à un traitement extensif au sein de régions unisensorielles primaires. Cependant, plusieurs découvertes anatomiques, électrophysiologiques et de neuroimageries remettent en question ce postulat, suggérant l'existence d'une convergence et d'interactions multisensorielles précoces.Les travaux présentés dans cette thèse sont destinés à mieux comprendre les bases neuronales et les mécanismes impliqués dans la combinaison d'informations sensorielles entre les sens de l'audition et du toucher chez l'homme. Des interactions neuronales non-linéaires précoces audio-somatosensorielles ont été observées à maintes reprises chez l'homme et le singe dans des circonstances aussi variées que sous anes- thésie, avec stimulation passive, et lors de tâches nécessitant un comportement (une détection simple de stimuli, par exemple). Ainsi, le rôle fonctionnel joué par ces interactions à une étape du traitement de l'information si précoce demeure une question ouverte. Il a également été démontré que les temps de réaction en réponse à des stimuli audio-somatosensoriels sont facilités par rapport à leurs homologues unisensoriels indépendamment de leur position spatiale. Ce résultat suggère que l'intégration audio- somatosensorielle se produit dans des régions cérébrales possédant des représentations spatiales à large échelle. Cependant, des expériences qui ont exigé un traitement spatial des stimuli ont produits des effets limités à des conditions où les stimuli multisensoriels étaient, alignés dans l'espace ou encore comme pouvant varier selon la partie de corps stimulée. Il n'a pas été établi à ce jour si ces divergences pourraient être dues aux contraintes liées à la tâche et/ou à la nécessité d'un traitement de l'information spatiale.Hypothèse et résultats expérimentauxDans une première étude, nous avons émis l'hypothèse que les interactions audio- somatosensorielles précoces sont pertinentes pour le comportement. En effectuant un partage des temps de réaction par rapport à la médiane d'un sous-ensemble de données comportementales et électroencépha- lographiques, nous avons constaté que les interactions multisensorielles qui se produisent à des latences précoces (entre 40-83ms) sont spécifique aux temps de réaction rapides indiquant une corrélation directe entre ces interactions neuronales précoces et le comportement.Dans une deuxième étude, nous avons émis l'hypothèse que si l'information spatiale devient perti-nente pour la tâche, elle pourrait exercer une influence sur des mesures comportementales de l'intégration audio-somatosensorielles. Dans deux expériences psychophysiques, nous montrons que même si les participants prêtent attention à l'information spatiale, une facilitation de la détection se produit et ce toujours indépendamment de la configuration spatiale des stimuli. Cependant, la performance de discrimination, quantifiée à l'aide d'un index de sensibilité (d') est altérée suite aux essais multisensoriels en général et de manière plus significative pour les essais multisensoriels non-alignés dans l'espace.Dans une troisième étude, nous avons émis l'hypothèse que des améliorations comportementales pourraient différer selon la partie du corps qui est stimulée (la main vs. la nuque). Des résultats préliminaires suggèrent une dissociation possible entre une facilitation comportementale et les potentiels évoqués. Les temps de réactions étaient influencés par la configuration spatiale uniquement dans le cas ou les stimuli somatosensoriels étaient sur la nuque alors que les potentiels évoqués étaient modulés par l'alignement spatial pour les deux types de stimuli somatosensorielles.ConclusionCette thèse apporte des éléments nouveaux concernant le rôle fonctionnel joué par les interactions multisensorielles précoces de bas niveau. En combinant la psychophysique et la neuroimagerie électrique, nous démontrons la pertinence comportementale des ces interactions dans le système humain normal. Par ailleurs, nous montrons que ces interactions précoces sont hermétiques aux influences dites «top-down» sur le traitement spatial suggérant leur occurrence dans des régions cérébrales ayant accès à des représentations spatiales de grande échelle. Nous soulignons enfin des interactions spécifiques entre l'espace auditif et la stimulation somatosensorielle sur différentes parties du corps. Approfondir la connaissance concernant les bases neuronales et les mécanismes impliqués dans l'intégration multisensorielle dans le système normale est d'une importance centrale car elle permettra d'examiner et de mieux comprendre comment le cerveau déficient pourrait bénéficier d'une réhabilitation avec la stimulation multisensorielle.

A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin, and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.

Comparable Survival Between HIV+ and HIV- non-Hodgkin and Hodgkin Lymphoma Patients Undergoing Autologous Peripheral Blood Stem Cell Transplantation.

Autologous stem cell transplantation (ASCT) has been successfully used in HIV-related lymphoma (HIV-Ly) patients on highly active antiretroviral therapy. We report the first comparative analysis between HIV-Ly and a matched cohort of HIV(-) lymphoma patients. This retrospective European Group for Blood and Marrow Transplantation study included 53 patients (66% non-Hodgkin and 34% Hodgkin lymphoma) within each cohort. Both groups were comparable except for the higher proportion of males, mixed-cellularity Hodgkin lymphoma and patients receiving granulocyte colony-stimulating factor before engraftment and a smaller proportion receiving total body irradiation-based conditioning within the HIV-Ly cohort. Incidence of relapse, overall survival, and progression-free survival were similar in both cohorts. A higher nonrelapse mortality within the first year after ASCT was observed in the HIV-Ly group (8% vs 2%), predominantly because of early bacterial infections, although this was not statistically significant and did not influence survival. Thus, within the highly active antiretroviral therapy era, HIV patients should be considered for ASCT according to the same criteria adopted for HIV(-) lymphoma patients.

Manipulating keratinocyte stem cell proliferation and differentiation using RNA interference

ABSTRACT : The epidermis, the outermost compartment of the skin, is a stratified and squamous epithelium that constantly self-renews. Keratinocytes, which represent the main epidermal population, are responsible for its cohesion and barrier function. Epidermal renewal necessitates a fine equilibrium between keratinocyte proliferation and differentiation. The keratinocyte stem cell, located in the basal cell layer, is responsible for epidermal homeostasis and regeneration during the wound healing process. The transcription factor p63 structurally belongs to the p53 superfamily. It is expressed in the basal and supra-basal cell layers of stratified epithelia and is thought to be important for the renewal or the differentiation of keratinocyte stem cells (Yang et al., 1999; Mills et al., 1999). In order to better understand its function, we established an in vitro model of p63 deficient human keratinocyte stem cells using a shp63 mediated RNA interference. Knockdown of endogenous p63 induces downregulation of cell-adhesion genes as previously described (Carroll et al., 2006). Interestingly, the replating of attached p63-knockdown keratinocytes on a feeder layer results in a loss of attachment and proliferation. They are no longer clonogenic. However, if the same population are replated in a fibrin matrix, extended fibrinolysis is reported, a common process in wound healing, suggesting that p63 regulates the fibrinolytic pathway. This result was confirmed by Q-PCR and shows that the urokinase pathway, which mediates fibrinolysis, is upregulated. Altogether, these findings suggest a mechanism in which the fine tuning of p63 expression promotes attachment or release of the keratinocyte stem cell from the basement membrane by inducing genes of adhesion and/or of fibrinolysis. This mechanism may be important for epidermal self-renewal, differentiation as well as wound healing. Its misregulation may be partly responsible for the p63 knockout phenotype. The downregulation of p63 also induces a decrease in LEKTI expression. LEKTI (lymphoepithelial Kazal-type serine protease inhibitor) is a serine protease inhibitor encoded by the Spink5 gene. It is expressed and secreted in the uppermost differentiated layers of stratified epithelia and plays a role in the desquamation process. When this gene is disrupted, humans develop the Netherton syndrome (Chavanas et al., 2000b). It is a dermatosis characterized by hair dysplasias, ichtyosiform erythroderma and impairment in epidermal barrier function promoting inflammation similarly as in psoriasis with inflammatory infiltrate in excess. TNFα (tumor necrosis factor alpha) and EDA1 (ectodysplasin A1) are two transmembraneprecursors that belong to the TNF superfamily, which is involved in immune and inflammation regulation (Smahi et al., 2002). We suggest that the secreted serine protease inhibitor LEKTI plays a role in the regulation of TNFα and EDA1 precursor cleavage and absence of LEKTI induces excess of inflammation. To investigate this hypothesis, we induced downregulation of Spink5 expression in rat keratinocyte stem cells by using a shSpink5 mediated RNA interference approach. Interestingly, expression of TNFα and EDA1 is modified after knockdown of Spink5 by Q-PCR. Moreover, downregulation of Spink5 induces loss of cohesiveness between keratinocytes and colonies adopt a scattered phenotype. Altogether, these preliminary data suggest that downregulation of LEKTI may play a role in the inflammatory response in Netherton syndrome patients, by regulating TNFα expression.

The role of Pten in the Hematopoietic System and the Hematopoietic Stem Cells

Résumé Identification, localisation et activation des cellules souches hématopoiétiques dormantes in vivo Les cellules souches somatiques sont présentes dans la majorité des tissus régénératifs comme la peau, l'épithélium intestinal et le système hématopoiétique. A partir d'une seule cellule, elles ont les capacités de produire d'autres cellules souches du même type (auto-renouvellement) et d'engendrer un ensemble défini de cellules progénitrices différenciées qui vont maintenir ou réparer leur tissu hôte. Les cellules souches adultes les mieux caractérisées sont les cellules souches hématopoiétiques (HSC), localisées dans la moelle osseuse. Un des buts de mon travail de doctorat était de caractériser plus en profondeur la localisation des HSCs endogènes in vivo. Pour ce faire, la technique "label retaining assay", se basant sur la division peu fréquentes et sur la dormance des cellules souches, a été utilisée. Après un marquage des souris avec du BrdU (analogue à l'ADN) suivi d'une longue période sans BrdU, les cellules ayant incorporés le marquage ("label retaining cells" LCRs) ont pu être identifiées dans la moelle osseuse. Ces cellules LCRs étaient enrichies 300 fois en cellules de phenotype HSC et, en utilisant de la cytofluorométrie, il a pu être montré qu'environ 15% de toutes les HSCs d'une souris restent dormantes durant plusieures semaines. Ces HSCs dormantes à long terme ne sont probablement pas impliquées dans la maintenance de 'hématopoièse. Par contre, on assiste à l'activation rapide de ces HSCs dormantes lors d'une blessure, comme une ablation myéloide. Elles re-entrent alors en cycle cellulaire et sont essentielles pour une génération rapide des cellules progénitrices et matures qui vont remplacer les cellules perdues. De plus, la détection des LCRs, combinée avec l'utilisation du marqueur de HSCs c-kit, peut être utilisée pour la localisation des HSCs dormantes présentes dans la paroi endostéale de la cavité osseuse. De manière surprenante, les LCRs c-kit+ ont surtout étés trouvées isolées en cellule unique, suggérant que le micro-environement spécifique entourant et maintenant les HSCs, appelé niche, pourrait être très réduit et abriter une seule HSC par niche. Rôles complexes du gène supresseur de tumeur Pten dans le système hématopoiétique La phosphatase PTEN disparaît dans certains cancers héréditaires ou sporadiques humains, comme les gliomes, les cancers de l'utérus ou du sein. Pten inhibe la voie de signalisation de la PI3-kinase et joue un rôle clé dans l'apoptose, la croissance, la prolifération et la migration cellulaire. Notre but était d'étudier le rôle de Pten dans les HSC normale et durant la formation de leucémies. Pour ce faire, nous avons généré un modèle murin dans lequel le gène Pten peut être supprimé dans les cellules hématopoiétiques, incluant les HSCs. Ceci a été possible en croissant l'allèle conditionnelle ptenflox soit avec le transgène MxCre inductible par l'interféron α soit avec le transgène Scl-CreERt inductible par le tamoxifen. Ceci permet la conversion de l'allèle ptenflox en l'allèle nul PtenΔ dans les HSCs et les autres types cellulaires hématopoiétiques. Les souris mutantes Pten développent une splénomégalie massive causée par une expansion dramatiques de toutes les cellules myéloides. De manière interessante, alors que le nombre de HSCs dans la moelle osseuse diminue progressivement, le nombre des HSCs dans la rate augmente de manière proportionnelle. Etrangement, les analyses de cycle cellulaire ont montrés que Pten n'avait que peu ou pas d'effet sur la dormance des HSCs ou sur leur autorenouvellement. En revanche, une augmentation massive du niveau de la cytokine de mobilisation G-CSF a été détéctée dans le serum sanguin, suggérant que la suppression de Pten stimulerait la mobilisation et la migration des HSC de la moelle osseuse vers la rate. Finallement, la transplantation de moelle osseuse délétée en Pten dans des souris immuno-déficientes montre que Pten fonctionnerait comme un suppresseur de tumeur dans le système hématopoiétique car son absence entraîne la formation rapide de leucémies lymphocytaires. Summary Identification, localization and activation of dormant hematopoietic stun cells in vivo Somatic stem cells are present in most self-renewing tissues including the skin, the intestinal epithelium and the hematopoietic system. On a single cell basis they have the capacity to produce more stem cells of the same phenotype (self-renewal) and to give rise to a defined set of mature differentiated progeny, responsible for the maintenance or repair of the host tissue. The best characterized adult stem cell is the hematopoietic stem cell (HSC) located in the bone marrow. One goal of my thesis work was to further characterize the location of endogenous HSCs in vivo. To do this, a technique called "label retaining assay» was used which takes advantage of the fact that stem cells (including HSCs) divide very infrequently and can be dormant for months. After labeling mice with the DNA analogue BrdU followed by a long BrdU free "chase", BrdU "label retaining cells" (CRCs) could be identified in the bone marrow. These CRCs were 300-fold enriched for phenotypic HSCs and by using flow cytometry analysis it could be shown that about 15% of all HSCs in the mouse are dormant for many weeks. Our results suggest that these long-term dormant HSCs are unlikely to be involved in homeostatic maintenance. However they are rapidly activated and reenter the cell cycle in response to injury signals such as myeloid ablation. In addition, detection of LRCs in combination with the HSC marker c-Kit could be used to locate engrafted dormant HSCs close to the endosteal lining of the bone marrow cavities. Most surprisingly, c-Kit+LRCs were found predominantly as single cells suggesting that the specific stem cell maintaining microenvironment, called niche, has limited space and may house only single HSCs. Complex roles of the tumor suppressor gene Pten in the hematopoietic system. The phosphatase PTEN is lost in hereditary and sporadic forms of human cancers, including gliomas, endometrial and breast cancers. Pten inhibits the PI3-kina.se pathway and plays a key role in apoptosis, cell growth, proliferation and migration. Our aim was to study the role of Pten in normal HSCs and during leukemia formation. To do this, we generated a mouse model in which the Pten gene can be deleted in hematopoietic cells including HSCs. This was achieved by crossing the conditional ptenflox allele with either the interferona inducible MxCre or the tamoxifen inducible Scl-CreERT transgene. This allowed the conversion of the ptenflox allele into a pterr' null allele in HSCs and other hematopoietic cell types. As a result Pten mutant mice developed massive splenomegaly due to a dramatic expansion of all myeloid cells. Interestingly, while the number of bone marrow HSCs progressively decreased, the number of HSCs in the spleen increased to a similar extent. Unexpectedly, extensive cell cycle analysis showed that Pten had little or no effect on HSC dormancy or HSC self-renewal. Instead, dramatically increased levels of the mobilizing cytokine G-CSF were detected in the blood serum suggesting that loss-of Pten stimulates mobilization and migration of HSC from the BM to the spleen. Finally, transplantation of Pten deficient BM cells into immuno-compromised mice showed that Pten can function as a tumor suppressor in the hematopoietic system and that its absence leads to the rapid formation of T cell leukemia.