Indiana University-Purdue University Indianapolis (IUPUI) / SIRT1 (Sirtuin 1) is a founding member of a family of seven proteins and histone deacetylases. It is involved in cellular resistance to stress, metabolism, differentiation, aging, and tumor suppression. SIRT1-/- mice demonstrate embryonic and postnatal development defects. We examined hematopoietic and endothelial cell differentiation of SIRT1-/- mouse embryonic stem (mES) cells in vitro, and hematopoietic progenitors in SIRT1+/+, SIRT1+/-, and SIRT1-/- mice. SIRT1-/- ES cells exhibited markedly delayed/immature formation of blast colony-forming cells (BL-CFCs). When individual blast colonies were analyzed for hematopoietic and endothelial potential, replated SIRT1-/- BL-CFC possessed limited hematopoietic potential, whereas endothelial potential was essentially unaltered. The ability of SIRT1-/- ES cells to form primitive erythroid progenitors was not only delayed but greatly decreased. Moreover, after differentiation of SIRT1-/- mES cells, there were also significant decreases in granulocyte-macrophage (CFU-GM) and multipotential (CFU-GEMM) progenitor cells. Differentiation delay/defects were associated with delayed capacity to switch off Oct4, Nanog and Fgf5, decreased β-H1 globin, β-major globin, and Scl gene expression and reduced activation of the Erk1/2 pathway upon SIRT1-/- ES cell commitment. Reintroduction of WT SIRT1 into SIRT1-/- cells partially rescued the primitive erythroid progenitor formation of SIRT1-/- cells and the expression of hemoglobin genes, Hbb-bh1 and Hbb-b1, suggesting that the defect of hematopoietic commitment is due to deletion of SIRT1, and not to genetic drifting of SIRT1-/- cells. To confirm the requirement for SIRT1 for normal development of hematopoietic progenitor cells, we assessed embryonic and adult hematopoiesis in SIRT1+/+, SIRT1+/- and SIRT1-/- mice. Yolk sacs from SIRT1 mutant embryos generated fewer primitive erythroid precursors compared to wild-type (WT) and heterozygous mice. Moreover, knockout of SIRT1 decreased primary bone marrow hematopoietic progenitor cells (HPCs) in 5 week and 12 month old mice, which was especially notable at lower (5%) O2 tension. In addition these progenitors survived less well in vitro under conditions of delayed growth factor addition. Taken together, these results demonstrate that SIRT1 plays a role in ES cell hematopoietic differentiation and mouse hematopoiesis.

Myelodysplastic syndromes (MDS) comprise a heterogeneous group of hematologic malignancies characterized by clonal hematopoiesis, one or more cytopenias such as anemia, neutropenia, or thrombocytopenia, abnormal cellular maturation, and a high risk of progression to acute myeloid leukemia. The bone marrow microenvironment (BMME) in general and mesenchymal stromal cells (MSCs) in particular contribute to both the initiation and progression of MDS. However, little is known about the role of MSC-derived extracellularmatrix (ECM) in this context. Therefore, we performed a comparative analysis of in vitro deposited MSC-derived ECM of different MDS subtypes and healthy controls. Atomic force microscopy analyses demonstrated that MDS ECM was significantly thicker and more compliant than those from healthy MSCs. Scanning electron microscopy showed a dense meshwork of fibrillar bundles connected by numerous smaller structures that span the distance between fibers in MDS ECM. Glycosaminoglycan (GAG) structures were detectable at high abundance in MDS ECM as white, sponge-like arrays on top of the fibrillar network. Quantification by Blyscan assay confirmed these observations, with higher concentrations of sulfated GAGs in MDS ECM. Fluorescent lectin staining with wheat germ agglutinin and peanut agglutinin demonstrated increased deposition of N-acetyl-glucosamine GAGs (hyaluronan (HA) and heparan sulfate) in low risk (LR) MDS ECM. Differential expression of Nacetyl- galactosamine GAGs (chondroitin sulfate, dermatan sulfate) was observed between LR- and high risk (HR)-MDS. Moreover, increased amounts of HA in the matrix of MSCs from LR-MDS patients were found to correlate with enhanced HA synthase 1 mRNA expression in these cells. Stimulation of mononuclear cells from healthy donors with low molecular weight HA resulted in an increased expression of various pro-inflammatory cytokines suggesting a contribution of the ECM to the inflammatory BMME typical of LR-MDS. CD34+ hematopoietic stem and progenitor cells (HSPCs) displayed an impaired differentiation potential after cultivation on MDS ECM and modified morphology accompanied by decreased integrin expression which mediate cell-matrix interaction. In summary, we provide evidence for structural alterations of the MSC-derived ECM in both LR- and HR-MDS. GAGs may play an important role in this remodeling processes during the malignant transformation which leads to the observed disturbance in the support of normal hematopoiesis.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23772号 / 医博第4818号 / 新制||医||1056(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 村川 泰裕, 教授 松田 文彦, 教授 篠原 隆司 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Chez les humains, un large pourcentage de leucémies myéloïdes et lymphoïdes exprime des gènes Homéobox (Hox) de façon aberrante, principalement ceux du groupe des gènes Hoxa. Cette dérégulation de l’expression des gènes Hox peut provenir directement des translocations impliquant des gènes Hox ou indirectement par d’autres protéines ayant un potentiel oncogénique. De plus, plusieurs études indiquent que les gènes Hox jouent un rôle essentiel dans l'initiation de diverses leucémies. Comprendre le fonctionnement des gènes Hox dans l'hématopoïèse normale est donc une condition préalable pour élucider leurs fonctions dans les leucémies, ce qui pourrait éventuellement conduire à l’élaboration de nouveaux traitements contre cette maladie. Plusieurs études ont tenté d’élucider les rôles exacts des gènes Hox dans l'hématopoïèse via l’utilisation de souris mutantes pour un seul gène Hox. Or, en raison du phénomène de redondance fonctionnelle chez cette famille de gènes, ces études ont été peu concluantes. Il a été précédemment démontré que dans une population de cellules enrichies en cellules souches hématopoïétiques (CSH), les gènes du cluster Hoxa sont plus exprimés que les gènes Hox des autres clusters. Aussi, il a été établi que les gènes du cluster Hoxb sont non essentiels à l’hématopoïèse définitive puisque les CSH mutantes pour les gènes Hoxb1-9 conservent leur potentiel de reconstitution à long terme. En nous basant sur ces données, nous avons émis l'hypothèse suivante : les gènes Hoxa sont essentiels pour l'hématopoïèse normale adulte. Pour tester notre hypothèse, nous avons choisi d’utiliser un modèle de souris comportant une délétion pour l’ensemble des gènes Hoxa. Dans le cadre de cette recherche, nous avons démontré que les CSH, les progéniteurs primitifs et les progéniteurs des cellules B sont particulièrement sensibles au niveau d'expression des gènes Hoxa. Plus particulièrement, une baisse de la survie et une différenciation prématurée semblent être à l’origine de la perte des CSH Hoxa-/- dans la moelle osseuse. L’analyse du profil transcriptionnel des CSH par séquençage de l'ARN a révélé que les gènes Hoxa sont capables de réguler un vaste réseau de gènes impliqués dans divers processus biologiques. En effet, les gènes Hoxa régulent l’expression de plusieurs gènes codant pour des récepteurs de cytokine. De plus, les gènes Hoxa influencent l’expression de gènes jouant une fonction dans l’architecture de la niche hématopoïétique. L’expression de plusieurs molécules d’adhésion est aussi modulée par les gènes Hoxa, ce qui peut affecter la relation des CSH avec la niche hématopoïétique. L’ensemble de ces résultats démontre que les gènes Hoxa sont d'importants régulateurs de l'hématopoïèse adulte puisqu’ils sont nécessaires au maintien des CSH et des progéniteurs grâce à leurs effets sur plusieurs processus biologiques comme l'apoptose, le cycle cellulaire et les interactions avec la niche. / In humans, a large percentage of myeloid and lymphoid leukemias exhibit aberrant Homeobox (Hox) genes expression, predominantly Hoxa genes. This aberrant expression is known to be caused by either translocations involving Hox genes or indirect activation of Hox genes. In addition, evidence now indicates a critical role for Hox genes in the initiation of leukemias. Clearly, understanding how Hox genes function in normal hematopoiesis is prerequisite to elucidate their involvement in leukemogenesis and this may eventually lead to new treatments for this disease. Attempts to determine the precise role(s) of Hox genes in normal hematopoiesis using single gene loss of function mutants have shown little success due to functional complementation by the remaining Hox genes. We previously showed that the Hoxa genes are much higher expressed in enriched hematopoietic stem cell (HSC) populations than the other members of the Hox gene family. Moreover, Hoxb cluster genes were found to be dispensable for HSCs long-term repopulation of irradiated mice. Thus, we hypothesize that Hoxa genes are critical for normal adult hematopoiesis. We have used a multi-gene knockout (KO for the entire Hoxa cluster) approach to thoroughly evaluate this issue. In this thesis, we showed that HSC, primitive progenitors and B cell progenitors are particularly sensitive to the levels of Hoxa gene expression. Furthermore, a lower survival and a premature differentiation account for the loss HSC Hoxa-/- in bone marrow. Differential expression profiling by RNASeq revealed that Hoxa genes are capable of regulating a broad array of genes involved in various biological processes. Indeed, Hoxa genes regulate the expression of several genes coding for cytokine receptors. Furthermore, Hoxa genes modulate the expression of genes implicated in the regulation and formation of the niche architecture. The expression of several adhesion molecules is also modulated by the Hoxa genes, which can affect the relationship of HSC with the hematopoietic niche. Through their action on several biological processes such as apoptosis, cell cycle and niche interactions, Hoxa genes are necessary for maintenance of HSC and progenitors. Taken together, these results demonstrate that Hoxa genes are important regulators of adult hematopoiesis.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

La leucémie lymphoïde représente environ 30% des cas de cancer chez l’enfant. Elle est souvent causée par des réarrangements chromosomiques impliquant des gènes encodant des facteurs de transcription, qui contrôlent des programmes génétiques complexes. Par exemple, LMO2 (LIM-only 2) est un facteur de transcription oncogénique fréquemment exprimé de façon aberrante dans les leucémies lymphoblastiques aigues des cellules T (T-ALL). Dans l’hématopoïèse normale, LMO2 est essentiel à la génération des cellules souches hématopoïétiques à l’origine de toutes les cellules sanguines. D’ailleurs, certaines cellules leucémiques possèdent des propriétés normalement réservées aux cellules souches hématopoïétiques. Ainsi, l’étude de la fonction de LMO2 dans les cellules souches hématopoïétiques peut être pertinente autant dans le contexte hématopoïétique normal que leucémique. Afin de mettre en évidence de nouvelles fonctions moléculaires pour LMO2, j’ai choisi d’identifier les protéines qui s’y associent. En plus de ses partenaires connus, j’ai identifié plusieurs protéines de transcription/remodelage de la chromatine, en accord avec son rôle transcriptionnel. Plusieurs nouvelles fonctions potentielles ont été révélées, indiquant que cette protéine adaptatrice pourrait faire partie de complexes non transcriptionnels, régulant d’autres processus cellulaires. Les oncogènes comme LMO2 pourraient être des régulateurs à large spectre. Particulièrement, j’ai identifié des interactions entre LMO2 et des protéines de réplication de l’ADN. J’ai montré que LMO2 contrôle la réplication de l’ADN dans les cellules hématopoïétiques, et possiblement durant la leucémogenèse, indépendamment de son rôle transcriptionnel. Ensemble, ces études ont donc permis de révéler de nouvelles fonctions pour LMO2, et pourraient servir de paradigme pour d’autres facteurs de transcription oncogéniques, particulièrement aux autres protéines de la famille LMO, qui sont aussi des oncogènes puissants. / Lymphoid leukemia represents about 30% of childhood cancer cases. It is often caused by chromosomal rearrangements involving genes coding for transcription factors, controlling complex genetic programs. As an example, the oncogenic transcription factor LMO2 (LIM-only 2) is often aberrantly expressed in T cell acute lymphoblastic leukemia (T-ALL). In normal hematopoiesis, LMO2 is essential for the generation of hematopoietic stem cells that give rise to all blood cells. Moreover, some leukemic cells possess properties normally reserved to hematopoietic stem cells. Thus, studying the role of LMO2 in hematopoietic stem cells could be relevant to the contexts of normal hematopoiesis and leukemogenesis. To reveal new molecular functions for LMO2, I chose to identify its associated proteins. In addition to its known protein partners, I identified many proteins involved in transcription/chromatin remodeling, in agreement with its transcriptional role. In addition, several new potential functions have been revealed, indicating that this scaffold protein could be part of non-transcriptional protein complexes, regulating different cell processes. Oncogenes like LMO2 could be master regulators in normal hematopoietic and leukemic cells. Particularly, I identified protein-protein interactions between LMO2 and DNA replication proteins. I demonstrated that LMO2 controls S phase progression in hematopoietic cells, independently of its association in transcriptional complexes. LMO2 overexpression in mice induces T-ALL and affects specifically the cell cycle status of thymocyte progenitors, which are targets of transformation by LMO2. Thus, LMO2 promotes DNA replication in hematopoietic cells, and possibly in leukemogenesis. Together, these studies allowed to reveal new functions for LMO2, and could serve as a paradigm for other oncogenic transcription factors, especially for other LMO proteins which are all potent oncogenes.

Veränderungen in der lokalen Chromatinstruktur beeinflussen die dynamische Regulation von Genen, welche für die Differenzierung notwendig sind. PU.1 ist ein Master-Transkriptionsfaktor in der Hämatopoese und wird streng reguliert, um ein zelllinienspezifisches Expressionsmuster zu erzielen. Hohe Konzentrationen von PU.1 sind für myeloische Differenzierung erforderlich. In B-Zellen wird PU.1 mittelstark exprimiert und muss aktiv runterreguliert werden, um eine Ausdifferenzierung der multipotenten Vorläuferzellen zu T-Zellen zu ermöglichen. Derzeit ist wenig über die Regulierung von PU.1 in T-Zellen bekannt. Darüber hinaus wurde eine abnormale Expression von PU.1 in verschiedenen Leukämieerkrankungen beobachtet. Mittels eines genome-wide Chromatin-Interaktions-Screens konnten wir einen cis-Repressor mit insulierender Kapazität identifizieren, welcher mittels eines Chromatinloops die Promotoraktivität von PU.1 in T-Zellen, jedoch nicht in myeloischen oder B-Zellen blockiert. Sowie Looping als auch Insulation erfordern die Bindung des Chromatin-Regulatorprotein CTCF. Im Gegensatz zu normalen myeloischen Zellen finden wir, dass Krebszellen aus myeloischen Leukämie Patienten diese T-Zell-spezifische repressive Chromatinstruktur aufweisen, was einen räumlichen Kontakt des Insulator mit dem PU.1 Promotor ermöglicht. Die Ergebnisse dieser Arbeit beschrieben das CTCF gesteuerte „long distance looping“ als ein neuer molekularer epigenetischer Mechanismus, um Transkriptionsfaktor PU.1 in T-Zellen runterzuregulieren, und zeigen zum ersten mal, dass Krebszellen die Chromatinstruktur anderer Zelllinien imitieren können, um die Expression von Differenzierungsgenen zu blockieren. / Alterations in the local chromatin structure orchestrate the dynamic regulation of differentiation promoting genes. PU.1 is a master transcription factor in hematopoiesis. PU.1 gene must be tightly regulated to achieve lineage specific expression pattern. High levels of PU.1 are required for myeloid commitment: it is expressed at intermediate level in B-cells and must be actively silenced to permit T cell development from early multipotent progenitors. However, little is known of how PU.1 is regulated in T-cells. Moreover, aberrant PU.1 expressions have been observed in multiple leukemias. Using a genome-wide chromatin interaction screen we identified a cis-repressor with insulating capacity that undergoes long-distant chromatin looping to block PU.1 promoter activity in T cells but not myeloid or B cells. Looping and repression requires binding of the chromatin regulator protein CTCF. In contrast to normal myeloid cells, we found that cancer cells from myeloid leukemia patients adopt the T cell specific repressive chromatin structure bringing the insulator into spatial contact with the PU.1 promoter. These results identify CTCF controlled long-distant insulator looping as a novel mechanism to silence lineage-opposing transcription factor expression, and reveal that cancer cells can mimic the chromatin confirmation of another lineage to block expression of differentiation driving genes.

Das Eukaryotische-Initiations faktor-4E Bindungsprotein (4E-BP) ist ein Inhibitor der Translationsinitiation. Nicht-phosphoryliertes 4E-BP bindet an den eukaryotischen Initiationsfaktor 4E (eIF4E). Diese Bindung blockiert die Rekrutierung des Initiationskomplexes eIF4F an die Cap-Struktur des 5´Endes von eukaryotischen zellulären mRNAs, was die Initiation der Translation verhindert. Phosphorylierung von 4E-BP durch die mTOR Kinase führt zur Dissoziation des 4E-BP/eIF4E Komplexes und erhöht die Verfügbarkeit von eIF4E, dies wird mit Zellproliferation assoziiert. Die Aktivität von eIF4E wird nicht nur von 4E-BP, sondern auch durch Phosporylierung reguliert, welche wiederum durch die "MAP-Kinase-Interacting-Protein-Kinase" (MNK) reguliert wird. Drei Isoformen von 4E-BP sind bekannt: 4E-BP1, 4E-BP2 and 4E-BP3. 4E-BP1 und 4E-BP2 sind an oxidativem und adipogenetischen Stress beteiligt. Beide Proteine werden im h?matopoetischen System gleich exprimiert, wohingegen 4E-BP3 nicht detektiert wird. 4E-BP1 wird während der Erythroblasten-Proliferation phosphoryliert. Aus diesem Grund habe ich die Hämatopoese und die durch Phenylhydrazine (PHZ) induzierte Stress-Erythropoese in 4E-BP1 und 4E-BP2 Knock-Out Mäusen und 4E-BP1,2 Doppel-Knock-Out Mäusen analysiert. Ich konnte zeigen, dass die Hämatopoese in 4E-BPs defizienten Mäusen nicht beeinflusst wird. Allerdings zeigten 4E-BP1,2-/- und 4E-BP2-/- Mäuse eine verspätete Antwort auf Phenylhydrazin (PHZ) induzierten erythropoetischen Stress. Gleichzeitig war die mRNA Translation von GATA-1, ein essentieller erythropoetischer Transkriptionsfaktor in Erythroblasten runterreguliert. Die Signaltransduktionswege mTOR und MNK1 waren bei erythropoetischen Stress aktiviert. Diese Daten zeigen, dass 4E-BP2, aber nicht 4E-BP1, notwendig ist um auf erythropoetischen Stress zu reagieren und deuten an, dass die 4E-BP gesteuerte translations-regulierende Maschinerie eine Rolle in der Stress-Erythropoese spielt. / Translational regulation allows an organism to generate fast responses to environmental changes quickly. Eukaryotic initiation factor 4E binding protein (4E-BP) is an inhibitor of translation initiation. Unphosphorylated 4E-BP binds to eukaryotic initiation factor 4E (eIF4E) blocking recruitment of the initiation complex eIF4F to the cap structure at the 5´ terminus of eukaryotic cellular mRNAs. Thus initiation of translation is blocked. Phosphorylation of 4E-BP by the mTOR kinase causes disassociation of the 4E-BP/eIF4E complex and increases the availability of eIF4E. EIF4E activity is not only regulated by 4E-BP, but also phosphorylation which is regulated by MAP kinase - interacting protein kinase (MNK). Three isoforms of 4E-BP are known, termed 4E-BP1, 4E-BP2 and 4E-BP3. 4E-BP1 and 4E-BP2 are involved in oxidative and adipogenetic stresses in vivo. They are equally expressed in hematopoietic system, whereas 4E-BP3 is not detected. 4E-BP1 is phosphorylated during erythroblast proliferation. Erythroid differentiation is blocked by overexpresssion of eIF4E in tissue culture. These studies implied that 4E-BPs might play role in response to erythropoietic stress. I examined hematopoiesis and phenylhydrazine (PHZ) induced stress erythropoiesis in 4E-BP1 and 4E-BP2 individual knock out mice and 4E-BP1,2 compound knock out mice. I found that the hematopoiesis of 4E-BPs deficient mice were unaffected. However, 4E-BP1,2-/- and 4E-BP2-/- mice showed delayed response to phenylhydrazine (PHZ) induced erythropoietic stress. Simultaneously, the mRNA translation of GATA-1, which is the essential erythroid transcription factor, was downregulated in their erythroblasts. The signaling pathways through the mTOR and MNK1 were activated in erythropoietic stress. These data showed that 4E-BP2 but not 4E-BP1 was required for the response to erythropoietic stress and suggested that 4E-BP related translation regulatory machinery played a role in stress erythropoiesis.

Der CCAAT enhancer binding protein beta (C/EBPβ) Transkriptionsfaktor reguliert die Differenzierung, Proliferation und Funktion vieler Zelltypen, einschließlich verschiedener Zellen des Immunsystems. Eine detaillierte molekulare Analyse des Mechanismus, wie C/EBPβ alternative Zellschicksale steuert, wurde jedoch bisher noch nicht unternommen. Es wurde gezeigt, dass die ektopische Expression von C/EBPβ in determinierten B- Vorläuferzellen diese zu inflammatorischen Makrophagen reprogrammieren kann. Wir haben dieses Reprogrammierungsystem verwendet, um die Strukturelemente in C/EBPβ, die für die Regulation der (Trans)Differenzierung durch C/EBPβ wichtig sind, zu untersuchen. Um die maßgeblichen C/EBPβ Proteinmodule für die Reprogrammierung zu bestimmen, wurden entweder C/EBPβ Wildtyp Isoformen oder Mutanten in primären murinen B-Vorläuferzellen ektopisch exprimiert. Die Analysen ergaben, dass die translational regulierten langen Isoformen LAP* and LAP, jedoch nicht die kurze Isoform LIP lymphoide Zellen zu myeloischen Zellen reprogrammieren können. Des weiteren haben wir gezeigt, dass die konservierten Regionen 2, 3 und 4 der C/EBPβ Transaktivierungsdomäne essentiell und ausreichend für die Konvertierung von B Zellen zu myeloischen Zellen sind. Die reprogrammierten myeloischen Zellen setzten sich aus einer heterogenen Population verschiedener myeloischer Zelltypen zusammen. Detaillierte Analysen von CD11b+ reprogrammierten Zellen zeigten, dass diskrete konservierte Regionen von C/EBPβ verschiedene pro- und anti-inflammatorische Gene und divergente Entwicklungsprogramme aktivierten. Des Weiteren führten nicht nur strukturelle C/EBPβ Mutanten sondern auch Puktmutationen an Stellen, die posttranslationalen Modifikationen (PTM) unterliegen, zu verschiedenen Reprogrammierungsergebnissen. Diese Daten zeigen, dass die C/EBPβ abhängige myeloische Diversifikation durch die Integration von strukturellen C/EBPβ Proteinmodulen und deren signalabhängigen PTMs erreicht wird. / The CCAAT enhancer binding protein beta (C/EBPβ) transcription factor regulates differentiation, proliferation, and functionality of many cell types, including various cells of the immune system. A detailed molecular understanding of how C/EBPβ directs alternative cell fates remains largely elusive. Ectopic expression of C/EBPβ has been previously shown to reprogram committed B cell progenitors into inflammatory macrophages. We took advantage of this reprogramming system in order to examine how C/EBPβ regulates (trans)differentiation. To determine which C/EBPβ protein modules are important for reprogramming, C/EBPβ wild type isoforms and mutants were ectopically expressed in primary mouse B cell progenitors. The data showed that the translationally regulated long isoforms LAP* and LAP, but not the N-terminally truncated isoform LIP can reprogram lymphoid cells into myeloid cells. Furthermore, we found that conserved regions 2,3 and 4 in the C/EBPβ protein transactivation domain are necessary and sufficient for B-to-myeloid cell conversion. Interestingly, the reprogrammed myeloid cells were found to represent a heterogeneous mixture of different myeloid cell types. Detailed analyses of the reprogrammed CD11b+ cells revealed that discrete conserved regions in C/EBPβ activated distinct pro- and anti-inflammatory genes and triggered divergent differentiation programs. Moreover, not only structural C/EBPβ mutants, but also post-translational modification (PTM) site mutations led to different reprogramming outcomes. These data suggest that C/EBPβ orchestrates myeloid diversification by integrating PTMs with structural plasticity as signal dependent adaptable modular properties to determine cell fate.

A desnutrição proteico-energética (DPE) provoca anemia e leucopenia decorrente da redução de precursores hematopoéticos e comprometimento da produção de mediadores indutores da hematopoese, bem como alterações estruturais e ultra-estruturais na matriz extracelular medular. A hematopoese ocorre em nichos medulares distintos - endosteal e perivascular - que modulam os processos de diferenciação, proliferação e auto-renovação da célula tronco hematopoética (CTH). As células tronco mesenquimais (CTM) tem um papel importante na formação destes nichos, através da sua diferenciação nos diversos tipos celulares que os compõe. Adicionalmente, a CTM pode modular a função de outras células, como a CTH e a célula endotelial (CE) medular, através da liberação de diversos fatores de crescimento e citocinas. As CE expressam proteínas que regulam a diferenciação e movimentação das CTH na MO. Há sinais que a CTM pode ser a precursora da CE medulares, pois in vitro a CTM pode se diferenciar em CE-like. Desta forma, a CTM é um ponto chave no estudo das alterações causadas pela DPE no nicho perivascular e sobre a regulação da hematopoese. Neste trabalho, investigamos se a DPE afeta a diferenciação in vitro da CTM medular em CE-like e avaliamos se essas células apresentam diferentes capacidades em produzir alguns mediadores regulatórios da hematopoese (CXCL-12, SCF, Ang-1, IL-11, GM-CSF e TFG-β), bem como possíveis alterações no perfil de expressão gênica de marcadores de função das CTM e CE-like. Utilizamos camundongos C57BL/6 machos, divididos em grupos Controle e Desnutrido, sendo que o grupo Controle recebeu ração normoprotéica (12% caseína) e o grupo Desnutrido recebeu ração hipoprotéica (2% caseína), ambos durante 5 semanas. Após este período, os animais foram eutanasiados, foi realizada a avaliação nutricional e hematológica, caracterizando a DPE. As CTM foram isoladas, caracterizadas e diferenciadas in vitro em CE-like, o que foi evidenciado pela maior expressão gênica de NT5E, FLT1, KDR, PECAM1 e VCAM1. Avaliamos a expressão dos genes CDH5, CSPG4, LEPR, NES, CSF1, CSF2, CSF3, MCAM, PROM1, ANGPT1, CXCL12, ENG, IGF1, IL3, IL11, KITL, TGFB1, WNT3A, WNT5A, ICAM1, PDGFB1 e VWF. Encontramos alterações causadas pela DPE na expressão gênica e quantificação de CXCL-12, SCF e Ang-1, os quais mostraram que as células avaliadas do grupo Desnutrido encontram-se em um estado \"pró-proliferativo\", em um esforço para restabelecer a hematopoese na DPE. Entretanto, foi observado neste trabalho e nos demais trabalhos do grupo que há hipoplasia medular na DPE e, portanto, pode-se inferir que as alterações hematopoéticas observadas na DPE não são ocasionadas por alterações na síntese de SCF, CXCL-12 ou Ang-1. / Protein-energy malnutrition (PEM) causes anemia and leukopenia as it reduces hematopoietic precursors, impairs the production of mediators that induce hematopoiesis and alters structural and ultrastructural changes in bone marrow (BM) extracellular matrix. Hematopoiesis occurs in distinct BM niches - endosteal and perivascular - which modulate the processes of differentiation, proliferation and self-renewal of hematopoietic stem cell (HSC). Mesenchymal stem cells (MSC) play an important role in the formation of these niches through their differentiation in several cell types that compose them. Additionally, MSC can modulate the function of other cells, such as HSC and endothelial cells (EC), through the release of several growth factors and cytokines. The EC express proteins that regulate the differentiation and migration of HSC in the BM. MSC seem to be the precursor of medullary EC because in vitro MSC can differentiate into EC-like cells. Thus, MSC are a key point in the study of changes caused by DPE on the perivascular niche and on the regulation of hematopoiesis. In this study, we investigated whether PEM would affect BM-MSC in vitro differentiation into EC-like cells and evaluated whether these cells would have distinct capacities of producing some regulatory mediators of hematopoiesis (CXCL- 12, SCF, Ang-1, IL-11, GM -CSF and TFG-β), as well as analyzed possible changes in the gene expression profile of MSC function and EC-like cells related markers. C57BL/6 mice were divided into Control and Malnourished groups, which received for 5 weeks, respectively, a normal protein diet (12% casein) and a low protein diet (2% casein). After this period, animals were euthanized, nutritional and hematological evaluations were performed, featuring the PEM. MSC were isolated, characterized and differentiated in vitro into EC-like cells, which were evidenced by increased gene expression of NT5E, FLT1, KDR, PECAM1 and VCAM1. The expression of CDH5, CSPG4, LEPR, NES, CSF1, CSF2, CSF3, MCAM, PROM1, ANGPT1, CXCL12, ENG, IGF1, IL3, IL11, KITL, TGFB1, Wnt3a, WNT5A, ICAM1, PDGFB1 and VWF genes was also evaluated. Changes caused by PEM on gene expression and quantification of CXCL-12, SCF and Ang-1 were found, indicating that tested cells from the Malnourished group were in a \"pro-proliferative\" state in an effort to restore hematopoiesis. However, our results are in accordance to the literature regarding bone marrow hypoplasia as a consequence of PEM. Therefore, we infer hematopoietic changes observed in this work are not related to changes in the synthesis of SCF, 12 CXCL-12 or Ang-1.