Thesis advisor: Thomas Chiles / Thesis advisor: James Kim / The purpose of this thesis is to describe some of the research I conducted in Dr. Markmann’s laboratory at Massachusetts General Hospital in the field of transplantation immunology. The first portion provides background information on the immune system and its different components, eventually providing an in-depth look at regulatory T cells, and their role in transplant immunology. The second portion of the thesis is dedicated to my experiments. This part presents the materials and methods used, the previous findings that led to my experiments, the data analysis and results, and a conclusive discussion. The research I did that is included in this thesis pertains to regulatory T cells. It is believed that a potential pathway to prolonging allograft survival is to drive antigen-specific CD4+Foxp3- T cells to convert to CD4+Foxp3+ T cells. I studied the effects of retinoic acid (RA) and anti-CD45RB treatment on Foxp3 expression in CD4+ T cells. These studies were done using different lines of transgenic mice that provided models of antigen-specificity. My in vitro data demonstrated that RA is capable of boosting TGF-β-mediated Foxp3 upregulation in a synergistic manner and that anti-CD45RB is also able to increase Foxp3 expression. In vivo RA experiments were inconclusive, and due to some misfortune and time constraints in vivo anti-CD45RB experiments were not conducted. Research funded by the following sources: 2R56AI048820, 5R01AI057851, and 5K01DK079207. / Thesis (BS) — Boston College, 2011. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: College Honors Program. / Discipline: Biology.

O ácido retinóico (AR) é sintetizado a partir de oxidações sucessivas do retinol. A última etapa de oxidação é catalisada pelas enzimas RALDHs. O estabelecimento da polaridade ântero-posterior (AP) do coração é crítico para a demarcação das regiões de efluxo (ventrículos) e de influxo (átrios). Nosso grupo propõe que uma onda caudo-rostral (CR) de RALDH2 seja o mecanismo responsável por esta padronização em vertebrados. Para testar o papel da sinalização pelo AR na padronização AP do peixe zebra, manipulamos sua via com o inibidor das enzimas RALDHs, DEAB, com AR e com um inibidor da enzima CYP26 (IC), que cataboliza o AR. Os tratamentos com DEAB durante o período da onda de RALDH2 produziram átrios reduzidos, enquanto que os tratados com AR e com IC apresentam o domínio atrial expandido. Animais tratados com DEAB e AR em um estágio posterior a onda não mostraram diferença significativa ao controle. Estes dados sugerem uma forte correlação entre o evento da onda e a padronização das câmaras cardíacas, semelhante ao que ocorre nos demais amniotos. / The retinoic acid (RA) is synthetized by successive oxidations of retinol. The last oxidation step is catalyzed by the RALDHs enzymes. The establishment of the anteroposterior (AP) polarity is critical for the demarcation of outflow (ventricle) and inflow (atrium) regions. Our group proposes that a caudorostral (CR) wave of RALDH2 is the mechanism responsible for this patterning in vertebrates. In order to test the role of the RA signaling in zebrafish AP patterning, we manipulated its pathway with a RALDH enzymatic inhibitor, DEAB, with RA and with a CYP26 (IC) enzymatic inhibitor, that catabolises the RA. The DEAB treatments during the manifestation of the RALDH wave produced reduced atriums; meanwhile, the treatments with RA and IC presented an atrium expansion. Animals treated with DEAB and RA during a stage posterior to the wave did not present any significant difference. These data suggest a strong correlation between the wave event and the cardiac chamber patterning, similar to the mechanism observed in others amniotes.

Os gliomas representam o tipo de tumor primário mais comum do sistema nervoso central. Apesar dos avanços nas técnicas cirúrgicas e de radioterapia e nos protocolos de quimioterapia, não há tratamento eficiente disponível. O ácido retinóico na forma all-trans (atRA) é um agente anti-proliferativo utilizado para o tratamento de alguns tipos de lesões pré-malígnas e tumores, como a leucemia promielocítica. Entretanto, sua utilização no tratamento clínico de tumores sólidos, incluindo os gliomas, é limitada, devido ao rápido desenvolvimento de resistência aos efeitos anti-tumorais do atRA. Para endereçar este problema, foi proposto: a) clonar e analisar a função de cyp26b1 de rato, uma enzima da família dos citocromos P450 envolvida na metabolização de ácido retinóico, previamente descrito em nosso laboratório como potencialmente envolvido na resposta de células de glioma ao tratamento com atRA; b) gerar e caracterizar uma variante da linhagem celular C6 de glioma de rato resistente aos efeitos anti proliferativos de atRA. O cDNA de cyp26b1 de rato, até então desconhecido, foi amplificado, clonado e seqüenciado com sucesso. A seqüência protéica correspondente é conservada e agrupa-se no dendrograma com outras proteínas Cyp26B1, mesmo de organismos distantes como zebrafish, em detrimento de outras proteínas parálogas da família Cyp26. Cyp26b1 é fortemente induzida por atRA em células de glioma de rato e humano, de forma dose-dependente. Embora seja expresso em amostras de cérebros humanos normais e de glioma, não foram encontradas diferenças significativas entre os diferentes graus de matlignidade tumoral, ou em comparação com cérebro normal. A super-expressão de cyp26b1 de rato através de transfecção estável de células de glioma de rato e humano, bem como de células P19 de teratocarcinoma de camundongo, não alterou significativamente as características de crescimento celular, tanto em substrato sólido quanto em substrato semi-sólido. O tratamento prolongado de células C6 de glioma de rato com atRA levou à obtenção de uma população policlonal resistente aos efeitos anti-proliferativos de atRA, a partir da qual uma linhagem celular clonal resistente foi isolada com sucesso. Esta variante clonal, denominada linhagem celular C6R, apresenta inibição de crescimento por atRA diminuída, quando comparada com a linhagem celular C6 parental e com a variante ST1. Além disso, esta variante também mostrou uma tendência, embora não significativa estatisticamente, de gerar tumores maiores quando injetados no subcutâneo de camundongos do tipo nude. A expressão de genes previamente descritos como induzidos pelo tratamento com atRA em células ST1 é fortemente inibida na linhagem celular C6R. Além disso, menor expressão de RARβ, RARγ e CRABP-1 foi também observada em células C6R, enquanto que expressão muito maior de RXRγ e CRABP-2 foi encontrada nas células resistentes a atRA, em comparação com as células C6 parentais. Como conclusões gerais deste trabalho, foi proposto que cyp26b1 parece estar envolvido no metabolismo e detoxificação de atRA e parece não ser efetor da inibição de crescimento induzida por atRA, nem estar relacionado com a resistência de células de glioma ao tratamento com atRA. Por outro lado, o isolamento e caracterização de uma linhagem celular resistente ao tratamento com atRA sugere que a resistência está relacionada à expressão diferencial de receptores de retinóides e de proteínas de ligação ao ácido retinóico. / Gliomas are among the most common primary tumors of the central nervous system. Despite the advances in surgical and radiotherapy procedures and chemotherapy protocols, efficient treatment is still not available. Retinoic acid is a anti-proliferative agent used for treatment of a number of pre-malignant lesions and tumors, as promyelocytic leukemia. However, its clinical use in treatment of solid tumors, including gliomas, is impaired by the rapid development of resistance to the anti-tumoral effects of atRA. In order to address this problem, we proposed: a) to clone and analyze the role of rat cyp26b1, a member of cytochrome P450 superfamily envolved in the metabolization of retinoic acid, which has previously been described by our group as being potentially involved in the response of glioma cells to retinoic acid treatment; b) to generate and characterize a variant of rat C6 glioma cell line resistant to anti proliferative effects of atRA. The previously unknown cDNA of rat cyp26b1 was successfully amplified, cloned and sequenced. The protein is conserved and clustered in dendrograms with other cyp26b1 proteins, even from distant organisms as zebrafish, in detriment of other cyp26 paralogs. Cyp26b1 is strongly induced by atRA in rat and human glioma cells, in a dose-dependent fashion. Although expressed in human normal brains and glioma samples, no significant differences were found among different tumor grades nor comparing to normal brain. Stable-transfection and overexpression of rat cyp26b1 in rat and human glioma cell lines, as well as P19 mouse teratocarcinoma cell line, did not significantly modified cell growth properties, either on solid or semi-solid substrates. Prolonged treatment of C6 rat glioma cell line with atRA leads to isolation of an atRA-resistant polyclonal cell population, from which a resistant clonal cell line was successfully isolated. This clonal variant, named C6R cell line, displayed diminished growth inhibition by atRA compared to the parental C6 cell line and the variant ST1 cell line. In addition, this variant also showed a trend, although not quite statistically significant, to generate larger tumors when xenografted s.c. into nude mice. Expression of genes previously described as being induced by atRA-treatment in ST1 cells is strongly impaired in the C6R resistant cell line. In addition, lower expression of RARβ, RAR&#947 and CRABP-1 was also observed in C6R cells, whereas a much higher expression of RXRγ and CRABP-2 was found in the resistant cells when compared to the parental C6 cells. As general conclusions of this work, we found that cyp26b1 is more likely to be involved in primary atRA metabolism and detoxification and does not seem to be an effector of atRA-induced cell growth arrest nor to be related to the resistance of glioma cells to atRA treatment. On the other hand, isolation and characterization of an atRA-resistant cell line suggests that atRA resistance is more likely to be due to differential expression of retinoid receptors and retinoic acid binding proteins.

Les cellules souches embryonnaires (ES) murines peuvent être différenciées in vitro en une population homogène de neurones glutamatergiques semblables aux neurones présents dans le cortex in vivo, suite à un traitement par l’acide rétinoïque (AR). Bien que le rôle de l’AR soit bien étudié, les facteurs qui spécifient le destin neuronal ne sont pas connus. Nous montrons ici que Pou3f2 (Brn2) est un facteur essentiel à la différenciation neuronale des cellules ES in vitro. L’utilisation de l’approche de différenciation in vitro associée à des techniques de génomique à haut débit (RNA-seq, ChIP-seq) a permis d’identifier des gènes régulés directement ou indirectement par Brn2. Parmi ces gènes se trouvent Ascl1, Hes5 ou Pou6f1, qui sont des gènes clés dans la neurogenèse. La comparaison de nos données avec des expériences précédemment publiées nous a permis d’identifier un nombre restreint de gènes cibles de Brn2 quelque soit le protocole de différenciation utilisé. Parmi ces gènes se trouve Zic1. Nous montrons que Zic1 coopère avec Brn2 pour spécifier le destin neuronal des cellules ES in vitro. / Mouse embryonic stem (ES) cells can be differentiated in vitro into a highly homogenous population of glutamatergic neurons, similar to those present in the cerebellar cortex by treatment with retinoic acid (RA). While the role of RA in differentiation is well studied, the downstream factors that specify the neural fate of the ES cells are not known. Here we show that Pou3f2 (Brn2), with a known role in neuronal differentiation in vivo, is essential for neuronal differentiation of ES cells in vitro. Using our in vitro differentiation protocol combined with high throughput techniques (RNA-seq, ChIP-seq) we show that Brn2 directly and indirectly regulates a set of target genes with essential roles in neurogenesis such as Ascl1, Hes5 or Pou6f1. Integration of these results with previously published datasets allowed us to identify a core set of Brn2 target genes common to each differentiation model. Amongst these is transcription factor Zic1. We show that Zic1 and Brn2 cooperate to specify the neural fate of RA-treated ES cells in vitro.

Cancer immunotherapy is an approach to cancer therapy that involves the enhancement of the cancer patient’s own innate and/or adaptive immune systems to attack their own cancer. Clinically available cancer immunotherapies rely on different strategies: infusion of ex vivo manipulated autologous dendritic cells (DCs), infusion of genetically engineered autologous cytotoxic CD8+ T lymphocytes, stimulation of T lymphocyte proliferation, or inhibition of immunosuppressive pathways to improve T lymphocyte effector function. Nonetheless, only a small percentage of cancer patients receive benefit from immunotherapies and thus further improvements in clinical outcomes are required. Among numerous other therapeutic immunotherapies strategies being developed and tested, adenovirus serotype 5-based vectors (Ad5) have been well studied in preclinical and clinical settings. Preclinical research has shown that vaccination of mice with Ad5-OVA (an Ad5 encoding a model tumor antigen, chicken ovalbumin (OVA)) results in activation and proliferation of OVA-specific CD8+ T lymphocytes capable of specific killing of tumor cells that express OVA. This dissertation evaluates the potential of polymeric nanoparticles (NP) loaded with all-trans retinoic acid (ATRA), a vitamin A derivative with potent immunostimulatory effects, to improve the immunostimulatory and therapeutic effects of Ad5-OVA in a murine E.G7-OVA tumor model, a well described model that can be used for studying the immune response to Ad5-based immunotherapies. In the first part of this work, poly(lactide-co-glycolide) NP loaded with ATRA (ATRA-PLGA-NP) were prepared and characterized. Next, the antitumor effect and the magnitude of the OVA-specific immune response due to Ad5-OVA vaccination versus ATRA-PLGA-NP (or ATRA soluble) plus Ad5-OVA combination treatment were compared in vivo. The results showed that the combination treatment using ATRA-NP, but not ATRA soluble, resulted in enhanced survival and enhanced levels of OVA-specific CD8+ T lymphocytes in peripheral blood, spleen, and tumor. Next, cRGD- and mannose-functionalized PLGA-PEG NP were developed in an attempt to actively target the tumor neovasculature and DC-rich organs, respectively. The functionalization efficacy was confirmed by ex vivo fluorescence imaging studies. In vivo studies using E.G7-OVA-challenged mice showed that treatment with ATRA-loaded cRGD-functionalized PLGA-PEG-NP + Ad5-OVA, despite not enhancing the levels of OVA-CD8+ T lymphocytes in peripheral blood, substantially enhanced survival compared to either the combination of Ad5-OVA + non-functionalized ATRA-PLGA-PEG-NP or Ad5-OVA + conventional ATRA-PLGA-NP. On the contrary, treatment with mannose-functionalized PLGA-PEG-NP + Ad5-OVA, despite optimally enhancing the levels of OVA-CD8+ T lymphocytes in peripheral blood (compared to all other treatment groups), did not lead to enhanced survival compared to either the combination of Ad5-OVA + non-functionalized ATRA-PLGA-PEG-NP, Ad5-OVA + conventional ATRA-PLGA-NP, and over Ad5-OVA treatment alone. Although not investigated further in this dissertation, it was speculated that the observed trend in survival benefit provided by ATRA-PLGA-PEG-cRGD-NP + Ad5-OVA over the other NP formulations may have been due to higher levels of ATRA within the TME due to actively targeting the tumor vasculature, corroborating previous studies which demonstrated that ATRA functions as a potent stimulator of anti-tumor cellular immune responses within the tumor. The paradoxical results obtained with mannose-functionalized PLGA-PEG-NP are less readily explained. In conclusion, it was demonstrated in this work that the co-administration of Ad5-OVA and ATRA-loaded NP formulations enhanced the tumor specific cellular immune response and the survival of tumor challenged mice compared to vaccination with Ad5-OVA alone.

Murine neural precursor cells have been a well studied model for neural cell fate determination and stem cell function both in vivo and in primary culture. However, factors such as cell number, the presence of multiple cell populations and of niche intrinsic factors made it difficult to dissect the mechanisms regulating cortical development. To overcome this issue, we have developed a culture system where mouse embryonic stem cells (ES) are differentiated to cortical radial precursors through retinoic acid treatment of embryoid bodies. One day after plating in neural differentiation conditions, ~70% of cells in the culture are cortical radial precursors (RPs) as indicated by the definitive cortical marker Emx1, and over 8 days in culture, these RPs differentiate to pyramidal glutamatergic neurons of the cortex mimicking in vivo development. Astrocyte differentiation can be observed later as the culture progresses, which again mimics the typical timed genesis of cells in the cortex. The stem cell properties and cell fate of these RPs can be manipulated with growth factors in culture as they are in vivo. In particular, FGF2 promotes proliferation and survival, while ciliary neurotrophic factor (CNTF) induces precocious astrocyte formation. Thus, our ES-derived cortical RP cultures can serve as an alternate and complementary in vitro model to examine neural precursor biology during early development.

Murine neural precursor cells have been a well studied model for neural cell fate determination and stem cell function both in vivo and in primary culture. However, factors such as cell number, the presence of multiple cell populations and of niche intrinsic factors made it difficult to dissect the mechanisms regulating cortical development. To overcome this issue, we have developed a culture system where mouse embryonic stem cells (ES) are differentiated to cortical radial precursors through retinoic acid treatment of embryoid bodies. One day after plating in neural differentiation conditions, ~70% of cells in the culture are cortical radial precursors (RPs) as indicated by the definitive cortical marker Emx1, and over 8 days in culture, these RPs differentiate to pyramidal glutamatergic neurons of the cortex mimicking in vivo development. Astrocyte differentiation can be observed later as the culture progresses, which again mimics the typical timed genesis of cells in the cortex. The stem cell properties and cell fate of these RPs can be manipulated with growth factors in culture as they are in vivo. In particular, FGF2 promotes proliferation and survival, while ciliary neurotrophic factor (CNTF) induces precocious astrocyte formation. Thus, our ES-derived cortical RP cultures can serve as an alternate and complementary in vitro model to examine neural precursor biology during early development.

名古屋大学博士学位論文 学位の種類:博士（医療技術学） (課程) 学位授与年月日　平成22年3月25日

Embryonic stem cells (ESCs) hold great promise for treatment of degenerative disorders such as Parkinson's and Alzheimer's disease, diabetes, and cardiovascular disease. The ability of ESCs to differentiate to all somatic cell types suggests that they may serve as a robust cell source for production of differentiated cells for regenerative medicine and other cell-based therapeutics. In order for ESCs to be used effectively in clinical settings, efficient and reproducible differentiation to targeted cell types must be demonstrated. The overall objective of this project was to engineer microenvironmental control over differentiating ESCs through the formation of embryoid bodies (EBs) uniform in size and shape, and through the incorporation of morphogen-containing polymer microspheres within the interior of EBs. The central hypothesis was that morphogen delivery through incorporated polymer microspheres within a uniform population of EBs will induce controlled and uniform differentiation of ESCs. Rotary suspension culture was developed in order to efficiently produce uniform EBs in high yield. Compared to static suspension culture, rotary suspension significantly improved the production of differentiating cells and EBs over the course of 7 days, while simultaneously improving the homogeneity of EB size and shape compared to both hanging drop and static EBs. The diffusive transport properties of EBs formed via rotary suspension were investigated using a fluorescent, cell permeable dye to model the movement of small morphogenic molecules within EBs. Confocal microscopy, cryosections and EB dissociation all demonstrated that the dye was not able to fully penetrate EB, and that the larger EBs at later time points (7 days) retarded dye movement to a greater extent than earlier EBs (days 2 and 4). Polymer microspheres capable of encapsulating morphogenic factors were incorporated into EBs in order to overcome the diffusional limitations of traditional soluble delivery. The size of microspheres, microsphere coating, microsphere to cell ratio, and rotary mixing speed were all observed to influence incorporation within EBs. The use of microsphere-mediated delivery within EBs to direct cell differentiation was examined. Microsphere-mediated delivery of retinoic acid (RA) induced formation of uniquely cystic spheroids with a visceral endoderm layer enveloping a pseudo-stratified columnar epithelium, and with spatial localization of transcriptional profiles similar to the early primitive streak stage of mouse development. Continued differentiation of RA MS EBs in defined media conditions was assessed. Gene expression demonstrated that regular serum enhanced endoderm induction, serum-free media supported ectoderm differentiation, while mesoderm was most prominent in untreated EBs in full serum. In summary, this work has realized a unique approach for stem cell differentiation through modification of the internal microenvironment of ESC spheroids. This novel inside-out method toward engineering EBs demonstrated that the mode of morphogen delivery significantly affected the course of differentiation. These studies provide the basis for ongoing work, which will utilize the choice of microsphere material, coating, and morphogen in order to uniquely study mechanisms of ESC differentiation and achieve unparalleled engineering of the EB microenvironment.

Thesis (Ph. D.)--University of Oklahoma. / Bibliography: leaves 150-172.