There is a growing need for bone cells to be used in the generation and repair of bone tissues that have become destroyed as a result of disease (e.g. osteoporosis), injuries and genetic deficiencies. Human umbilical cord blood has been a source of cells that is being investigated for its ability to generate cells for repair of various tissues including bone, which is the focus of this investigation. Similarly cord blood can also be used as a source of progenitors for osteoclasts that might be used for management of other diseases of bone characterized by defects in these cells (e.g. osteopetrosis). Moreover, there are few reliable in vitro models for human osteoclasts and so the ability to develop a cell model for human osteoclasts will also permit more meaningful studies on regulation of these cells than possible in the past. One of the setbacks suffered by researchers in designing therapeutic uses for stem cells is that in laboratory animals tumours can arise from these transplanted and pleuripotential cells. This suggests that stem cells might also have a dysregulated apoptosis pathway, which is critically important to understand before the safe use of stem cells can be assured. In order to increase our understanding of apoptosis in bone cells in particular, studies were done to try to understand the role of harakiri, a pro-apoptotic gene, in the development of osteoblasts and osteoclasts. Therefore, this series of study had two main foci; the development of human bone cells, particularly osteoclasts and osteoblasts, from human umbilical cord blood, and to understand further the mechanisms regulating apoptosis in bone cells. This study showed that osteoblasts could not be derived easily from human cord blood cells while it was possible to generate fully functional osteoclasts, which demonstrated unique properties. Studies done on harakiri deficient mice showed that absence of this gene caused an increase in osteoblast formation and a decrease in osteoclast formation. These findings can be exploited when considering the development of pharmacological agents that might be used in the future to modulate osteoclast cell development, function, and apoptosis.

There is a growing need for bone cells to be used in the generation and repair of bone tissues that have become destroyed as a result of disease (e.g. osteoporosis), injuries and genetic deficiencies. Human umbilical cord blood has been a source of cells that is being investigated for its ability to generate cells for repair of various tissues including bone, which is the focus of this investigation. Similarly cord blood can also be used as a source of progenitors for osteoclasts that might be used for management of other diseases of bone characterized by defects in these cells (e.g. osteopetrosis). Moreover, there are few reliable in vitro models for human osteoclasts and so the ability to develop a cell model for human osteoclasts will also permit more meaningful studies on regulation of these cells than possible in the past. One of the setbacks suffered by researchers in designing therapeutic uses for stem cells is that in laboratory animals tumours can arise from these transplanted and pleuripotential cells. This suggests that stem cells might also have a dysregulated apoptosis pathway, which is critically important to understand before the safe use of stem cells can be assured. In order to increase our understanding of apoptosis in bone cells in particular, studies were done to try to understand the role of harakiri, a pro-apoptotic gene, in the development of osteoblasts and osteoclasts. Therefore, this series of study had two main foci; the development of human bone cells, particularly osteoclasts and osteoblasts, from human umbilical cord blood, and to understand further the mechanisms regulating apoptosis in bone cells. This study showed that osteoblasts could not be derived easily from human cord blood cells while it was possible to generate fully functional osteoclasts, which demonstrated unique properties. Studies done on harakiri deficient mice showed that absence of this gene caused an increase in osteoblast formation and a decrease in osteoclast formation. These findings can be exploited when considering the development of pharmacological agents that might be used in the future to modulate osteoclast cell development, function, and apoptosis.

Bone tissue is continuously remodeled by bone-forming osteoblasts and bone-resorbing osteoclasts, in processes tightly regulated by hormones, cytokines and growth factors. CD47, a ubiquitously expressed protein, and one of its ligands, signal-regulatory protein alpha (SIRPα), are two cell-surface proteins belonging to the immunoglobulin (Ig)-superfamily. The interaction between CD47 and SIRPα is important for, amongst other processes, the fusion of macrophages into giant cells, which are closely related to osteoclasts. The aim of the present study was to gain knowledge about the role of CD47–SIRPα interaction and resultant downstream signaling pathways in bone cell differentiation, formation and function. The addition of antibodies against CD47 or SIRPα inhibited the formation of multinucleated osteoclasts from bone marrow monocytes (BMMs) in culture. Moreover, a significant decrease in the number of osteoclasts was detected in CD47-/- BMM cultures compared to CD47+/+ cultures. In line with these in vitro results, we found fewer osteoclasts in vivo in the trabecular bone of CD47-/- mice, as compared to CD47+/+ bone. Interestingly, an extended analysis of the trabecular bone of CD47-/- mice revealed that the bone volume, mineralizing surface, mineral apposition rate, bone formation rate and osteoblast number were also significantly reduced compared with CD47+/+ mice, indicating the importance of CD47 in osteoblast differentiation. In vitro studies of bone marrow stromal (BMS) cells from CD47-/- mice or SIRPα-mutant mice (mice lacking the signaling domain of SIRPa) showed a blunted expression of osteoblast-associated genes. Moreover, these altered genotypes were associated with reduced activity of the bone mineralization-associated enzyme alkaline phosphatase as well as a reduced ability to form mineral. To reveal the molecular mechanisms by which CD47 activation of SIRPα is important for BMS cell differentiation, we studied signaling downstream of SIRPα in the absence of CD47. In BMS cells lacking CD47, a considerable reduction in the levels of tyrosine phosphorylated SIRPα was detected, and the subsequent recruitment of the Src-homology-2 (SH2) domain-containing protein tyrosine phosphatase (SHP-2)–phosphoinositide 3-kinase (PI3K)–Akt2 signaling module was nearly abolished. In conclusion, the interaction between CD47 and SIRPα results in the activation of the SHP-2–PI3K–Akt2 pathway, which is necessary for normal osteoblast differentiation. In CD47-/- mice and SIRPα-mutant mice, this interaction is perturbed, which prevents normal osteoblast differentiation and subsequent mineral formation. In addition, the altered BMS cell phenotype results in an impaired ability to stimulate osteoclast differentiation.

Inflammatory bone diseases like rheumatoid arthritis and periodontal disease lead to increased bone loss in the inflamed areas. The multinucleated bone resorbing cells, the osteoclasts, present in these diseases are larger than normal, and these larger osteoclasts (10+ nuclei) resorb more bone and more often than smaller osteoclasts (2-5 nuclei). Thus, the focus of this thesis was to determine if there are differences in mechanisms of osteoclast activation between large and small osteoclasts. Experiments using authentic rabbit osteoclasts and RAW 264.7-derived osteoclasts revealed differences in the expression of a number of activating factors; with large osteoclasts expressing higher levels of activating receptors (RANK, IL-1RI, TNFR1 and integrins αv and β3), as well as enzymes involved in cellular resorption, while small osteoclasts expressed higher levels of an alleged fusion receptor and the inhibitory receptor, IL-1RII. Further studies revealed that large osteoclasts more readily responded to stimulation by IL-1 compared to small osteoclasts and at lower concentrations suggesting this is a result of their higher expression of activating receptors. Differences in responses to the IL-1 isoforms, IL-1α and IL-1β, were also seen in large osteoclasts: IL-1α generated more large osteoclasts over the course of differentiation, while IL-1β induced changes in cell morphology and in the induction of integrin β3 phosphorylation. These observations suggested that differences in osteoclast responses are induced by IL-1α and IL-1β and it led to the hypothesis that there are differences in signaling between large and small osteoclasts. To elucidate differences in signaling mechanisms a signaling pathway microarray was used which revealed higher expression of Vegfa in large compared to small osteoclasts. Osteoclast differentiation with RANKL increased Vegfa gene expression in a time-dependent manner and VEGF-A secretion was elevated in populations enriched for large osteoclasts. Furthermore, mechanistic studies with inhibitors of transcription factors involved in differentiation revealed that RANKL-mediated Vegfa expression in large osteoclasts was regulated by the NF-κB pathway via induction of Hif1α. These results support the hypothesis that signaling differences exist between large and small osteoclasts and implicates VEGF-A in osteoclast hyperactivity in inflammatory conditions.

Osteoblasts and osteoclasts are crucial to maintaining bone homeostasis. These specialized cells rely on various environmental signals and cross talk from one another in order to model, remodel and repair bone. Exogenous chemicals such as the therapeutic drug rosiglitazone, a peroxisome proliferator activated receptor gamma (PPARγ) agonist, can interfere with bone-forming and bone-resorbing pathways, causing osteoporosis and increasing the risk of bone fracture. Evidence is emerging that environmental toxicants induce similar toxic endpoints in bone, both through PPARγ-dependent and PPARγ-independent mechanisms. To date, these toxicants have only ever been considered in isolation or in limited co-exposure studies. This comprehensive review relating these toxicants and their effect on bone health will help guide future studies and illuminate gaps in our knowledge. Five toxicant classes (organotins—mainly tributyltin; TBT, heavy metals—lead, cadmium, and arsenic, dioxin-like chemicals, phthalates, and perfluoroalkyl compounds; PFAs) were examined, with emphasis on molecular targets, osteoclast- and osteoblast-specific effects, animal models and epidemiological data. It was concluded that organotins (TBT) act via PPARγ and RXR agonism, phthalates act via PPARγ agonism, heavy metals act at least through ERK-mediated pathways, and dioxin-like chemicals act through aryl hydrocarbon receptor interaction. The molecular targets of PFAs remain unknown. Additional targets are still being investigated. These findings emphasize the importance of co-exposures, as these toxicants act through diverse molecular mechanisms that may share toxic endpoints, making co-exposure consequences particularly severe. While the evidence available strongly suggests that lead, cadmium, and dioxin-like chemicals are negative modulators of human bone health, evidence supporting this conclusion for organotins, phthalates, arsenic, and PFAs is somewhat lacking. There are still significant gaps in our understanding that must be filled to gain a holistic understanding of these threats to human bone health.

Abstract The role of vacuolar H+-ATPase (V-ATPase) in exocytic and endocytic membrane transport processes was studied by using its specific inhibitor, bafilomycin A1 (Baf A1), as a tool. On the exocytic pathway, both brefeldin A- and nocodazole-induced retrograde transport of Golgi proteins to the endoplasmic reticulum (ER) were inhibited by Baf A1. Furthermore, p58/ERGIC-53, which normally cycles between the ER, the intermediate compartment (IC), and cis-Golgi, was arrested in pre-Golgi tubules and vacuoles, and the number of p58-positive 80-nm Golgi (COPI) vesicles was reduced, suggesting that the drug inhibits the vesicle-mediated retrieval of the protein from post-ER compartments. The small GTPase rab1p was efficiently recruited to the tubules, accumulating in the presence of Baf A1. In contrast, these tubules showed no enrichment of anterogradely transported proteins, indicating that they participate in retrograde transport. Interestingly, acidic lumenal pH could only be detected in the more central pre-Golgi elements. The forward (anterograde) transport of newly synthesized Semliki Forest virus (SFV) and vesicular stomatitis virus (VSV) glycoproteins from the ER to the cis-Golgi was largely unaffected by Baf A1. However, maturation processes occurring in the trans-Golgi were inhibited, and the amounts of viral glycoproteins appearing at the cell surface were reduced. Newly synthesized VSV glycoprotein accumulated into rab1p-positive Golgi membranes in the presence of Baf A1, indicating that the transport from cis-Golgi was affected. Furthermore, O-glycosylation of the expressed CD8 chimeras and lectin cytochemistry experiments indicate that Baf A1 affects the transport from cis-Golgi. Instead, Baf A1 did not affect the transport of viral glycoproteins from the trans-Golgi network to the cell surface. We propose, that anterograde intra-Golgi traffic may be affected indirectly by Baf A1, as it inhibits retrograde vesicle-mediated transport and thus cisternal maturation. Baf A1 inhibited the entry of SFV into BHK-21 cells. Thus, V-ATPase was responsible for the acidification of the endosomes needed for virus entry. In cells infected with VSV and subsequently treated with Baf A1, virus particles were found to be accumulated in tubular membrane structures, which also contained endocytosed BSA-gold. Neither VSV nor BSA-gold particles were detected in lysosomal glycoprotein (lgp) 120-positive lysosomes, however. Thus, secreted and further endocytosed virus particles accumulate into tubulated endocytic organelles, apparently early endosomes, in Baf A1-treated cells. We conclude that the transport from endosomes to lysosomes is inhibited by Baf A1. The bulk of rab7 GTPase, which participates in vesicle fusion to late endosomes, was localized to the ruffled border (RB) membrane of bone-resorbing osteoclast. This indicates that the membrane has some characteristics of late endosomal membranes and that endocytic membrane transport is oriented towards the RB. Consistently, both endocytosed lumenal horseradish peroxidase and receptor-bound transferrin were delivered to the RB. The delivery of membrane-associated transferrin to the RB further indicates that the RB has some endosomal characteristics and suggests that the endocytic pathway contributes to the maintenance of functional RB. The endocytic pathway could act in balancing the membrane traffic associated with transcytosis from the RB to the basal plasma membrane. Endocytic processes in osteoclasts appeared to be very sensitive to Baf A1. Thus, blocking of the endocytic membrane traffic towards the RB could explain the inactivation of cells by low concentrations of the drug.

Inflammatory arthritis (IA) is a debilitating disease that is characterized by joint destruction. This destruction is caused by osteoclasts (OCLs) degrading bone within the synovium, however the exact cellular origins of these synovial OCLs is not well understood. We hypothesize that the synovial OCLs seen in IA are independently derived from two contributing cell lineages: the canonical source of OCL which are Hematopoietic Stem Cell (HSC) derived monocytes and the newly described Erythro-Myeloid Progenitor (EMP). To explore the contribution of these two lineages to synovial OCL, we used Cx3cr1CreERT2;Rosa26LSL-tdTomato mice to label EMP-derived cells, and Flt3Cre;Rosa26LSL-YFP to label HSC-derived cells. Using immunofluorescent histology, we found that synovial OCLs formed under arthritic conditions derive from both HSC and EMP-progenitors, suggesting the possibility that regulatory mechanisms unique to each developmental lineage promote OCL differentiation in arthritic joints. In support of these observations, in IA we detected two populations of mononuclear cells, as possible osteoclast precursors, that express both the OCL marker TRAP and the monocytic/macrophage marker CD68. These mononuclear TRAP+ CD68+ cell populations are found within the inflamed synovium and in bone periosteal surfaces of arthritic joints, and are mostly EMP derived, however, the HSC lineage significantly contributes to osteoclast formation, as suggested by our lineage-tracing strategy. We are currently investigating the dynamics of these cell populations during the early, peak, and resolution stages in an acute murine model of IA. To better understand regulatory differences between OCL derived from each of these lineages, we plan to isolate EMP- and HSC-derived OCLs directly from joints to study their precise phenotype, cytokine differentiation requirements, resorptive capacity and transcriptional activity, by flow cytometry, in vitro cell cultures and single cell RNA sequencing, respectively.

Indiana University-Purdue University Indianapolis (IUPUI) / Many factors are present in the development of skeletal tissue. Some factors lead to an increase in bone mass while some lead to a decrease. One factor that is known to have an influence on skeletal development is Signal Transducer and Activator of Transcription 3 (Stat3). This knowledge arose because of a mutation in the Stat3 gene in humans causing a disease called Hyper-IgE Syndrome. This mutation leads to a variety of issues, including decreased bone mass. Because of this, our lab has sought to study Stat3 in its relation to bone. Many studies have already been conducted that discern how Stat3 influences skeletal biology by observing its role in osteoclasts, osteoblasts, and other bone cells. Its role is still unclear, and many studies have provided seemingly contradictory results in how it works on bone tissue. Our lab set up several different studies in order to further elucidate what role Stat3 plays in skeletal development by looking at its effects on osteoblasts and osteoclasts, the bone-forming and bone-destroying cells of the body, respectively. We conditionally knocked out Stat3 in the osteoblasts of mice and compared several different bone parameters to their wild type counterparts at 8 weeks of age. Differences were noted in bone phenotype, including decreased femur length, weight, bone mineral density, and bone mineral content in the cKO compared to their WT counterparts. While no significant difference in trabecular integrity was noted, several differences were observed in cortical bone. These differences indicate that Stat3 has a positive role in osteoblast differentiation, leading to an overall positive effect on bone mass. To observe the role of Stat3 in osteoclasts, in vitro experiments were set up in which pre-osteoclast RAW 264.7 cells were manipulated with Stat3 siRNA or a Stat3 overexpression construct and RANKL to induce differentiation. Using qPCR and western blot assays, it was determined that when Stat3 is knocked down, several important genes in osteoclastogenesis and osteoclast function are more highly expressed than in the control groups. When Stat3 is overexpressed, a similar pattern is observed where these same genes are downregulated in the presence of higher Stat3 levels. These results indicate that Stat3 has an overall inhibitory effect on osteoclastogenesis and osteoclast function, indicating it has a positive effect on bone mass. Future studies could be performed to further elucidate the effects of Stat3 on skeletal development. Isolating the osteoblasts from cKO and WT mice and performing qPCR and western blot assays could be useful in finding out how Stat3 is influencing these cells. Further studies could also be done on the RAW 264.7 cells to find where Stat3 is interacting with the RANKL pathway. A resorption assay could be done with these cells to better understand how function might be influenced by Stat3.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第24190号 / 医博第4884号 / 新制||医||1060(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 伊藤 能永, 教授 安達 泰治, 教授 椛島 健治 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.