Human Bone Marrow Microenvironment Research Articles

Generation of complex bone marrow organoids from human induced pluripotent stem cells

The human bone marrow (BM) niche sustains hematopoiesis throughout life. We present a method for generating complex BM-like organoids (BMOs) from human induced pluripotent stem cells (iPSCs). BMOs consist of key cell types that self-organize into spatially defined three-dimensional structures mimicking cellular, structural and molecular characteristics of the hematopoietic microenvironment. Functional properties of BMOs include the presence of an in vivo-like vascular network, the presence of multipotent mesenchymal stem/progenitor cells, the support of neutrophil differentiation and responsiveness to inflammatory stimuli. Single-cell RNA sequencing revealed a heterocellular composition including the presence of a hematopoietic stem/progenitor (HSPC) cluster expressing genes of fetal HSCs. BMO-derived HSPCs also exhibited lymphoid potential and a subset demonstrated transient engraftment potential upon xenotransplantation in mice. We show that the BMOs could enable the modeling of hematopoietic developmental aspects and inborn errors of hematopoiesis, as shown for human VPS45 deficiency. Thus, iPSC-derived BMOs serve as a physiologically relevant in vitro model of the human BM microenvironment to study hematopoietic development and BM diseases.

Establishment of Human Bone Marrow-on-Chip As a Preclinical Model to Evaluate Drug-Induced Toxicities and Myelofibrosis Patient-Specific Pathophysiology

Hematopoiesis takes place in the bone marrow (BM) and is supported by a complex cellular and molecular network in the BM microenvironment. The inaccessibility of living human BM hampers the study of its pathophysiology under stress induced by drugs, radiation, or genetic mutations. Commonly used models of the human BM microenvironment rely on mouse models and 2D or 3D static tissue cultures. Although existing in vitro hematopoiesis models have led to advances in the field, they fail to recapitulate the biology of cellular interactions, cell extrinsic factors, and complex functions of living human BM hematopoiesis. This has also limited our understanding of human BM biology and the influence of BM niche cells in BM defects in diseases such as myeloproliferative neoplasms (MPNs) and leukemias. Consequently, there is an unmet need to develop a preclinical human BM hematopoiesis model that supports growth and differentiation of multilineage hematopoietic stem cells to enable evaluation of investigational compounds for myeloerythroid toxicity and efficacy. Microphysiologic systems (MPS), such as organ-on-chips, emulate the complex interplay between intrinsic and extrinsic cell factors in the BM microenvironment, the 3D tissue architecture, and are more clinically relevant compared with static co-culture approaches (PMID: 31988457; PMID: 37359774). Here we present the development of a human BM-on-chip MPS model to recapitulate both healthy and MPN disease and evaluate the alterations in the BM niche environment. We utilized a commercially available organ-on-chip microfluidic platform (Emulate, Boston, MA) to develop a living human BM with a functional hematopoietic niche in vitro by perfusion with culture medium. This fluidic-controlled modular chip comprises an osteogenic apical channel separated from a vascular basal channel by a semiporous membrane, which allows for media perfusion, cellular crosstalk, and intercellular signaling. The osteogenic apical channel is filled with a fibrin gel that allows the co-culture of all critical cells of the BM niche, including osteoblasts, endothelial cells, mesenchymal stem cells, hematopoietic stem and progenitor cells. The parallel basal channel is lined by human vascular endothelium and is perfused with culture medium optimized to allow survival and development of functional hematopoietic lineage niche cell populations. This unidirectional fluidic MPS model allows for collection of cell effluents for evaluation of cell-extrinsic factors such as cytokines, extracellular vesicles, and metabolites. The modular and accessible nature of this system provides an advantage to investigate the effects of the microenvironment and drug candidates individually on the osteogenic and vascular compartments. This human BM-on-chip has been tested with cells from several healthy and myelofibrosis donors. Utilizing imaging, flow cytometry, cytokine profiling, and transcriptomics, we demonstrate that the vascularized human BM-on-chip recapitulates the living human hematopoietic niche, with BM relevant cytokine levels, retains hematopoietic stem and progenitor cells in in vivo-like proportions for over 3 weeks in culture, allows for the differentiation of BM niche lineage cell populations similar to human BM aspirate, and captures human donor variabilities. We also show that BM-on-chips derived from MPN patients recapitulate the hematopoietic defects to understand the biology and disease response in myelofibrosis. Our robust experimental workflow and bioinformatics platform will be utilized to evaluate candidate compounds for BM myelotoxicities, drug mechanisms of action, and biomarkers in a multifactorial manner to support preclinical programs. This biomimetic platform offers a new approach for analysis of drug responses and toxicities in BM as well as for study of hematopoiesis and hematologic diseases such as myelofibrosis.

Combined Single-Cell and Spatial Transcriptomics Unveil the Complex Organization of the Non-Immune Human Bone Marrow Microenvironment during Aging

Hematopoietic stem cells (HSCs), reside in a specialized bone marrow (BM) microenvironment known as the hematopoietic stem niche. Despite the critical role of the niche in tightly controlling the processes of normal and malignant hematopoiesis, its cellular composition remains only partially understood. Recent studies, including our own, using single-cell RNA sequencing (scRNA-seq) approaches have provided valuable insights into the profiling of the human BM niche. However, a comprehensive effort to describe the cellular heterogeneity and regulatory circuitry of the aged BM microenvironment in elderly humans is still lacking. As individual age, multiple systems and organs experience a progressive loss of anatomical and physiological integrity. Aging of the HSC niche is accompanied by a reduction in the numbers and function of its constituents and a decrease in the levels of HSC-supporting factors. Whether niche aging can contribute to the defects observed in aged hematopoiesis, such as the clonal shift towards myelopoiesis, the decrease in immune surveillance, or age-associated metabolic diseases, remains unresolved. To address this, we utilized scRNA-seq and spatial transcriptomics to provide a detailed characterization of the molecular landscape and stromal interactions in the aged non-immune BM microenvironment. First, we performed scRNA-seq profiling on fluorescence-activated cell sorting-purified endothelial cells (ECs, TO-PRO-3 -/CD45 -/CD235 -/Lin -/CD31 +/CD9 +) and mesenchymal stromal cells (MSCs, TO-PRO-3 -/CD45 -/CD235 -/Lin -/CD271 +) from human BM samples of young (n=4) and elderly healthy donors (n=5). We analyzed a total of 1514 ECs and 3848 MSCs from older adults, grouped into 7 and 10 subclusters, respectively, defining distinct functional cell states. ECs and MSCs cells from young individuals were annotated using SingleR, utilizing the gene signature of each functional state in the elderly as a reference. Our results revealed significant shifts in the distribution of the functional states of BM niche cells during aging. Regarding ECs, we noticed reduced pathways associated with the cell cycle and RNA transcription pathways, indicating impaired cell cycle activity, coupled with decreased antioxidant defense (Figure 1A). In contrast, there was an increase in subgroups related to the response to foreign molecules, immune system activation, and vascular remodeling, suggesting an inflammatory response and increased vascular remodeling processes associated with aging. For MSCs, we observed a decline in osteogenesis and a reduction in the proportion of the early mesenchymal group related to immunity and the extracellular matrix (Figure 1B). These changes could contribute to the decreased ability of MSCs in elderly individuals to maintain HSCs function. On the contrary, we noticed increased adipogenic differentiation to the detriment of osteogenesis, which could be responsible for osteoporosis problems. Next, and to further explore the location and the direct interactions between niche cells, we are currently integrating spatially resolved transcriptomics with our scRNA-seq data to accurately comprehend the spatial distribution and interactions among the BM niche cells, using for the first time 10x Genomics Visium Spatial Gene Expression on bone BM tissue in humans. As a final validation step, we are using a mouse model to validate further the mechanisms of age-associated changes in the BM microenvironment. In summary, our results provide valuable insights into age-related transcriptional alterations in human BM ECs and MSCs, suggesting altered behavior of these niche cells during the aging of the hematopoietic system. This deeper understanding of the architecture of the aged hematopoietic system and its microenvironment offers the potential for developing novel therapeutic strategies preventing the detrimental effects of aging.

Human Induced Pluripotent Stem Cell-Derived Bone Marrow Organoids to Model Hematopoietic Development

The human bone marrow niche plays an essential role in supporting and regulating hematopoiesis throughout life and inherited bone marrow failure syndromes are often linked to dysfunction of the niche. Recently, methods for generating induced pluripotent stem cell (iPSC)-derived hematopoietic organoids (Motazedian et al., 2020) and bone marrow organoids (Khan et al., 2022) have been developed. We recently presented our approach to generate a model of a complex human bone marrow microenvironment in vitro (ASH annual meeting 2022 ). Here, we characterize our model in greater detail and analyse functionality of bone marrow organoid (BMO)-derived HSPCs. We devised a feeder- and serum-free protocol to generate a complex human iPSC-derived BMO within three weeks. In comparison to the protocol by Khan et al., we incorporated a sequential combination of Wnt activation and Activin/Nodal inhibition to induce mesoderm formation and patterning. We replicated the method with five different iPSC lines, including fibroblast- and PBMC-derived iPSCs, as well as iPSCs generated from renal epithelial cells isolated from urine samples. Using confocal and two-photon microscopy we visualized the spatial architecture of differentiated BMOs and detected CD45 + cells embedded into a network of CD31 + vascular cells and CD271 + stromal cells. PDGFRβ expressing pericytes were enwrapping the vascular cells. Moreover, stromal cells expressed CXCL12, Leptin receptor and Nestin. Electron microscopy, HE staining of organoid sections and 3D surface rendering showed lumen-forming vascular structures containing hematopoietic cells. To further elucidate the cell type composition, we performed single-cell RNA sequencing (scRNA-seq) of differentiated BMOs. Clustering of the scRNA-seq data by marker gene expression yielded three main populations: Cells in the endothelial cluster expressed canonical genes of arterial ECs. In the mesenchymal cluster we could distinguish pericytes, vascular smooth muscle cells and osteochondrogenic precursors. In the hematopoietic compartment, we identified cells expressing molecular signatures of granulocytes, monocytes and their progenitors, but also megakaryocytic-like cells and lymphoid progenitors. Remarkably, a cluster of cells expressed genes of fetal HSCs. This suggests that BMOs may recapitulate definitive hematopoiesis. To test T-cell differentiation potential in vitro, we sorted CD34 + HSPCs from dissociated BMOs and differentiated them within an artificial thymic organoid system (ATO). After three weeks we detected CD45 +CD4 +CD8 + cells, some of them expressing CD3 +TCRαβ +. Thus, we postulate that T-cells phenotypically resembling conventional T-cells emerge from BMO-derived HSPCs. To assess functional properties of BMO-derived HSPCs in vivo, we transplanted sorted CD34 + HSPCs into immunodeficient NSG mice and are currently analysing their engraftment potential. In summary, we provide a novel bone marrow niche model to study hematopoietic development in a complex three-dimensional organoid derived from human iPSCs.

Three-Dimensional Human Bone Marrow Organoids for the Study and Application of Normal and Abnormal Hematoimmunopoiesis.

Hematoimmunopoiesis takes place in the adult human bone marrow (BM), which is composed of heterogeneous niches with complex architecture that enables tight regulation of homeostatic and stress responses. There is a paucity of representative culture systems that recapitulate the heterogeneous three-dimensional (3D) human BM microenvironment and that can endogenously produce soluble factors and extracellular matrix that deliver culture fidelity for the study of both normal and abnormal hematopoiesis. Native BM lymphoid populations are also poorly represented in current invitro and invivo models, creating challenges for the study and treatment of BM immunopathology. BM organoid models leverage normal 3D organ structure to recreate functional niche microenvironments. Our focus herein is to review the current state of the art in the use of 3D BM organoids, focusing on their capacities to recreate critical quality attributes of the invivo BM microenvironment for the study of human normal and abnormal hematopoiesis.

Identification of phenotypically, functionally, and anatomically distinct stromal niche populations in human bone marrow based on single-cell RNA sequencing.

Hematopoiesis is regulated by the bone marrow (BM) stroma. However, cellular identities and functions of the different BM stromal elements in humans remain poorly defined. Based on single-cell RNA sequencing (scRNAseq), we systematically characterized the human non-hematopoietic BM stromal compartment and we investigated stromal cell regulation principles based on the RNA velocity analysis using scVelo and studied the interactions between the human BM stromal cells and hematopoietic cells based on ligand-receptor (LR) expression using CellPhoneDB. scRNAseq led to the identification of six transcriptionally and functionally distinct stromal cell populations. Stromal cell differentiation hierarchy was recapitulated based on RNA velocity analysis and in vitro proliferation capacities and differentiation potentials. Potential key factors that might govern the transition from stem and progenitor cells to fate-committed cells were identified. In situ localization analysis demonstrated that different stromal cells were localized in different niches in the bone marrow. In silico cell-cell communication analysis further predicted that different stromal cell types might regulate hematopoiesis through distinct mechanisms. These findings provide the basis for a comprehensive understanding of the cellular complexity of the human BM microenvironment and the intricate stroma-hematopoiesis crosstalk mechanisms, thus refining our current view on human hematopoietic niche organization.

Bioengineering the human bone marrow microenvironment in liquefied compartments: A promising approach for the recapitulation of osteovascular niches

Recreating the biological complexity of living bone marrow (BM) in a single in vitro strategy has faced many challenges. Most bioengineered strategies propose the co-culture of BM cellular components entrapped in different matrices limiting their migration and self-organization capacity or in open scaffolds enabling their escaping. We propose a methodology for fabricating a “human bone marrow–in–a–liquefied-capsule” to overcome these challenges, embracing the most important BM components in a single platform. Since free dispersion of the cells within the BM is an essential feature to maintain their in vivo properties, this platform provides a liquefied environment for the encapsulated cells to move freely and self-organize. Inside liquefied capsules, an engineered endosteal niche (eEN) is co-cultured with human umbilical cord cells, including endothelial cells and hematopoietic stem and progenitor cells (HSPCs). Two different human-like BM niches were recreated under static and dynamic systems. Although the culture of the engineered BM capsules (eBMC) in these different environments did not change the structural and compositional features of the BM niches, the biophysical stimulation potentiated the cellular intercommunication and the biomolecules secretion, demonstrating an enhanced in vitro bio performance. Moreover, while the eBMC without HSPCs provided the secretion of hematopoietic supportive factors, the presence of these cells recapitulated more closely the biological complexity of the native BM niches. This functional eBMC approach is an innovative platform capable of investigating several components and interactions of BM niches and how they regulate BM homeostasis and hematopoiesis. Statement of significanceThe recapitulation of the multifaceted bone marrow (BM) microenvironment under in vitro conditions has gained intensive recognition to understand the intrinsic complexity of the native BM. While conventional strategies do not recapitulate the BM osteovascular niches nor give the cellular components a free movement, we report for the first time the development of human bone marrow–in–a–liquefied-capsule to overcome such limitations. Our engineered BM capsules (eBMC) partially mimic the complex structure, composition, and spatial organization of the native osteovascular niches present in the BM. This strategy offers a platform with physiological relevance to exploit the niches’ components/networks and how they regulate the hematopoiesis and the initiation/progression of various BM-related pathologies.

Bone Marrow Microenvironment-On-Chip for Culture of Functional Hematopoietic Stem Cells.

Hematopoiesis takes place in the bone marrow and is supported by a complex cellular and molecular network in the bone marrow microenvironment. Commonly used models of the human bone marrow microenvironment include murine models and two-dimensional and three-dimensional tissue cultures. While these model systems have led to critical advances in the field, they fail to recapitulate many aspects of the human bone marrow. This has limited our understanding of human bone marrow pathophysiology and has led to deficiencies in therapy for many bone marrow pathologies such as bone marrow failure syndromes and leukemias. Therefore, we have developed a modular murine bone marrow microenvironment-on-chip using a commercially available microfluidic platform. This model includes a vascular channel separated from the bone marrow channel by a semi-porous membrane and incorporates critical components of the bone marrow microenvironment, including osteoblasts, endothelial cells, mesenchymal stem cells, and hematopoietic stem and progenitor cells. This system is capable of maintaining functional hematopoietic stem cells in vitro for at least 14 days at frequencies similar to what is found in the primary bone marrow. The modular nature of this system and its accessibility will allow for acceleration of our understanding of the bone marrow.

3D Multicellular Spheroid for the Study of Human Hematopoietic Stem Cells: Synergistic Effect Between Oxygen Levels, Mesenchymal Stromal Cells and Endothelial Cells.

IntroductionThe human bone marrow microenvironment is composed of biological, chemical and physical factors that act in a synergistic way to modulate hematopoietic stem cell biology, such as mesenchymal stromal cells (MSCs), endothelial cells (ECs) and low oxygen levels; however, it is difficult to mimic this human microenvironment in vitro.MethodsIn this work, we developed 3D multicellular spheroid (3D-MS) for the study of human hematopoietic stem cells (HSCs) with some components of perivascular niche. HSCs were isolated from umbilical cord blood, MSCs were isolated from human bone marrow and a microvasculature EC line (CC-2811, Lonza®) was used. For the formation of a 3D structure, a magnetic levitation culture system was used. Cultures were maintained in 21%, 3% and 1% O2 for 15 days. Culture volume, sphericity index and cell viability were determined. Also, human HSC proliferation, phenotype and production of reactive oxygen species were evaluated.ResultsAfter 15 days, 3D-MS exhibited viability greater than 80%. Histology results showed structures without necrotic centers, and higher cellular proliferation with 3% O2. An increase in the expression of the CD34 antigen and other hematopoietic antigens were observed to 1% O2 with MSCs plus ECs and low ROS levels.ConclusionThese findings suggest that 3D-MS formed by MSCs, ECs and HSCs exposed to low concentrations of oxygen (1–3% O2) modulate human HSC behavior and mimics some features of the perivascular niche, which could reduce the use of animal models and deepen the relationship between the microenvironment of HSC and human hematological diseases development.

Acute myeloid leukemia–induced remodeling of the human bone marrow niche predicts clinical outcome

AbstractMurine models of myeloid neoplasia show how leukemia infiltration alters the hematopoietic stem cell (HSC) niche to reinforce malignancy at the expense of healthy hematopoiesis. However, little is known about the bone marrow architecture in humans and its impact on clinical outcome. Here, we dissect the bone marrow niche in patients with acute myeloid leukemia (AML) at first diagnosis. We combined immunohistochemical stainings with global gene expression analyses from these AML patients and correlated them with clinical features. Mesenchymal stem and progenitor cells (MSPCs) lost quiescence and significantly expanded in the bone marrow of AML patients. Strikingly, their HSC- and niche-regulating capacities were impaired with significant inhibition of osteogenesis and bone formation in a cell contact–dependent manner through inhibition of cytoplasmic β-catenin. Assessment of bone metabolism by quantifying peripheral blood osteocalcin levels revealed 30% lower expression in AML patients at first diagnosis than in non-leukemic donors. Furthermore, patients with osteocalcin levels ≤11 ng/mL showed inferior overall survival with a 1-year survival rate of 38.7% whereas patients with higher osteocalcin levels reached a survival rate of 66.8%. These novel insights into the human AML bone marrow microenvironment help translate findings from preclinical models and detect new targets which might pave the way for niche-targeted therapies in AML patients.

Induction of a Leukemic Bone Marrow Niche Mediated By TGFB1 of Leukemic Blasts in Pediatric Acute Mekaryoblastic Leukemia: Results of an in Vitro and In Vivo Analysis

Background: An increasing knowledge about the bone marrow niche demonstrates the high complexity of leukemogenesis. Mesenchymal stromal cells (MSC) are important members of the bone marrow niche and source of fibrosis. Further, the microenvironment seems to be regulated by megakaryocytes and platelets via cytokines, such as transforming growth factor beta 1 (TGFB1). Despite extensive research, the pathogenesis of the bone marrow niche in childhood leukemia and the therapeutic potential is still unclear. We focus on acute childhood megakaryoblastic leukemia (AMKL) as a disease model and include patients with (ML-DS) and without Down syndrome. Based on similar clinical progressions - myelofibrosis occurs as a side-effect of both leukemia subtypes; these two diseases suit to characterize the leukemic bone marrow niche. Methods: We performed a comprehensive characterisation of MSC from ML-DS (n=9), AMKL patients (n=5) and healthy donors (HD; n=6) via e.g. differentiation assays (adipogenic, osteogenic), gene expression profiles and western blot analysis. In addition, we established an in vivo model with humanized ossicles, representing a human bone marrow microenvironment (as described by Chen et al. 2012; Reinisch et al. 2015): We injected MSC mixed with pooled human umbilical vein endothelial cells (HUVEC) and Matrigel subcutaneously into NOD scid gamma (NSG) mice. After 8 weeks, the engrafted ossicles were injected with megakaryoblastic cells (CMK cell line); injected ossicles (n=16); uninjected ossicle (n=27), MSC from ML-DS (n=19 ossicles), AML M1 (n=15 ossicles) and HD (n=9 ossicles). After 4 more weeks, histopathology evaluation of fibrosis in the ossicles was performed in accordance with the European Consensus on Grading Bone Marrow criteria from an independent pathologist. Results: The detailed characterisation of MSC with ML-DS and AMKL demonstrated a high similarity to MSC of HD: morphology, osteogenic differentiation potential, colony forming unit-fibroblast assay, proliferation and gene expression profiles. However, two differences emerged in our analysis: MSC showed a decreased adipogenic differentiation potential in ML-DS and AMKL compared to HD (ML-DS vs. HD=0.26-fold, p<0.05; AMKL vs. HD=0.50-fold). Gene expression profiling identified an upregulation of IGF2BP3, an oncofetal RNA binding protein, in MSC of ML-DS compared to HD confirmed by qRT-PCR (2.6-fold, p<0.05). IGF2BP3 is known to be highly expressed in many cancers and seems to be associated with proliferation. The increased level of IGF2BP3 (protein: IF2B3) was confirmed at protein-level detected by western blot analysis (ML-DS vs HD: 37.3-fold, p<0.05 and AMKL-MSC vs HD: 13.1-fold, p<0.05). TGFB1 - known to be secreted by leukemic megakaryoblasts - induced a fibrotic state in MSC regardless of their origin indicated by decreased adipogenic differentiation potential (treated vs. untreated: ML-DS 0.22-fold; AMKL 0.08-fold; HD 0.06-fold, p<0.05) and increased expression of collagen genes (qRT-PCR; COL1A1: ML-DS=1.63-fold, AMKL=1.80-fold (p<0.01), HD=1.66-fold (p<0.05); COL3A1: ML-DS=1.31-fold, AMKL=1.52-fold (p<0.05), HD=1.24-fold). The humanized bone marrow niche in our mouse model demonstrated a development of myelofibrosis after injection of the megakaryoblastic cell line (CMK): Grade 1 or 2 in 81% of the ossicles. The induction was independent of the MSC entity (HD/ML-DS). Of note, a monocytic subpopulation, which engrafted unexpectedly in ossicle from HD-MSC (n=3 ossicle), did not induce fibrotic fibers. Conclusion: Our data impressively illustrate the mutual influence between MSC and leukemic blasts that leads to a fibrotic microenvironment. This correlation has been observed in vitro but also in a unique mouse model. The interaction of MSC and leukemic blasts seems to be the key factor for the development of the leukemic niche in AMKL mediated inter alia by the TGFB pathway. However, we could identify several disease specific characteristics of MSC. Our model offers a unique opportunity to fundamentally examine of the leukemic niche in order to subsequently evaluate the potential therapeutic use in further studies. Disclosures Reinhardt: Novartis: Other: Participation in Advisory Boards; CSL Behring: Research Funding; Jazz: Other: Participation in Advisory Boards, Research Funding; Roche: Research Funding.

Abstract 3756: Highly multiplexed proteomic assessment of the human acute myeloid leukemia bone marrow microenvironment

Abstract Acute myeloid leukemia (AML) is a genetically heterogenous and often fatal cancer of the hematopoietic system. Even after achieving an initial complete remission, more than half of such AML patients experience clinical relapse due to the persistence of “minimal” residual disease (MRD). Ex-vivo studies have hypothesized that the interactions between residual leukemic cells and the local microenvironment in the bone marrow may play a key role in their survival and chemoresistance. Therefore, we performed for the first time a global examination of the proteomic profile of the bone marrow microenvironment in AML patients using a highly multiplexed method based on the ability of slow off-rate modified aptamers (modified small single-stranded oligonucleotides) to bind target proteins with high specificity and affinity at slow dissociation rates. Ten relapsed/refractory adult AML patients and age-matched healthy subject controls were recruited, under an IRB-approved protocol, for research bone marrow aspirate (BMA) and blood serum collection. The supernatant resulting from centrifugation of BMA and serum samples were processed and analyzed on a SOMAscan™ hybridization microarray platform for the detection and quantification of 1,305 target proteins (Somalogic, CO). Data was corrected using Hybridization Control and Median Signal Normalization methods. Significant differences were found between AML and healthy donor bone marrow, such that 133 analytes were differentially expressed in the AML group (Wilcoxon rank sum test FDR p<0.05, fold change >1.5); 85 over-expressed and 48 under-expressed. In addition to proteins already known to be elevated in AML patients (eg: Erythropoietin, Thrombopoietin, Hepcidin and Ferritin) and dysregulation of pathways previously identified as disease relevant (eg: Arginase), we also discovered multiple statistically significant differences in levels of soluble proteins that are not currently known to be associated with AML pathogenesis or treatment. Comparative analysis between blood serum and bone marrow determined that 82 of the 133 candidates have differential expression that was specifically restricted to the bone marrow. The STRING database was queried for pathway analysis of enriched protein sets in the AML group and clustered analytes with molecular functions including cytokine activity (GO:0005125, n=11, p=1.93e-08), cytokine receptor binding (GO:0005126, n=12, p=3.17e-08) and signal transducer activity (GO:0004871, n=19, p=9.23e-05). Using a high-throughput proteomic technology we have identified an AML bone marrow microenvironment-specific profile comprised of both proteins with known implications in leukemic pathogenesis and also several novel candidates from biologically plausible pathways that, once validated, may provide mechanistic insight and opportunity for therapeutic targeting. Citation Format: Bogdan Popescu, Katherine Lindblad, Giovanna Fantoni, Gege Gui, Janet Valdez, Meghali Goswami, Christin DeStefano, Catherine Lai, Angélique Biancotto, Julián Candia, Foo Cheung, Julie Thompson, Laura W. Dillon, Christopher S. Hourigan. Highly multiplexed proteomic assessment of the human acute myeloid leukemia bone marrow microenvironment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3756.

Acute Myeloid Leukemia Induced Remodelling of the Human Bone Marrow Niche Predicts Clinical Outcome

The hematopoietic stem cell (HSC) niche consists of different cellular and non-cellular constituents which regulate HSC maintenance and retention in the bone marrow. It has been shown in a number of murine models of myeloid neoplasia how leukemia infiltration alters the HSC niche to reinforce malignancy. Acute myeloid leukemia (AML) is characterized in human by a high relapse rate indicating that leukemia initiating cells are protected by its niche. However, despite our knowledge in murine models little is known about the bone marrow architecture in human and the impact of the leukemic niche on clinical outcome.In this study, we combined immunohistochemical stainings (IHC) with protein and global gene expression analyses together with clinical data to dissect the human bone marrow architecture in AML and assess its clinical impact. Human bone marrow was collected from AML patients at first diagnosis and matching non-leukemic donors. To evaluate the bone marrow architecture CD271+ mesenchymal stem and progenitor cells (MSPCs) were automatically quantified on bone marrow sections. In fact, AML patients showed 1.5-fold increase in bone marrow MSPCs compared to non-leukemic donors (Median (IQR), AML: 5.5% (2.8-9.5), n=36; control: 3.7% (2.1-5.7), n=58; p < 0.01). MSPCs proved to produce reticular fibers, an extracellular matrix protein frequently associated with different malignancies. In AML bone marrow these fibers were also found to be more abundantly expressed (Median (IQR), AML: 3.4% (1.8-4.5), n=37; control: 1.6% (1.1-3.3), n=19; p < 0.05).Next, to globally assess the gene expression profile of MSPCs in AML bone marrow we performed microarray analyses (ClariomTM S Human Assay) of freshly isolated uncultured lineage- CD146+ CD271+ MSPCs. Strikingly, HSC-regulating genes in particular CXCL12, ANGPT1 and VCAM1 showed lower expression in AML MSPCs which correlated with the degree of hematopoietic failure in AML patients. Along with the increased number of MSPCs, geneset enrichment analysis (GSEA) revealed higher proliferation of MSPCs in AML. In murine models loss of quiescence of MSPCs was previously found to be due to bone marrow sympathetic neuropathy. We therefore measured catecholamines and neurotrophic factor in the bone marrow extracellular fluid of AML patients and non-leukemic donors at first diagnosis. In fact, noradrenalin and brain-derived neurotrophic factor (BDNF) showed a 2-fold (p=0.26) resp. 4-fold (p<0.0001) lower expression in AML bone marrow. Importantly, BDNF is proved to be essential for sympathetic neuron proliferation and differentiation.In order to get an overview of alterations of canonical pathways in bone marrow MSPCs upon AML infiltration, we applied QIAGEN's Ingenuity® Pathway Analysis software. Several of the major differently regulated pathways proved to involve differentiation and mineralization of MSPCs. We therefore assessed bone metabolism in AML patients at first diagnosis and quantified serum osteocalcin levels. Notably, AML patients showed 30% lower osteocalcin levels than non-leukemic donors (Median (IQR), AML, 12.15ng/ml (7.53-16.28) n=58; control, 17.2ng/ml (12.5-23.45) n=31; p < 0.05). To evaluate if the deficiency in osteoblast mineralization is specifically due to AML infiltration we performed in vitro co-culture assays. Both MSPCs and an osteoblast-like cell line (SaOS2) showed significant impaired mineralization in presence of certain AML cell lines as well as primary human AML cells, while healthy mononuclear cells did not affect mineralization. Strikingly, this AML-induced defect in osteoblast mineralization proved to be of clinical significance. Patients with low osteocalcin levels (<11ng/ml) showed inferior overall survival with 1-year survival rate of 38.7% while patients with high osteocalcin levels reached 66.8% (n=58; median duration of follow-up 9.7 months).In summary, we globally characterized the bone marrow architecture in AML patients in comparison to non-leukemic donors and assessed its clinical significance. This increasing understanding of the human AML bone marrow microenvironment might open the window for new niche-targeted therapies to eradicate leukemic stem cells and eventually decrease the high relapse rate in AML. DisclosuresDuehrsen:Amgen: Research Funding; Roche: Honoraria, Research Funding; Gilead: Consultancy, Honoraria; Celgene: Honoraria, Research Funding; AbbVie: Consultancy, Honoraria; Janssen: Honoraria.

Bioengineering of Humanized Bone Marrow Microenvironments in Mouse and Their Visualization by Live Imaging.

Human hematopoietic stem cells (HSCs) reside in the bone marrow (BM) niche, an intricate, multifactorial network of components producing cytokines, growth factors, and extracellular matrix. The ability of HSCs to remain quiescent, self-renew or differentiate, and acquire mutations and become malignant depends upon the complex interactions they establish with different stromal components. To observe the crosstalk between human HSCs and the human BM niche in physiological and pathological conditions, we designed a protocol to ectopically model and image a humanized BM niche in immunodeficient mice. We show that the use of different cellular components allows for the formation of humanized structures and the opportunity to sustain long-term human hematopoietic engraftment. Using two-photon microscopy, we can live-image these structures in situ at the single-cell resolution, providing a powerful new tool for the functional characterization of the human BM microenvironment and its role in regulating normal and malignant hematopoiesis.

Bioengineering of Humanized Bone Marrow Microenvironments in Mouse and Their Visualization by Live Imaging

Human hematopoietic stem cells (HSCs) reside in the bone marrow (BM) niche, an intricate, multifactorial network of components producing cytokines, growth factors, and extracellular matrix. The ability of HSCs to remain quiescent, self-renew or differentiate, and acquire mutations and become malignant depends upon the complex interactions they establish with different stromal components. To observe the crosstalk between human HSCs and the human BM niche in physiological and pathological conditions, we designed a protocol to ectopically model and image a humanized BM niche in immunodeficient mice. We show that the use of different cellular components allows for the formation of humanized structures and the opportunity to sustain long-term human hematopoietic engraftment. Using two-photon microscopy, we can live-image these structures in situ at the single-cell resolution, providing a powerful new tool for the functional characterization of the human BM microenvironment and its role in regulating normal and malignant hematopoiesis.

Genetically engineered mesenchymal stromal cells produce IL-3 and TPO to further improve human scaffold-based xenograft models

Recently, NOD-SCID IL2Rγ-/- (NSG) mice were implanted with human mesenchymal stromal cells (MSCs) in the presence of ceramic scaffolds or Matrigel to mimic the human bone marrow (BM) microenvironment. This approach allowed the engraftment of leukemic samples that failed to engraft in NSG mice without humanized niches and resulted in a better preservation of leukemic stem cell self-renewal properties.To further improve our humanized niche scaffold model, we genetically engineered human MSCs to secrete human interleukin-3 (IL-3) and thrombopoietin (TPO). Invitro, these IL-3- and TPO-producing MSCs were superior in expanding human cord blood (CB) CD34+ hematopoietic stem/progenitor cells. MLL-AF9-transduced CB CD34+ cells could be transformed efficiently along myeloid or lymphoid lineages on IL-3- and TPO-producing MSCs. Invivo, these genetically engineered MSCs maintained their ability to differentiate into bone, adipocytes, and other stromal components. Upon transplantation of MLL-AF9-transduced CB CD34+ cells, acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) developed in engineered scaffolds, in which a significantly higher percentage of myeloid clones was observed in the mouse compartments compared with previous models. Engraftment of primary AML, B-cell ALL, and biphenotypic acute leukemia (BAL) patient samples was also evaluated, and all patient samples could engraft efficiently; the myeloid compartment of the BAL samples was better preserved in the human cytokine scaffold model. In conclusion, we show that we can genetically engineer the ectopic human BM microenvironment in a humanized scaffold xenograft model. This approach will be useful for functional study of the importance of niche factors in normal and malignant human hematopoiesis.

Abstract B18: B-cell lymphoma patient-derived xenograft models: The road to personalized therapy

Abstract Lymphoma is the most common hematological malignancy, and B-cell lymphoma accounts for 85% of all lymphomas. The current overall cure rate for B-cell lymphoma is estimated at ~30%, even with the development and application of novel therapies, and the majority of patients relapse after treatment due to the development of drug resistance. The evaluation of novel drug targets using established B-cell lymphoma cell lines is limited by the inexact correlation between responsiveness observed in the cell line versus the patient sample. However, patient-derived xenograft (PDX) mouse models have been shown to recapitulate the diversity of growth, metastasis, and histopathology of the original tumor, overcoming the limitations of cell lines. Furthermore, recent studies have indicated that PDXs can recapitulate treatment responses of the parental tumor and can successfully choose a therapeutic target and regimen for the patient. We previously established a SCID-hu in vivo human primary mantle cell lymphoma (MCL) PDX model, the first human primary MCL animal model for biological and therapeutic research, and investigated the disease biology and potential novel human MCL therapies using this model. In this MCL PDX model, the engraftment and growth of patient MCL cells were dependent on the human bone marrow microenvironment supplied by an implanted human fetal bone chip. Our clinical information and reports show that numerous B-cell lymphoma subtypes involve bone marrow; therefore, we expanded our PDX model to include various B-cell lymphomas to study the clonal evolution of tumors and to evaluate novel therapies for the treatments of these diseases. PDX models (n=20) were established for multiple different types of B-cell lymphoma, including MCL (n=12), Burkitt's lymphoma (n=1), marginal zone lymphoma (n=2), follicular lymphoma (n=2), chronic lymphocytic leukemia (n=1) and diffuse large B-cell lymphoma (n=2). The engraftment rate was high at 95%, the tumor xenografts rapidly grew at 3-4 weeks/generation, and 68% of the tumor xenografts were passaged for multiple generations, even up to 14 generations. Further demonstrating the utility of this model to recapitulate the characteristics of the original patient tumor, the tumor xenograft cells migrated to the lymph nodes, spleen, bone marrow, and gastrointestinal tract of the host mice, mimicking the disease progression observed in humans, and H&amp;E staining showed similar histological features between the original patient tumor and the tumor xenograft cells such as splenomegaly. To identify novel targeted treatments for these various lymphomas, we assessed the in vitro cell viability of MCL, DLBCL, and Burkitt's lymphoma cells from the patients and isolated from the PDXs of different generations. The responses to these agents were very similar between the PDXs and the original patient tumor, and the treatment responses were the same between different generations, showing that the PDX models recapitulate the drug response of the original tumor. Finally, in 34% of relapsed/refractory MCL patients who were treated with ibrutinib, transient lymphocytosis was observed, and this phenomenon was also observed in our MCL PDX mouse models treated with ibrutinib. Furthermore, the combination of ibrutinib and rituximab quenched lymphocytosis in the animal model, which provided the basis of using this combination in a clinical trial in which the MCL patients had an overall response rate of 88%. In conclusion, these data demonstrate the ability of PDX mouse models to successfully recapitulate and the characteristics of the original patient tumor as well as mimic the progression of the disease in vivo. In addition, the drug responses of the original patient tumor to various agents were similar to the responses of the tumor xenograft, demonstrating that these PDX models can be used as a way to personalize the treatment of numerous types of B-cell lymphomas. Citation Format: Leo Zhang, Lan Pham, Hui Zhang, Jingmeng Xie, Wenjing Tao, Taylor Bell, Zhihong Chen, Bingliang Fang, Michael Wang, Krystle Nomie. B-cell lymphoma patient-derived xenograft models: The road to personalized therapy. [abstract]. In: Proceedings of the AACR Special Conference: Patient-Derived Cancer Models: Present and Future Applications from Basic Science to the Clinic; Feb 11-14, 2016; New Orleans, LA. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(16_Suppl):Abstract nr B18.

Involvement of urokinase receptor in the cross-talk between human hematopoietic stem cells and bone marrow microenvironment.

Hematopoietic stem cells (HSCs) reside in bone marrow (BM) and can be induced to mobilize into the circulation for transplantation. Homing and lodgement into BM of transplanted HSCs are the first critical steps in their engraftment and involve multiple interactions between HSCs and the BM microenvironment.uPAR is a three domain receptor (DIDIIDIII) which binds urokinase, vitronectin, integrins. uPAR can be cleaved and shed from the cell surface generating full-length and cleaved soluble forms (suPAR and DIIDIII-suPAR). DIIDIII-suPAR can bind fMLF receptors through the SRSRY sequence (residues 88-92).We previously reported the involvement of soluble uPAR in HSC mobilization. We now investigate its possible role in HSC homing and engraftment.We show similar levels of circulating full-length suPAR in healthy donors and in acute myeloid leukemia (AML) patients before and after the pre-transplant conditioning regimen. By contrast, levels of circulating DIIDIII-suPAR in AML patients are higher as compared to controls and significantly decrease after the conditioning.We found that suPAR and uPAR84-95, a uPAR-derived peptide which mimics active DIIDIII-suPAR, induce a significant increase in Long Term Culture (LTC)-Initiating Cells (ICs) and in the release of clonogenic progenitors from LTCs of CD34+ HSCs. Further, suPAR increases adhesion and survival of CD34+ KG1 AML cells, whereas uPAR84-95 increases their proliferation.Thus, circulating DIIDIII-suPAR, strongly increased in HSC mobilization, is indeed down-regulated by pre-transplant conditioning, probably to favour HSC homing. BM full-length suPAR and DIIDIII-suPAR may be involved in HSC lodgement within the BM by contributing to a suitable microenvironment.

A humanized bone marrow ossicle xenotransplantation model enables improved engraftment of healthy and leukemic human hematopoietic cells.

Xenotransplantation models provide powerful tools for the investigation of human normal and malignant hematopoiesis. However, current models do not extensively recapitulate human bone marrow (BM) microenvironment components and exhibit limited engraftment of many human leukemias and other disorders. Here, we describe a xenotransplantation model bearing subcutaneous humanized ossicles with an accessible BM microenvironment formed by in situ differentiation of human BM-derived mesenchymal stromal cells. In these humanized ossicles, we observed robust engraftment of normal human hematopoietic stem and progenitor cells, and detected extensive engraftment of diverse primary acute myeloid leukemia samples at levels much greater than in unmanipulated mice. Direct intraossicle transplantation accelerated engraftment and resulted in substantially higher leukemia-initiating cell frequencies. We also observed robust engraftment of acute promyelocytic leukemia and myelofibrosis leading to the identification of leukemia-initiating cells in hematopoietic stem cells in myelofibrosis. This humanized ossicle xenotransplantation approach provides a novel system to model human hematologic disease.

Novel In Vivo Model of Multiple Myeloma Bone Disease Using ß-Tricalcium Phosphate Artificial Bone Grafts

Abstract Model systems to study Multiple Myeloma (MM) related bone disease exist but have a number of limitations. Disseminated MM models have variable cell homing and do not precisely recapitulate the human microenvironment interactions with myeloma cells. Severe combined immunodeficiency (SCID) mice engrafted with human fetal bone (SCID-hu) have been used by us, and are able to recapitulate the human bone marrow microenvironment. The fetal bone chips are however difficult to obtain, and vary in size and shape, complicating inter-sample comparison. Similarly, the poly-ε-caprolactone polymeric scaffold, previously used to seed murine or human stromal compartment, may not correctly reproduce bone destruction and inhibition of osteogenesis by MM as seen in patients, making this model difficult to test therapies targeting the MM niche. β-tricalcium phosphate (β-TCP) is a biocompatible and biodegradable bone graft substitute that is uniform in structure and easily available, and may be a viable alternative to overcome SCID-hu difficulties in modeling MM bone disease. Here, we utilized β-TCP bone graft substitute to develop a novel in vivo MM model where β-TCP permits the development of the bone microenvironment, supports MM development, and is technically feasible and highly reproducible. Using this model, we aim to better understand the biology of the niche in MM by genetically modifying its components and by testing new niche-targeting therapies. Our initial results show that osteogenesis takes place in the β-TCP bone graft, and the implant is supportive of MM tumor growth. Inter-scapular subcutaneous implantation of β-TCP alone, or co-implantation with human-derived stromal cell line HS27A in immunocompromised recipients resulted in the expression of osteogenic markers Runx2, alkaline phosphatase (ALP), Col1A1, and Osteocalcin (OCN), as well as a marker of bone resorption. Further, implants supported the growth of human-derived MM1.S and murine 5TGM1 cells, as visualized directly in vivo by serial luciferase bioluminescence imaging (BLI) and by immunohistochemistry. Modifying the niche compartment in Cre/iDTR animals with MM disease is an exciting novel strategy to understand which niche component in vivo may be targeted to suppress MM development. Mouse strains with promoter-specific Cre recombinase that induces the expression of the diphtheria toxin (DT) receptor (iDTR) can be utilized to selectively ablate a cell population of interest in vivo, via intraperitoneal DT injection. Here, we first utilized OCN-Cre/iDTR mice to test the deletion of mature osteoblasts in β-TCP artificial bone graft post-implantation. Our data show a dose-dependent reduction in osteoblastic markers OCN, ALP, Runx2, Sclerostin, Osteoprotegerin and RANKL. Importantly, DT ablation of osteoblasts in the β-TCP implant resulted in a significantly increased 5TGM1 tumor growth, as judged by BLI and tumor weight. Our data show that the mature osteocalcin-positive niche population is protective against MM disease. Ongoing studies of the β-TCP mouse model will address the relative contribution of various osteogenic populations to the course of MM development in vivo, and test the efficacy of novel MM drugs. Disclosures Raje: BMS: Consultancy; Amgen: Consultancy; Celgene Corporation: Consultancy; Takeda: Consultancy; Onyx: Consultancy; Takeda: Consultancy; Amgen: Consultancy; Onyx: Consultancy; BMS: Consultancy; AstraZeneca: Research Funding; Eli Lilly: Research Funding; AstraZeneca: Research Funding; Millenium: Consultancy; Eli Lilly: Research Funding; Novartis: Consultancy; Acetylon: Research Funding; Millenium: Consultancy; Novartis: Consultancy; Acetylon: Research Funding.