Thesis: Ph. D., Massachusetts Institute of Technology, Department of Biological Engineering, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 243-264). / Hematopoietic stem cell or bone marrow transplantation is a curative treatment for multiple hematologic malignancies. However, the myeloablative conditioning regimens preceding cell delivery have rendered the rapid and sustained hematopoietic recovery after transplantation an outstanding challenge. Successful long-term engraftment of hematopoietic stem cells is dependent largely on the surrounding stroma components or hematopoietic niche. Cell types within this niche that support hematopoietic recovery include two adherent cell types, mesenchymal stromal cells (MSCs) and vascular endothelial cells (VECs). The niche also contains many biophysical and mechanical cues including cell contractility against other cells or the matrix, pulsatile fluid flow, differences in localized niche stiffness, and occupation of fluid volume by macromolecules. This thesis aims to understand how VECs and MSCs respond to these cues ex vivo, and how these cues can be used to engineer VEC and MSC phenotypes that can predictably support hematopoietic recovery in vivo. VEC-mediated angiogenesis and angiocrine signaling are known to support hematopoietic recovery in vivo. In this thesis, we first explored how the biophysical cue of macromolecular crowding (MMC) and the mechanical cue of strain can regulate angiogenesis. The addition of synthetic MMC to in vitro cultures replicates the endogenous occupation of fluid space due to macromolecules. We explored how MMC affects the basement membrane formation of VECs, and determined that MMC can increase the deposition, areal spread, and alignment of basement membrane proteins. Even with the addition of biochemical signals from pericytes, this biophysical cue of MMC played a dominant role in the organization of the basement membrane. Pericytes that surround blood vessels and the basement membrane have been shown to exert contractile forces, which results in a hoop strain in the blood vessel wall. We translated this strain to in vitro VEC cultures by applying static, uniaxial strain to confluent VEC monolayers using a polydimethyl siloxane (PDMS) substrata, which allowed us to decouple the mechanical cue of pericytes from their chemical signaling. The application of 10% engineering strain was sufficient to induce cell-cycle re-entry in a quiescent monolayer. We then went on to demonstrate in a quasi-3D assay that straining the VECs also produced angiogeniclike sprouts. Together, these results show that biophysical and mechanical cues of the hematopoietic niche alone are sufficient to direct VEC-derived extracellular matrix formation and to induce angiogenic sprouting. Thus, future models of in vitro angiogenesis must include these cues to more comprehensively and accurately replicate the in vivo hematopoietic niche. Paracrine signaling from MSCs is crucial in regulating the self-renewal capacity and differentiation of hematopoietic stem and progenitor cells (HSPCs) that re-populate the bone marrow compartment in vivo. Thus, we then explored if and how to modulate MSC paracrine signaling or the MSC secretome. Like VECs, MSCs are known to respond to microenvironment cues such as substratum stiffness. We developed tissue-culture compatible PDMS-based substrata with tunable viscoelastic properties to assay potential mechanosensitivity. We characterized the bulk and surface properties of this substrata to verify that we could tune stiffness across three orders of magnitude without altering material surface biochemistry. When we expanded the MSCs on compliant substrata (elastic modulus ~I kPa), we found that we could increase the expression of osteopontin as well the expression of at least a dozen other secreted proteins without altering cell capacity for terminal differentiation. We observed changes in the MSC secretome that were significantly correlated to the viscoelastic properties (shear storage and loss moduli G' and G", respectively, and the ratio of G"/G' as tan [delta]) of the substratum material. These results suggested that we could mechanically modulate the MSC secretome using the viscoelastic properties of the extracellular substrata. Finally, we went on to explore how these mechanically modulated changes in MSC phenotype could regulate hematopoiesis in vitro and support hematopoietic recovery in vivo. To do so, we used statistical regression modeling (partial least squares regression or PLSR) to identify the components of the MSC secretome that were significantly correlated with improved radiation rescue and hematopoietic recovery in mouse models of hematopoietic failure. We then characterized the expression of these key secretome components in our mechanoprimed MSCs. The mechanoprimed MSCs expressed equal or higher concentrations of these proteins as a diameter-defined subpopulation of MSCs we previously identified to be therapeutically effective. Using the regression parameters from PLSR and the new expression data from our mechanoprimed MSCs, we then predicted how our mechanoprimed MSCs would elicit radiation recovery of the bone marrow compartment in vivo. From these computational predictions, we found that our mechanoprimed MSCs could potentially improve survival proportion in this in vivo model of hematopoietic failure. Thus, we tested mechanoprimed MSCs by expanding them in co-culture with HSPCs to determine if the MSCs could regulate hematopoiesis in vitro. We found that mechanoprimed MSCs could maximize the proliferation or expansion of HSPCs when co-cultured on top of our most compliant PDMS substrata (~I kPa). When grown on stiffer PDMS substrata (100 kPa), those MSCs could prime differentiation of the HSPCs down myeloid lineages, which include red blood cells. Together, these results demonstrate that these mechanoprimed MSCs can be used to modulate the ex vivo expansion and differentiation of HSPCs. Lastly, we tested these mechanoprimed MSCs in our sub-lethally irradiated mouse models of hematopoietic failure. Our mechanoprimed MSCs significantly increased the survival of the mice. Interestingly, this increased survival and improved hematopoietic recovery outperformed the survival predicted from our regression model. We also observed recovery of red blood cells, white blood cells, and platelets in mice treated with mechanoprimed MSCs, suggesting complete recovery of all hematopoietic lineages. In summary, we have explored how biophysical and mechanical cues can modulate VEC and MSC phenotype in vitro. In the case of VECs, the results presented in this thesis further the development of more accurate in vitro models of angiogenesis. Accurate in vitro models of angiogenesis are necessary to elucidate the mechanisms by which VECs regulate hematopoietic recovery in vivo. We also characterized the components of the MSC secretome correlated with improving hematopoietic recovery and demonstrated that we could engineer the expression of these same MSC secretome components using substratum viscoelastic properties. Lastly, we validated that these mechanically modulated MSCs led to improved survival outcome in vivo. The work presented in this thesis furthers our understanding of how biophysical and mechanical cues regulate hematopoietic niche components that participate in indirect repair of the bone marrow. We also demonstrated how these same cues can be applied in vitro to improve cell-based therapies for hematopoietic recovery in vivo. / by Frances D. Liu. / Ph. D.
| Identifer | oai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/119976 |
| Date | January 2018 |
| Creators | Liu, Frances D. (Frances Deen) |
| Contributors | Krystyn J. Van Vliet., Massachusetts Institute of Technology. Department of Biological Engineering., Massachusetts Institute of Technology. Department of Biological Engineering. |
| Publisher | Massachusetts Institute of Technology |
| Source Sets | M.I.T. Theses and Dissertation |
| Language | English |
| Detected Language | English |
| Type | Thesis |
| Format | 264 pages, application/pdf |
| Rights | MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission., http://dspace.mit.edu/handle/1721.1/7582 |
Page generated in 0.0023 seconds