Bone marrow transplants of hematopoietic stem cells have become standard treatment for a host of conditions including cancers of the blood and lymphatic systems, sickle cell anemia, inherited metabolic disorders, and radiation damage.
Unfortunately, many bone marrow transplants fail due to rejection by the patient’s immune system or graft-versus-host disease (in which the transplanted marrow cells attack the patient’s healthy cells), both of which can be fatal.
Mesenchymal stromal cells (MSCs) are known to secrete compounds that modulate the immune system and have shown promise in mitigating these problems in animal trials.
However, clinical results with MSCs have been disappointing thus far, as they are rapidly cleared from the body and can draw attack from patients’ immune systems, and efforts to encapsulate MSCs in protective biomaterials have resulted in large, bulky hydrogels that cannot be given intravenously and compromise the cells’ functions.
Today, in a scientific first, researchers from the Wyss Institute for Biologically Inspired Engineering, Harvard’s John A. Paulson School of Engineering and Applied Sciences (SEAS), and the Harvard Stem Cell Initiative (HSCI) demonstrate a single-cell encapsulation technology that effectively protects transplanted MSCs from clearance and immune attack and improves the success of bone marrow transplants in mice.
The work is published in PNAS.
“To our knowledge, this is the first example of single-cell encapsulation being used to improve cell therapies, which are becoming more widespread as treatments for a number of diseases,” said first author Angelo Mao, Ph.D., a former graduate student in the lab of Wyss Core Faculty member and lead of the Wyss Immuno-Materials Platform David Mooney, Ph.D. who is now a postdoc with Wyss Core Faculty member James Collins, Ph.D. “And, our encapsulated cells can be frozen and thawed with minimal impact on the cells’ performance, which is critical in the context of hospitals and other treatment centers.”
This advance builds on a method the team previously developed that uses a microfluidic device to coat individual living cells with a thin layer of an alginate-based hydrogel, creating what they term “microgels.”
The process encapsulates cells with 90% efficiency, and the resulting microgels are small enough that they can be delivered intravenously, unlike the bulky hydrogels created by other methods.
When injected into mice, MSCs encapsulated using this technique remained in the animals’ lungs ten times longer than “bare” MSCs, and remained viable for up to three days.
Because a large amount of MSCs’ clinical appeal lies in their secretion of compounds that modulate the body’s immune system, the researchers needed to test how microgel encapsulation affects MSCs’ ability to function and resist immune attack.
They modified their original alginate microgel by adding another compound that cross-links to the alginate and makes the microgel stiffer and better able to resist the body’s immune system and clearance mechanisms.
They also cultured the MSCs after encapsulation to encourage them to divide and produce more cells.
When these new microgels were injected into mice, their persistence increased five-fold over the previous microgel design and an order of magnitude over bare MSCs.
To induce an immune response against the MSCs, the team incubated encapsulated cells in a medium containing fetal bovine serum, which is recognized by the body as foreign, before introducing them into mice.
While the clearance rate of the encapsulated MSCs was higher than that observed without immune activation, it was still five times lower than that of bare MSCs.
The microgels also outperformed bare MSCs when injected into mice that had a preexisting immune memory response against MSCs, which mimics human patients who are given multiple infusions of stem cells.
MSCs exposed to inflammatory cytokines respond by increasing their expression of immune-modulating genes and proteins, so the researchers next tested whether encapsulation in their new microgels impacted this response.
They found that bare and encapsulated MSCs had comparable levels of gene expression when exposed to the same cytokines, demonstrating that the microgels did not impair MSC performance.
For their pièce de résistance, the team injected their MSC-containing microgels into mice along with transplanted bone marrow, half of which was immune-compatible with the recipient mouse and half of which was allogeneic, or an immune mismatch.
Mice that received encapsulated MSCs had more than double the fraction of allogeneic bone marrow cells in their marrow and blood after nine days compared with mice that did not receive MSCs.
Encapsulated MSCs also led to a greater degree of engraftment of the allogeneic cells into the host bone marrow compared to bare MSCs.
“One of the strong points of this work is that it uses a completely non-genetic approach to dramatically increase cell survival in transplant contexts, where it’s sorely needed,” said Mooney, who is also the Robert P. Pinkas Family Professor of Bioengineering at SEAS.
“This technology nicely complements genetic engineering approaches, and in fact could be more efficient than attempting to directly modify immune cells themselves.”
The Wyss Institute’s Validation Project Program is supporting advancement of this approach as a possible treatment for ischemia (narrowing of blood vessels) in human patients, and hopes to demonstrate clinical viability in the near future.
Validation Projects are technologies with potential high-impact applications that have successfully progressed through significant concept refinement and meet predefined technical, product development, and intellectual property criteria.
“This technology simultaneously resolves multiple issues with bone marrow transplants and stem cell therapies using an elegant, biomaterials-based approach that represents the kind of cross-disciplinary thinking that we value so greatly at the Wyss Institute,” said Wyss Founding Director Donald Ingber, M.D., Ph.D., who is also the Judah Folkman Professor of Vascular Biology at HMS and the Vascular Biology Program at Boston Children’s Hospital, as well as Professor of Bioengineering at SEAS. “We are excited to support this project as it moves toward clinical validation, and we look forward to other potential applications of microencapsulation to address drug and cell delivery problems.”
By merit of their regenerative secretome and their capacity for differentiation toward multiple mesenchymal lineages, the fibroblastic cell type termed mesenchymal stromal/stem cells (MSCs) shows promise for a wide range of tissue engineering and regenerative medicine applications (Figure 1).
As a result of their therapeutic versatility and the multitude of promising clinical results thus far, MSCs are poised to become an increasingly significant cell source for regenerative therapies as medicine evolves to focus on personalized and cell-based therapeutics.
Given their emerging importance, this review aims to provide an overview of historical and ongoing work aimed at understanding and better utilizing these cells for therapeutic purposes.
Initial Discoveries and the Evolving Definition of “MSC”
The initial discovery of MSCs is attributed to Friedenstein et al. who discovered a fibroblastic cell type derived from mouse and guinea pig bone marrow that could produce clonal colonies capable of generating bone and reticular tissue when heterotopically transplanted [1, 2].
The subsequent discovery that colonies of this cell type can generate cartilage and adipose tissue, in addition to bone, gave rise to the descriptor mesenchymal stem cells, as originally coined by Arnold Caplan .
Finally, Pittenger et al. established that human bone marrow also contains a subpopulation of stromal cells that are genuinely multipotent stem cells by demonstrating single colonies have trilineage mesenchymal potential .
Over time, the acronym MSC has come to take on multiple meanings including, mesenchymal stem cell, mesenchymal stromal cell, and multipotent stromal cell.
To help clarify this, the International Society for Cellular Therapy (ISCT) has officially defined MSCs as multipotent mesenchymal stromal cells and suggests this to mean the plastic-adherent fraction from stromal tissues, while reserving the term mesenchymal stem cells to mean the subpopulation that actually has the two cardinal stem cell properties (i.e., self-renewal and the capacity to differentiate down multiple lineages) .
Furthermore, ISCT has also defined MSCs as meeting several criteria including (i) being plastic adherent, (ii) having osteogenic, adipogenic, and chondrogenic trilineage differentiation potential, (iii) and being positive (>95%) and negative (<2%) for a panel of cell surface antigens.
Positive markers for human MSCs include CD73 (also present on lymphocytes, endothelial cells, smooth muscle cells, and fibroblasts), CD90 (also present on hematopoietic stem cells, lymphocytes, endothelial cells, neurons, and fibroblasts), and CD105 (also found on endothelial cells, monocytes, hematopoietic progenitors, and fibroblasts) .
Negative markers include CD34 (present on hematopoietic progenitors and endothelial cells), CD45 (a pan-leukocyte marker), CD14 or CD11b (present on monocytes and macrophages), CD79-α or CD19 (present on B cells), and HLA-DR unless stimulated with IFN-γ (present on macrophages, B cells, and dendritic cells) .
As these elaborate inclusionary and exclusionary criteria highlight, no single MSC-specific epitope has been discovered, unlike for some other stem cell populations (e.g., LGR5, which labels resident stem cells in hair follicles and intestinal crypts) [8, 9].
This unfortunate lack of a single definitive marker continues to confound the interpretation of a broad range of studies given that sorting out the canonical MSC population from the adherent fraction is rarely done, leading to the perennial question of which subpopulation in the adherent stromal fraction is actually eliciting the observed effects.
This lack of a definitive MSC marker has also contributed to the challenge of delineating the exact in vivo location, function, and developmental origin of MSCs.
MSC Adult Anatomical Location
In the bone marrow, where MSCs were first discovered, MSCs have been reported to typically localize near the sinusoidal endothelium in close association with the resident hematopoietic stem cells (HSCs) [14, 15].
In addition to serving as osteogenic progenitors, such MSCs have been shown to play an important role in regulating HSC function by maintaining the HSC niche and by secreting trophic factors such as angiopoietin 1 (Ang1), stem cell factor (SCF), and CXC ligand 12 (CXCL12) .
Such a broad anatomical distribution would suggest a common and ubiquitous MSC niche exists throughout the body.
Indeed, evidence suggests that many MSC populations are specifically located near blood vessels and are in fact a subpopulation of pericytes that reside on capillaries and venules .
Supporting observations include the fact that pericytes and MSCs express similar surface antigens, and that cells in perivascular positions were found to express MSC markers in human bone marrow and dental pulp [16, 21].
Perhaps most definitively, Crisan et al. found that cells positive for NG2, CD146, and PDGFR-βspecifically stained pericytes in multiple human tissues, and when cells with these markers were isolated, they were shown to have trilineage potential in vitro and were osteogenic once transplanted in vivo .
The converse, that all pericytes are MSCs, is not thought to be the case .
In addition to being abluminal to microvessels, it should be noted that a Gli1+ MSC-like population has also been found to reside within the adventitia of larger vessels in mice.
Similarly, a MSC population with a CD34+ CD31− CD146− CD45− phenotype has been discovered to reside within the adventitia of human arteries and veins suggesting that not all perivascular MSCs are pericyte-like cells in humans .
Furthermore, a MSC population has also been isolated from the perivascular tissue of umbilical cords (human umbilical cord perivascular cells (HUCPVCs)) which shows promise for tissue engineering applications given the cells’ noninvasive extraction and their relatively high abundance and proliferative capacity, compared to bone marrow-derived MSCs [26–28].
Finally, despite the prevalent view that MSCs reside in perivascular niches, some MSC populations may reside in avascular regions as well.
For example, a lineage tracing study focused on murine tooth repair demonstrated that while some odontoblasts descend from cells expressing the pericyte marker, NG2, the majority of odontoblasts did not, suggestive of a nonpericyte origin (or at least not from NG2-positive pericytes) .
However, there are fenestrated capillaries localized near the synovial surface , and diseased sclerotic and stenotic valves can be partially vascularized [35, 36], raising the possibility of MSCs trafficking from one anatomical location to another (e.g., synovium-associated vasculature to avascular cartilage) and innate differences in the local presence or absence of perivascular MSCs.
Future work focused on these questions will have important implications for understanding disease progression and potential regenerative avenues.
MSC Developmental Origins
Presently, there are considered to be multiple developmental origins of MSCs. Unsurprisingly, given their mesenchymal differentiation potential, certain subsets of MSCs are derived from mesodermal precursors, such as lateral plate mesoderm- (LPM-) derived mesoangioblast cells from the embryonic dorsal aorta [37, 38].
Support for this comes from the observation that mesoangioblast cells isolated from the mouse dorsal aorta and then grafted into chick embryos incorporated into several mesodermal tissues (bone, cartilage, muscle, and blood) .
Other reports suggest MSCs partly descend from a subpopulation of neural crest cells, with the remaining MSCs descending from unknown origins.
Support for this comes from the observation that a population of murine Sox1+ trunk neuroepithelial cells could undergo clonogenic expansion and maintain adipogenic, chondrogenic, and osteogenic differentiation in vitro .
This neural crest origin may help explain why MSCs have neural differentiation potential and why human bone marrow-derived MSCs can be enriched for using antibodies against nerve growth factor receptor [12, 38].
Given their lineage tracing results, the authors claimed that neural crest-derived MSCs are the earliest MSCs to arise in the embryo, but they did note that other MSCs must also arise later on in development as not all MSCs detected were found to be of a neural crest origin.
Corroborating this, a lineage tracing study using the promoter from Protein-0, a neural crest-associated marker, found that only a portion of bone marrow-derived MSCs were labeled in adult mice, suggestive of both a neural crest and nonneural crest origin .
It is possible that the indefinite nonneural crest source of MSCs observed in these studies may be mesoangioblasts or another mesoderm-derived cell type.
It has also been suggested that data indicative of a mesoangioblast origin may alternatively be explained by simply “contamination” of neural crest cells as the neural tube is close to dorsal aorta at day 9.5 .
With regard to human MSC origins, similar dual mesoderm and neural crest origins may also exist given that human iPSCs differentiated toward these two lineages can both give rise to MSC-like cells [42, 43].
Further study will be required to resolve these issues and to elucidate if any lasting functional dissimilarities exist between MSC subpopulations that arise from differing time periods and locations during development.
More information: Angelo S. Mao el al., “Programmable microencapsulation for enhanced mesenchymal stem cell persistence and immunomodulation,” PNAS (2019). www.pnas.org/cgi/doi/10.1073/pnas.1819415116
Journal information: Proceedings of the National Academy of Sciences
Provided by Harvard University