Dr. Prockop began by reviewing characteristics of MSCs, including the capability to adhere to plastic, ready differentiation into three lineages in culture, the capability for rapid expansion, and a lack of tumorigenicity under normal culture conditions. He noted, however, that no definitive epitopes have been identified, and there is little consensus on many of the properties of the cells (including their name). In the 1960s and 1970s, Friedenstein and Tavassoli discovered adherent cells from marrow (MSCs) that were present in 1:10,000 to 1:100,000 nucleated cells. These cells grew rapidly and differentiated into bone, fat, and cartilage in culture and in vivo. Moreover, they proved easy to clone and differentiate without fusion into a host of cell types, including adipocytes, osteoblasts, and chondrocytes. These cells have since been used for BMT and the repair/regeneration of numerous tissues, including muscle, heart, and bone, as supported by thousands of publications that describe numerous MSC-based therapeutic interventions.
Dr. Prockop then focused on the application of these cells to the treatment of osteogenesis imperfecta (OI), a disorder characterized by brittle bones. Because collagen is a major source of bone strength, it was hypothesized that mutations in human collagen genes could be responsible for the condition. Indeed, most OI patients were found to have mutations in several genes for type I collagen; in the most severe cases, a single glycine residue had been replaced with a larger residue that disrupted the protein’s triple-helical structure. Early treatments for OI (Horwitz EM, et.al. Nat Med 1999;5:309-313; Horwitz EM, et.al. Blood 2001;97:1227-1231) featured initial BMT from a normal sibling, followed four years later by the transplantation of a large number of MSCs from the same normal sibling. Results were encouraging, but the level of engraftment remained at one percent or less. An alternate approach is to take a small BM sample from an OI patient, grow the cells in a Good Manufacturing Practices facility, and create a large number of cells to replace functionality.
However, this approach initially failed. Adult stem/progenitor cells exist to repair tissue; although similar to ES cells in many respects, they are not identical. These cells often repair tissue through stimulatory responses and modulation of immune responses. This observation led to an experiment in which human MSCs were injected into the mouse hippocampus, a region rich in endogenous neural cells. The expected result--proliferation of human MSCs--was not observed; rather, endogenous neural stem cells proliferated (Munoz JR, et.al. PNAS 2005;102:18171-18176). While the proliferation of human cells was limited, their implantation promoted the proliferation, migration, and survival of the endogenous neural stem cells.
A second experiment was carried out by injecting human MSCs into diabetic mice that were treated with multiple low doses of streptozotocin to produce severe hypoglycemia. Human MSCs were injected into the left cardiac ventricle to avoid entrapment of the cells in the lung (Lee RH, et.al. PNAS 2006;103:17438-17443). The major effect observed in the pancreas was an increase in mouse islets that produced mouse insulin; a few human MSCs also produced human insulin. In the kidney, engraftment of cells and improved histology of glomeruli were observed. Additional islets appeared to bud off from pancreatic ducts, and only a few human cells were found. Possible explanations for these observations include stimulation of endogenous stem/progenitors or a decrease of inflammatory/immune reactions.
A third experiment injected human MSCs into the dentate gyrus of mice following transient common carotid occlusion, an injury that produces ischemia and reperfusion. Human cells disappeared within 4-7 days, but there were no significant differences between immunocompetent and immunodeficient mice. Decreases in neuron death and neurological deficits were observed. The presence of human cells upregulated several neuroprotective genes and downregulated nearly 80 non-neuroprotective genes. Microarray data suggested that the human MSCs reduced the inflammatory and immune responses to ischemia.
The anti-immune effects of MSCs in animal models and in patients with GVHD have been postulated to occur as follows (Ren G, et.al. Cell Stem Cell 2008;2:141-150). MSCs are activated by interferon-gamma and several other cytokines. MSCs then express cytokines that attract T-cells and express inducible nitric oxide synthase and nitric oxide. The nitric oxide then suppresses the T-cells.
Dr. Prockop summarized the state of the field by noting that the mechanisms through which MSCs and other cells repair tissues are not fully understood. Multiple clinical trials with MSCs and other bone marrow cells are underway to assess these cells as treatment tools for GVHD, spinal cord injury, stroke, and heart disease. Although some of these trials show encouraging results, it must be noted that many lack acceptable controls with regard to the patients’ conditions, the routes and doses of administration, and the nature of the cells employed. In conclusion, Dr. Prockop expressed great hope for these cells as agents in therapeutic interventions. However, more data are needed to understand their biology and to assess the risk/benefit of using the cells in different settings.
One attendee asked if beta-cells were regenerated by human MSCs in these experiments. Dr. Prockop stated that this was not clear; it is hypothesized that the human cells may be producing modulatory cytokines or chemokines. Another participant asked about the effects that xenografting may have on outcomes, and Dr. Prockop noted that human tissues were used in these experiments for their stability.