A “SMARTer” Way to Isolate Mesenchymal Stem Cells


The Singapore-MIT Alliance for Research and Technology or SMART employs a team of engineers and life scientists to design technologies that address problems in science and medicine. In particular, a SMART team has devised a new technique to identify mesenchymal stem cells from bone marrow cells on the basis of cell size, cell stiffness, and the deformation of the nucleus.

Mesenchymal stem cells (MSCs) constitute less than one percent of the total cells in bone marrow. Therefore, isolating these cells from the morass of cells that are in the bone marrow is somewhat of a challenge. Most of the procedures for isolating MSCs from bone marrow utilize cell surface proteins found on the surfaces of MSCs, but there are no few cell surface proteins that are unique to only MSCs. Therefore, such isolation procedures tend to be tedious and not completely efficient. Because MSCs can differentiate into cells that produce bone, cartilage, fat, or muscle, they have proven invaluable for tissue repair therapies.

This new study by the SMART team has identified three physical characteristics of MSCs that can distinguish them from other immature cells found in the bone marrow. These physical characteristics should help them invent devices that could rapidly isolate MSCs, and facilitate the isolation of sufficient numbers of stem cells to treat patients.

Presently there are no sure-fire ways to quickly and efficiently separate MSCs from bone marrow cells that have already begun to differentiate into other cell types, but share the same molecules on the cell surface. This caveat may explain why experimental results vary among labs, and why stem-cell treatments now in clinical trials are not as effective as they could be, said Krystyn Van Vliet, an MIT associate professor of materials science and engineering and biological engineering and a senior author of the paper, that appeared in the Proceedings of the National Academy of Sciences.

“Some of the cells that you’re putting in and calling stem cells are producing a beneficial therapeutic outcome, but many of the cells that you’re putting in are not,” Van Vliet said. “Our approach provides a way to purify or highly enrich for the stem cells in that population. You can now find the needles in the haystack and use them for human therapy.”

In bone marrow, MSCs exist alongside other immature cells, such as osteogenic cells, which have already begun the developmental path toward becoming cartilage- or bone-producing cells. Currently, researchers try to isolate MSCs based on protein markers found on the cell surfaces, but these markers are not specific to MSCs. Therefore isolation techniques that rely on cell surface proteins can also co-isolate other types of immature cells that are more differentiated.

“Conventional cell-surface markers are frequently used to isolate different types of stem cells from the human bone marrow, but they lack sufficient ‘resolution’ to distinguish between subpopulations of mesenchymal stromal cells with distinct functions,” Lee said.

The researchers set out to find biophysical markers for multipotency (the ability to differentiate into several different cell types). They hypothesized that cell size might be a factor, since fetal bone marrow stem cells, which tend to have a higher percentage of MSCs, are usually small in diameter.

Jongyoon Han, an MIT professor of electrical engineering and biological engineering, had previously invented a device that captures circulating tumor cells based on their size. The SMART team used Han’s machine to isolate bone marrow cells based on size and discovered that none of the larger cells were multipotent, but not all of the smaller cells were multipotent. Therefore, size alone in insufficient to distinguish MSCs.

After measuring several other physical traits, the SMART team observed that two other physical characteristics could be combined with cell size to completely distinguish MSCs from other stem cells: stiffness of the cell, and the degree of fluctuation in the cell’s nuclear membrane.

“You don’t need more than these three, but you also can’t use fewer than these three,” Van Vliet said. “We now have a triplet of characteristics that identifies populations of cells that are going to be multipotent versus populations of cells that are only going to be able to become bone or cartilage cells.”

These features seem to correspond to what is already known about stem cells, Van Vliet said. In contrast to cells that have already committed to their final fate, immature cells have genetic material that moves around inside the nucleus, producing more fluctuations of the nuclear cell membrane. Stem cells also have a less rigid internal cytoskeletal structure than those of highly differentiated cells, which makes them seem less stiff.

The researchers then tested the regenerative abilities of MSCs isolated on the basis of these three characteristics in mice. They found that immature MSCs could help repair both muscle and bone injuries, but cells identified as osteogenic stromal cells were able to repair bone but not muscle.

“We have provided the first demonstration that subpopulations of mesenchymal stromal cells can be identified and highly enriched for bone growth and muscle repair,” Lee said. “We envision that this approach would also be important in the selection and purification of bone marrow-derived stem cells for tissue repair in human patients suffering from a range of tissue-degenerative diseases.”

“This is potentially a big step forward in establishing a marker-free way of identifying mesenchymal stem cells with maximum differentiation capacity,” said Jochen Guck, a professor of cellular machines at the Dresden University of Technology. “Biophysical markers have long been discussed and sought as an alternative to antibody labeling. What sets this work apart from others is that it clearly said that no single marker (at least of those tried) alone is predictive enough, but that a combination of them is required.”

The SMART team is now working on high-speed methods for separating MSCs. Creating more pure populations of such cells should lead to more effective stem-cell treatments for tissue injuries, Van Vliet said.

“Instead of putting in 30 percent of the cells that you want, and 70 percent filler, you’re putting in 100 percent of the cells that you want,” she explains. “That should lead to more reliable patient outcomes, because you’re not going to have this variability from batch to batch, or patient to patient, in how many of each cell population are present.”

Van Vliet and Poon also hope to initiate a clinical trial that utilizes the osteogenic cells isolated in this study, which could potentially prove useful for treating bone injuries.

Advertisements