Magnetic Resonance Labeling of Stem Cells

Abstract

Recent meta-analyses of clinical cellular therapy trials in cardiovascular disease have shown that these therapies are safe and perhaps yield a positive therapeutic benefit (1-5). However, 1 issue that has plagued these clinical studies is the inability to determine the percentage engraftment of exogenously administered stem cells and the stem cell fate. In this issue of iJACC, Adler et al. (6) propose a cardiac magnetic resonance (CMR) contrast based on a paramagnetic agent, gadolinium, to track the engraftment of embryonic stem cell-derived cardiovascular progenitor cells.

Because of the high spatial resolution and soft tissue detail, CMR is ideally suited for imaging the myocardium. Thus, when techniques for direct labeling of stem cells with iron oxides for tracking stem cell engraftment using CMR were developed in the 1990s (7,8), they were rapidly adopted for cardiac use. Later methods using Food and Drug Administration-approved superparamagnetic iron oxides to label cells were hoped to speed clinical translation. However, adoption rates have been low. In part, this may be due to the negative image contrast provided by iron oxide-labeled cells. Discriminating hypointensities from labeled cells from other causes of signal loss, e.g., calcium, hemorrhage, or metallic implants, may be difficult. Furthermore, the loss of signal from iron-labeled cells obscures the underlying anatomy. Although techniques to convert the negative contrast produced by iron oxides to hyperintensities or positive contrast have been developed, these techniques are often difficult to implement and may lack sensitivity (9-11).

The overwhelming majority of CMR contrast agent usage today is gadolinium-based agents that provide positive image contrast. Because CMR is a signal-starved technique with many CMR artifacts resulting in a signal loss, contrast agents that enhance signal are valued. Although most contrast-enhanced CMR is performed with a pre-contrast scan in order to identify the effect of the contrast agent on the image, imaging of CMR-labeled cells is performed days after cells transplantation. Positive-contrast cell tracking would be beneficial when no reference scan is available.

However, contrast agents are not just black and white. The mechanism of contrast generation differs between iron oxide-based T2 agents and gadolinium-based T1 agents. Gadolinium probes need direct access to water to alter image contrast. In cell labeling, if the contrast agent is restricted to an endosomal compartment, then it will have little effect on most of the cellular water since water exchange in and out of the endosome is slow. On the other hand, contrast agent distributed through the cytosol will have a much stronger effect on T1. It appears that the chemical structure of the contrast agent as well as the means of cell uptake (e.g., pinocytosis vs. electroporation) strongly influence the relaxation properties of the labeled cell (12). A benefit of iron oxide probes is that they shorten T2* via a through-space mechanism; strong signal loss is observed regardless of the cellular location of the contrast agent.

Adler et al. (6) use a bimodal probe containing a carbocyanine dye and a gadolinium chelate to label embryonic stem cell-derived cardiovascular progenitor cells and track these cells in the myocardium. This so-called gadofluorine probe, Gadofluorine M (GdFM) has been previously reported for cell labeling and tracking studies in the brain (13,14). Prior work revealed that the probe is distributed throughout the cytosol and has excellent T1 relaxation properties (13,14). Adler et al. (6) extend this work to the myocardium and demonstrate that positive contrast can be used to track cells in the heart with this probe. In the current study, the authors work at very high field, 9.4-T, which is well suited to cardiac imaging in mice. However, at clinical field strengths, e.g., 1.5- or 3.0-T, the sensitivity should be much higher, as many gadolinium-based agents exhibit strong inverse field-dependent relaxivity. Thus, the conspicuity of the cells should improve if used clinically.

There are a few concerns with the present study. One of the pitfalls of any direct labeling technique is whether the label remains trapped in the intracellular space or is released and can be taken up by bystander cells. Adler et al. (6) have addressed this concern by coincubation of cells expressing green fluorescent protein with GdFM-labeled cells. Very little uptake of GdFM by the green fluorescent protein-expressing cells was seen.

However, another concern is if the cells should die, the contrast agent may either remain in situ or be taken up by bystander cells. Thus, the presence of the label may not represent the originally delivered exogenously labeled cells. Further complicating the issue is that the mechanism of GdFM uptake by cells is not well understood. GdFM has been used in several applications including the identification of metastatic cancer in lymph nodes (15,16) and components of the atherosclerotic plaque (17,18). From these studies, one can glean that GdFM can be taken up by phagocytic cells, such as macrophages, and has an affinity for non-lipid moieties, such as collagen/fibrous tissue. Since a large number of stem cells die shortly after administration, it is likely that the positive contrast provided by GdFM may persist long after the delivered stem cells are no longer there. Finally, with any gadolinium-based agent, there is the potential for the release of free gadolinium as the cells are degraded. Unchelated gadolinium is highly toxic. Further pre-clinical and clinical studies are required to prove the safety and utility of GdFM and this imaging technique. Although GdFM cell labeling yields a positive MR signal, it is likely that the utility of this labeling agent will remain in pre-clinical evaluation of stem cell therapies for the foreseeable future.