Hot Topics

Iron particles for Noninvasive Monitoring of Bone Marrow Mesenchymal Stem Cell In Infarcted Heart

(Stuckey DJ et al, Stem Cells 2006;24:1968)
Summarized by Erica Salerno and Chris Komurek

LAY SUMMARY

Bone marrow stromal cells (BMSCs), also known as mesenchymal stem cells, have shown the potential to differentiate into bone, fat, cartilage and even cardiac tissue.  With half of the patients diagnosed with heart failure dying within 5 years and billions of dollars spent for their healthcare, it is no wonder that much research has turned towards these BMSCs as the hope for solving this problem.  While clinical trials with BMSCs appear to be improving the heart’s ability to function, we have yet to identify the movement of the implanted cells, without invasive techniques. Stuckey, et al., attempted to use iron particles and MRI to enhance tracking of implanted stem cells over several months in hopes of optimizing methods to ensure better regeneration and safety to the recipient.

The group organized experiments up to a 20 week frame wherein they checked the cells’ location in vivo, in the live rat, with MRI.  Further tests were done after sacrificing the injected rats to confirm that what they saw was really their donated cells causing signals and not other cells. 

The group first obtained cells from the rats’ bone marrow and grew them, keeping only the cells that stuck to the plate, the mesenchymal stem cells (Figures A1-2).  These cells were analyzed according to their surface markers through flow cytometry and were exposed to a modified HIV-1 virus that transferred genetic information that coded for green fluorescent proteins (GFP), allowing the cells to be easily identified under the microscope.  These same cells were also incubated with many iron particles that were taken up by the BMSCs for MRI detection (Figures A3-5).  The heart was subjected to injury and the cells were injected near the injured site. After this, the mice were subjected to imaging by MRI (Figure B1).

As expected, the group detected the cells at the sites of injection with MRI.  This is significant because, not only can they now detect the cells, they can determine physiological changes that may occur in the damaged heart after injection, particularly the ejection fraction which indicates what portion of blood in a filled left ventricle is being pumped out.  This is a useful indicator of whether or not the failing heart is improving or not.  After detection, the hearts were removed and re-analyzed with MRI, as well as, observed for GFP-expression, to determine if the signals were indeed coming from the same cells (figures B2, 4).  The group even digested the heart with enzymes and, with a magnet, collected all the iron labeled cells to check for the GFP and see what the cells looked like (Figure B3, 5).  Not only did they find that the GFP and MRI signals were coming from the same regions but that the cells collected with the magnet were still very similar to what the cells looked like before the injection. 

Although the group did not find any significant improvement in the hearts, they did find an interesting correlation.  That is, the greater the infarction, the greater the signals recorded, implying that a greater number of the BMSCs with iron remained in these damaged areas than in less damaged hearts.

SCIENTIFIC SUMMARY

Much interest surrounds stem cell therapy for infarcted tissue of the heart.  However, hindrances such as the tracking of donor cells, or autologous stem cells, and the confirmation of engraftment, without the use of radioactively-labeled cells, must be overcome. This study monitored the homing and engraftment of donor BMSCs into the infarcted hearts of rats, noninvasively, using iron-labeled BMSCs, with hopes of improving cardiac perfusion and ejection fractions.   

BMSC isolation, culturing, characterization, and labeling
In this study, BMSCs were isolated from the tibia and femur of 8-week old male rats and plated in media for 24 hours.  The adherent population was cultured and used in experiments at passage 2. The BMSCs were characterized using flow cytometry and found to be CD4-/CD11-/CD31-/CD45R-/CD49-/CD90(Thy1)+ (Figures A1-3).  To monitor cells in vivo, BMSCs were transduced with a GFP lentiviral vector (LV-GFP) and incubated with iron particles (polystyrene beads containing magnetite core) (figure A4).  Phase contrast and fluorescent microscopy was performed to confirm labeling prior to injection.  To determine the effect iron-labeled BMSCs (Fe-BMSCs) had on MR images, BMSC suspensions (105 cells), containing either 0%, 1%, or 100% Fe-BMSCs were imaged via MRI, measuring the size of the signal void.  The percentage of Fe-BMSCs was reduced to identify the limit of detection, and it was found that signal voids from as few as 1,000 Fe-BMSCs (1%) can be detected (Figure A5).  

Infarction and BMSC injections
Myocardial infarctions were induced in female Wister rats via occlusion of the left anterior descending coronary artery.  Four injections of 1.25x105 iron- and GFP-labeled BMSCs were made into the myocardium of the infarcted tissue 10 minutes later.  Additional rats were either injected with nonlabeled BMSCs or unbound iron particles into infarcted hearts, or with Fe-BMSCs into noninfarcted hearts (Figure B).  One week later, MR images showed that the injections were successful, in that 2 or 3 of the injection sites could be located. The rat hearts were excised and GFP+ cells were detected at the sites where the MRI detected signal voids. Cells were also stained for CD68, a macrophage marker, showing non-iron-labeled macrophages and CD68(-) iron-labeled cells (Figures B2,5).

In vivo and ex vivo monitoring
Additional rats were infarcted, injected with GFP-Fe-BMSCs, and monitored in vivo via MRI at 1, 4, 10, and 16 weeks postinjection. Cine-MRI images of the entire left ventricle were obtained, while calculating ejection fraction for 40 minutes (Figure B1).  Signal voids were detected in the left ventricular wall, at the site of infarction. The size of the signal void decreased over time, indicating donor cells being lost from the heart, with the greatest reduction being between 1 and 4 weeks, and there was no significant improvement in ejection fraction.  The signal void from the donor Fe-BMSCs was found to be inversely correlated to the ejection fraction at 10 and 16 weeks, in that a greater number of donor cells were retained in the more severely infarcted rat hearts. 

Sixteen weeks postinfarction, hearts were excised and fixed to determine the degree of engraftment of the donor BMSCs into the rat infarcted heart.  Using high-resolution three-dimensional MR microscopy, signal voids were detected at the same position as in vivo.  No voids were detected in the hearts that received non-labeled BMSCs.  In noninfarcted hearts that were injected with Fe-BMSCs and in infarcted hearts injected with iron particles alone, signal voids were detected at 1 day and 1 week, but disappeared by 4 weeks, indicating non-engraftment (Figure B4). 

To further confirm that the BMSCs had actually survived and engrafted into the myocardium, immunohistochemical analysis was performed on heart sections 16 weeks after injection to detect GFP-expressing BMSCs (Figure B5). Iron particles and GFP were found to colocalize within cells.  Sections were also stained for CD68, and it was found that the number of macrophages decreased over 16 weeks.  Those present in week 1 did not contain iron and most of the iron-containing cells were CD68-.
           
Enzymatic isolation
To confirm long-term retention of iron particles in the GFP-BMSCs, at 6 and 20 weeks postinjection, enzymatic isolation of iron-containing cells was performed.  Rapidly excised hearts were attached to a Langendorff apparatus, as various salts, low-calcium solution, collagenase, and hyaluronidase perfused through.  Myocytes were pelleted, exposed to a magnet, and resuspended in medium for fluorescent photography (Figure B3).  GFP-Fe-BMSCs were located at regions of signal void detection and were also colocalized when stained.  The morphology of the grafted donor cells isolated from heart digests remained similar to that of the cells prior to injection.

Authors’ Conclusions
Noninvasive tracking of stem cell homing and engraftment is an ideal method used in treatments such as regeneration of heart tissue, with MRI tracking of iron-labeled cells currently being the most ideal.  This is the first time in which internally labeled donor cells were isolated from whole heart digests using their magnetic properties.  Subpopulations of BMSCs have the ability to differentiate into cardiomyocytes; however, no improvement in ejection fraction was seen over the 16-week period, perhaps because the degree of infarction in these rats was relatively low.

Personal Comments:
Stuckey, et. al., did prove that iron particles can be used to track cells in vivo via MRI and can also be used to re-isolate the transplanted cells for further experimentation, however, his research generated many concerns.  First, it was never proven whether the cultured BMSCs were mesenchymal stem cells or fibroblasts, although they phagocytized the iron particles.  Second, although rejection was not seen in this particular experiment, iron particles and GFP may result in immunoreactivity.  Third, it was not proven whether the cells differentiated or proliferated.  It seemed as if the cells’ morphology remained the same as prior to injection, indicating that the microenvironment had no influence.  Perhaps, after excision, flow cytometry should have been performed again to detect a change in surface marker expression.  Also, the cells could have been labeled with CFSE to detect proliferation.  Fourth, the transplanted BMSCs did not seem to improve cardiac function or ejection fraction.  Many cells were ejected in the less severe infarcted hearts.  Stuckey should have tracked the location of these cells to see if they settled in another organ, such as the lungs.  Fifth, many people with heart conditions may have an implanted device, such as a pacemaker, in which they would be unable to undergo MR imaging.