Heart Muscle Cells Produced from Induced Pluripotent Stem Cells Repair Heart Attacks in Pigs

When heart muscle cells are made from embryonic stem cells, they integrate into the heart and form proper connections with other heart muscle cells. Such experiments have been conducted in mice, guinea pigs, and nonhuman primates (i.e. monkeys). Chong and others earlier this year (Nature (2014) 510, 273-277) implanted heart muscle cells produced from embryonic stem cells into the hearts of nonhuman primates that had suffered from heart attacks. There was extensive evidence of engraftment of these cells, remuscularization of the heart, and electrical synchronization 2 to 7 weeks after transplantation. However, despite these successes, the hearts of some of these animals also showed abnormal heart beat patterns (known as arrhythmias). Such a problem has also been observed in other laboratory animals as well (see my book The Stem Cell Epistles), and this problem has to be addressed before derivatives of pluripotent stem cells can be used to treat damaged hearts (pluripotent means capable of differentiating into all the mature adult cell types).

Jianyi Zhang and his colleagues at the University of Minnesota have used induced pluripotent stem cells made from human skin cells to produce heart muscle cells that were used to treat pigs that had suffered from induced heart attacks.  Their results differed slightly from those of Chong and others.

Zhang and others noted that implanted heart muscle cells typically survive better if they are implanted with blood vessel cells (endothelial cells or ECs).  This was first shown in culture by Xiong and others in 2012 (Circulation Research 111, 455-468), but other work has confirmed this.  That is, Zhang’s coworkers in his laboratory co-transplanted heart muscle cells made from induced pluripotent stem cells with endothelial cells and smooth muscle cells (which are also a part of blood vessels), and saw that the co-transplanted cells survived much better than heart muscle cells that were transplanted without these other cell types.

On the basis of these experiments, Zhang and his crew decided that implanted heart muscle cells would do much better if they were implanted into pig hearts if they were implanted with endothelial and smooth muscle cells.  This was the hypothesis that Zhang and others wanted to test in this paper (which was published in Cell Stem Cell, Dec 4, 2014, 750-761).

Skin biopsies from human volunteers were used as a source of skin cells that were then genetically engineered and then cultured to form human induced pluripotent stem cells (hiPSCs).  These cultured hiPSCs were differentiated into heart muscle cells by means of the “Sandwich method,” which yielded beating heart muscle cells in about 30 days.  Additionally, their hiPSC lines were differentiated into smooth muscle and endothelial cells as well.

Next, Zhang and his colleagues and collaborators used 92 pigs and subjected them to experimentally-induced heart attacks.  Why pigs?  Pigs are a larger animal than rodents, and their hearts are larger and beat much slower than the hearts of rats and mice.  Therefore, they are a more expensive, but better experimental model system for the human heart.  Nevertheless, these pigs were divided into six different groups (3 pigs died from the procedure, so there were 89 pigs involved in this experiment).  Animals in the first group or SHAM group underwent the surgery to induce a heart attack, but no heart attack was induced.  The second group was called the MI group and this group received no other interventions after surgery.  The Patch group received a fibrin patch over the site of injury, but no cells.  The CM + EC + SMC group received injections of 2 million heart muscle cells, two million endothelial cells, and two million smooth muscle cells directly into the injured portion of the heart.  The Cell + Patch group received all three cell types in a fibrin patched that was imbued with a growth factor called Insulin-like growth Factor-1 (IGF-1) that had been loaded into microspheres.  This causes the growth factor to be released gradually and exert its effects over a much greater period of time.

That’s a lot of information so let’s review – six groups: 1) SHAM (no heart attack; 2) MI (heart attack and no treatment); 3) Patch (just the fibrin patch); 4) Cells + Patch (fibrin patch with the three cell types); 5) Cells (cells, but no patch), and a final group cells Patch + CM (just heart muscle cells in the patch).

Animals were evaluated one week after the heart attack and four weeks after their heart attacks. I am uncertain how soon after the heart attack the treatments were given, but in the paper it reads to me as though the treatments were given right after the heart attacks had been induced.  Because all implanted cells were engineered to glow in the dark, the number of surviving cells could be counted and tracked.

Only 4.2% of the cell survived in the Cells group, up to 9% of the cells in the Cell + Patch group survived.  32% of the cells in the CM + Patch group survived.  Thus, it seemed as though the presence of the other cell types did increase the survival of the heart muscle cells and the patch also increased cell survival rates.  Secondly, the heart function of all the treated groups was better than the MI group, but the hearts treated with Cells + Patch were clearly superior to all the others, with the exception of the SHAM group.  The hiPSC-derived heart muscle cells also clearly engrafted into the hearts of the pigs, but the big surprise in this paper is that THERE WERE NO INDICATIONS OF ARRHYTHMIAS!!!  Apparently the manner in which these hiPSC-derived heart muscle cells integrated and adapted to the native heart in such as way as to preclude irregular electrical activity.  Another indicator measured was ratio of phosphocreatine to ATP.  If that sounds like a language from outer space, it simply means a measurement of the efficiency of muscle mitochondria (the part of the cell that makes all the energy).  Again the Cells + Patch hearts had significantly more efficient mitochondria, and, hence, better energy production than the other hearts.  Damage to mitochondria also tends cause cells to up and die, which means that these cells were in better health that those from the MI group.

This paper shows that an ingenious tissue engineering innovation that uses a fibrin patch and a a combination of cells, not just heart muscle cells can significantly increase the healing after a heart attack.  Also, even though neither embryonic stem cell-derived cells nor iPSC-derived cells are ready for clinical trials, this paper shows that iPSCs are not as far behind iPSCs as some authors have suggested.  Furthermore, because iPSCs would not be subject to immunological rejection, they have an inherent superiority over embryonic stem cells.  The problem comes with the time required to make iPSCs and then derived heart muscle cells from them, which might put it outside the time window for treat of an acute heart attack.

Heart Cells Expressing Stem Cell Factor Show Less Cell Death After a Heart Attack

Stem Cell Factor is a cell surface protein that is expressed by several different cells, including tissue fibroblasts, heart cells, cells in the bone marrow, and blood vessel cells. Stem Cell Factor (SCF) plays important roles in the migration, proliferation, and adhesion of any cell that expresses the receptor for SCF, a molecule called c-kit. Cells that express c-kit include cardiac stem cells, endothelial progenitor cells, and hematopoietic stem cells. When c-kit binds to SCF, the SCF-containing cell activate their Akt /PI3K pathway, and this pathway prevents cells from dying and drives them to divide, differentiate, more, adhere, and even secrete new molecules.


Fu-Li Xang in the laboratory of Qingping Feng at the University of Western Ontario has done several experiments with SCF in the heart. His goal is to determine if heart cells that have SCF fare better after a heart attack than hearts that do not have quite so much SCF.

To that end, Feng and his team showed that SCF does help heal the heart after a heart attack in 2009 (Xiang et al, Circulation 120: 1065-74). The next step was to determine if SCF could attenuate cell death in the heart that results from a heart attack.



The strategy behind this experiments involved making genetically engineered mice that expressed lots of SCF in their heart muscle. The particular mouse strain that Feng and his crew made had the SCF gene activated by a heart muscle-specific promoter, but the expression of SCF could be shut off by giving the mice the drug doxycycline. These SCF transgenic mice and normal mice were given heart attacks and then some were treated with a doxycycline while others were given a drug called LY294002, which inhibits the Akt pathway. These animals were then analyzed three hours after the induced heart attack and the amount of cell death, the size of the infact, the number of stem cells that moved into the heart were all measured.


The upshot of all this work is this: SCF decreased the amount of cell death by about 40%. Also the size of the infarct was also smaller. These benefits were abrogated by the co-administration of either doxycycline or LY294002. When a search for molecules that are indicative of cell death were examined, the results were completely unsurprising: the markers of cells death like fragmented DNA or caspase-3 were decreased in the SCF mice and this attenuation was abrogated by co-administration with doxycycline or LY294002.

Other experiments examined the activation of the Akt/PI3K pathway in the SCF-expressing animals, and it was quite clear that the SCF-expressing animals showed a robustly active Akt/PI3K pathway compared to the non-SCF-expressing mice.

A different experiment examined the presence of c-kit-expressing cells in the hearts of these mice. Remember that c-kit expressing cells are stem cells that have been recruited to the heart by the SCF. Once again, it was exceedingly clear that the SCF-expressing mice had hearts with a large excess of c-kit-expressing cells and this recruitment of stem cells was abrogated by neutralizing c-kit with an antibody against it. The incoming stem cells also tend to secrete a host of interesting molecules that help heal the heart, and one of these molecules, HGF (hepatic growth factor), which also goes up in concentration in the hearts of the SCF-expressing mice, is blocked by a drug called crizotinib. If SCF-expressing mice were pre-treated with crizotinib, the infarct size tended to be just as large as the non-SCF-expressing cells.

Feng and his group also examined the resident stem cells in the heart, the cardiac stem cells population, which, by the way, also express c-kit. These cells also were induced to express HGF and IGF (insulin-like growth factor) as a result of SCF, and if the c-kit receptor was blocked with an antibody, then this effect was abrogated.

There is a lot of data in this paper, but the news is almost all good. Basically SCF will recruit stem cells to the heart after a heart attack and this recruitment happens quickly (within 3 hours) and does the heart a world of good. Translating this work into human patients will not be easy, but SCF is available. If it could be localized to the heart by some means soon after a heart attack, there is good reason to believe, based on these pre-clinical results that it would do the patient quite a bit of good. The next piece is figuring our how to go about doing just that.

Genomic Imprinting Maintains A Reserve Pool of Blood-Forming Stem Cells

Hematopoietic stem cells or HSCs reside in the bone marrow and give rise to the wide variety of specialized blood cells that inhabit our bloodstreams. Within the bone marrow, HSCs come in two varieties: an active arm of HSCs that proliferate continually to replace our blood cells and a reserve arm that sits and quietly waits for their time to come.

New research from the Stowers Institute at Kansas City, Mo, in particular a research team led by Linheng Li, discovered a mechanism that helps maintain the balance between those HSCs kept in reserve and those on active duty.

According to Dr. Li, genomic imprinting, a process that specifically shuts off one of the two gene copies found in each mammalian cell , prevents the HSCs held in reserve from being switched to active duty prematurely.

Li explained: “Active HSCs form the daily supply line that continually replenishes worn-out blood and immune cells while the reserve pool serves as a backup system that replaces damaged active HSCs and steps in during times of increased need. In order to maintain a long-term strategic reserve of hematopoietic stem cells that lasts a lifetime it is very important to ensure that the back-up crew isn’t mobilized all at once. Genomic imprinting provides an additional layer of regulation that does just that.”

Sexual reproduction produces progeny that have once set of chromosomes from the mother and one set of chromosomes from the father. The vast majority of genes are expressed from both sets of chromosomes. However, in placental mammals and marsupial mammals a small subset of genes are imprinted, which means that they receive a mark during the development of eggs and sperm and these marks shut down expression of those genes in either the sperm pronucleus or the egg pronucleus. Therefore, after the fusion of the sperm and the egg and the eventual fusion of the egg and sperm pronuclei, these imprinted genes are only expressed from one copy of genes. Some are only expressed from the paternal chromosomes and others are only expressed from the maternal chromosome. Imprinting is essential for normal development in mammals.

The importance of genetic imprinting is shown if an egg loses its pronucleus and is then fertilized by two sperms. The resulting zygote has two copies of paternal chromosomes and no copies of the maternal chromosomes. Such an embryo is called an andogenote, and the embryo fails to form but the placenta overgrows. If this occurs during human development, it can lead to a so-called “molar pregnancy” or “hydatiform mole.” This fast growing placental tissue can become cancerous and lead to uterine cancer. For that reason, molar pregnancies are usually dealt with expeditiously.

However, if the sperm that fertilizes the egg is devoid of a pronucleus, and the egg pronucleus duplicates, then the resulting zygotes can two copies of the maternal chromosomes, and this entity is known as a gynogenote, and it develops with a poorly formed placenta that dies early in development.

In previous experiments in mice, Li and his colleagues indicated that the expression of several imprinted genes changes as HSCs transition from quiescent reserve cells to multi-lineage progenitor cells.

In their current study, Li and other Stowers Institute researchers examined a differentially imprinted control region, which drives the reciprocal expression of a gene called H19 from the maternal chromosome and IGF2 (insulin-like growth factor-2) from the paternal chromosome.

The first author of this study, Aparna Venkatraman developed a mouse model that allowed her to specifically delete the imprinted copy from the maternal chromosome. Thus, in these mice, H19, which restricts growth, was no longer active and Igf2,, which promotes cell division, was now active from the paternal and the maternal chromosome. To access the effect of this loss of imprinting on the maintenance of HSCs, Venkatraman examined the numbers of quiescent HSCs and active HSCs. in mouse bone marrow.

Venkatraman explained: “A large number of quiescent HSCs was activated simultaneously when the epigenetic control provided by genomic imprinting was removed. It created a wave of activated stem cells that moved through different maturation stages.”

She followed this experiment with a closer look at the Igf2 gene. Misregulation of Igf2 leads to overgrowth syndromes such as Beckwith-Wiedmann Syndrome. It exerts its growth promoting effects through the Igf1 receptor, which induces an intracellular signaling cascade that stimulates cell proliferation.

IGF signaling pathway
IGF signaling pathway

The expression of the Igf1 receptor itself is regulated by H19, which encodes a regulatory microRNA (miR-675) that represses translation of the Igf1 receptor gene and therefore prevents production of Igf1 receptor protein. Venkatraman explained that once the “imprinting block is lifted, the Igf2-Igf1r signaling pathway is activated.” Venkatraman continued: “The resulting growth signal triggers the inappropriate activation and proliferation of quiescent HSCs, which eventually leads to the premature exhaustion of the reserve [HSC] pool.”

Interestingly, the roundworm, Caenorhabditis elegans, provided the first clues that diminished insulin/IGF signaling can increase lifespan and delay aging. Li again: “Here the IGF pathway is conserved by subject to imprinting, which inhibits its activation in quiescent reserve stem cells. This ensures the long-term maintenance of the blood system, which in turn supports the longevity of the host.”