The Impossible Heart

In 2004 she’d been in Minnesota for three years. She was building a team, raising funds, experimenting with cells, and following up on her work at Duke to strengthen muscle tissue after heart attacks when one day she got this crazy idea. (She would prefer the sentence to read “her team got this crazy idea.” She may be team leader, but she is, most emphatically, a team player, who says she owes a particular debt to Harald Ott, a young Austrian PhD student. “Harald drove the heart project,” she says of Ott, who left the lab before the story broke to do his surgery residency at Massachusetts General Hospital in Boston.)

As Taylor remembers the scenario, she and Ott were brainstorming as usual one day, mulling over how to take stem cell research “from bench to bedside,” she says, when one of them, possibly Taylor but maybe Ott, said, What if we give the cells something to grow on that they might recognize as their own heart even though it isn’t—some sort of organic structure that the recipient would be less likely to reject than, say, a plastic or metal scaffold? And the other one said, Why not?

At the time, Taylor had been thinking about pig valves, which are routinely used in humans for replacement valves—their cells removed in order for transplantation to succeed. The idea of growing new cells on a whole-organ scaffold emerged from that thought process. Taylor sought, in effect, to skip over the tissue-engineering research that was progressing at a snail’s pace and build a heart from scratch. So they started riffing on this superbly crazy idea—Taylor in her drawl and Ott with his clipped accent—marveling at the sheer audacity of it. If the lab could strip a dead heart of its cells, reseed it with new cells, attach a pacemaker, and get it to beat—well, they’d have something amazing.

“Harald tried all sorts of things [to remove the old cells] and one day he tried something like ordinary shampoo,” Taylor says, “and, lo and behold, it worked!”

In the Newsmaking article she later published in Nature Medicine, Taylor describes the decellularized hearts this way: “The fiber composition (waves, struts, and coils) and orientation of the myocardial/ECM [extracellular matrix] were preserved, whereas cardiac cells were removed in compressed constructs. Within the retained ventricular ECM we saw intact vascular basal membranes without endothelia or smooth muscle cells.” In plain language, the “new” heart retained the architecture required to pump blood. It appeared to be perfusable. Even the aortic valve was in good working order.

“So the next thing we did was inject cells into the wall of the heart,” she says, while seated at a round table in her office, talking to one of the many journalists she’s explained this to since the story became public in January. “Cardiac muscle cells, smooth muscle cells, a mixture of endothelial cells, fibroblasts, heart stem cells, and everything that makes up a heart. That those cells living in the wall of the heart would connect to each other and would move to the blood vessel areas and do the right thing—we had no idea would happen. This was all new information for us. Every piece of it was new, and again every piece of it is where it could’ve gone wrong . . . . We did everything we could to bias it in our favor. We took cells we thought would work, cells we could isolate in the lab and that were resistant to ischemia [the effects of too little blood flow to the heart], you know, to increase the odds. But, still, there were plenty of ways we could’ve messed it up.”

The reseeded scaffold was placed in a bioreactor that simulates the cardiac environment. The team attached a Medtronic pacemaker that had come to Minnesota from Duke with Taylor. It took only four days for a semblance of a heart contraction to appear, and another four days for the hearts to pump hard enough to equal the function of a neonatal heart at sixteen weeks (about 2 percent of adult heart function). The breakthrough was dramatic.

One night, Ott and lab tech Thomas Matthiesen were babysitting the heart in the lab. It was hanging like a piece of meat in a butcher’s shop window—a very small piece of meat, smaller than a golf ball. The heart was inside a glass jar connected to more glass jars, tubes, cylinders, and the electrical leads that would be needed to jump-start the heart from lifeless protein to pulsating organ. The heart had started out looking like a hunk of gristle, but in recent days had gone from ghostly white to flesh-colored to pink to pink shot through with streaks of crimson.

The deepening in color had been greeted with slaps on the back and high fives. It showed that, in Taylor’s words, “both the larger cardiac vessels and the smaller third- and fourth-level branches were patent.” The scientists figured they might as well enjoy the progress while it lasted because they were sure it wouldn’t—couldn’t possibly—last much longer. But tonight was different. The red dollop of cell-infused collagen was no longer inert but moving—expanding and contracting. In, out. In, out.

Ott couldn’t believe his eyes. Sure, they’d hoped for this, but, like contestants on a game show, they knew that with every step forward the stakes grew higher, the statistical odds stacking more formidably against them. Ott grabbed a video camera and e-mailed his “movies” to Taylor at her condo so she wouldn’t accuse him of pulling her leg when he told her why he was calling at 3 a.m.

The hearts were next transplanted into live rat abdomens to test their viability (they weren’t strong enough yet to keep the animal alive). “The recellularized construct was contracting and drug-responsive after eight days of culture,” Taylor wrote later. Given time to mature, the heart could “be transplanted either in part (for example, as a ventricle for congenital heart disease . . . ) or as an entire donor heart in end-stage heart failure . . . . This approach holds promise for virtually any solid organ.”

The team hung more rat hearts. And soon they were decellularizing a pig’s heart, which is as close as an animal heart gets to a human’s. Someone was assigned the task of coming up with a pancreas scaffold. Livers, kidneys, valves, and blood vessels were added to the inventory for future reseeding with stem cells.

Using a human cadaver heart may be the likeliest long-term scenario if this technique becomes common, but Taylor doesn’t rule out building a human heart on a pig heart scaffold. Either way, she says, the challenge is one of scale, not cell biology. And scale, according to tissue engineers who have been trying to build organs for years, is a whole lot easier to solve. Taylor’s core ambition is to show that stem cells that are moved from one organ to another, even from a heart to a kidney, will adapt to the new environment and change their characteristics to support the new organ. For human subjects, the stem cells will likely come from the recipient’s bone marrow or heart tissue.