A tale of shark bites at a Scottish pub has led us to some new ideas about rebuilding broken bodies. In the early 2000s American geneticist Michael Zasloff of Georgetown University had traveled to the University of St. Andrews to give a talk about several natural antibiotics found in animal skin. After the lecture, he and some of the university scientists went for a drink, and one of them, a marine biologist, began to talk about how dolphins were frequently savaged by sharks, sustaining some bite wounds 45 centimeters long and 12 centimeters deep. But remarkably the dolphins healed up in weeks, with no signs of infection.

Zasloff was struck by this swift recovery from terrible injuries, and he could not get the conversation out of his mind. He spent the next several years reading reports about bitten dolphins and talking to marine biologists who studied these animals. In 2011 he published a letter to the Journal of Investigative Dermatology entitled “Observations on the Remarkable (and Mysterious) Wound-Healing Process of the Bottlenose Dolphin.” He noted that the dolphins did not seem like they were simply patching torn flesh with a scar, which produces different kinds of cells, but instead might be actually regenerating the damaged tissue. And soon after that, he called one of us. Strange, at the time president of the MDI Biological Laboratory, was pushing the institution to investigate natural and synthetic compounds that stimulated regeneration, and Zasloff thought some of the antibiotics he had found in animal skin might also foster this kind of regrowth. Anything that helped the body replace or restore cells destroyed by disease or injury would be a major medical boon.

Six years after that phone call, the three of us (Yin, Strange and Zasloff) have shown that a natural antibiotic called MSI-1436, originally identified by Zasloff in a small shark, dramatically stimulates several types of damaged organs to regrow in zebra fish and prompts heart muscle to regenerate in mice. The compound appears to release some molecular “brakes” holding back a tissue’s natural ability to regenerate after sustaining damage. In mice that have a condition that mimics muscular dystrophy in people, it appears to slow down muscle degeneration. We are still experimenting in animals and have not shown these effects in humans, but MSI-1436 has an important advantage over the legion of drug candidates that look good in test tubes but fail in people: it has already been shown to be safe.

In 2007 this compound was tested in humans as a potential treatment for obesity and type 2 diabetes because it improves cell sensitivity to insulin. The studies, regulated by the U.S. Food and Drug Administration, demonstrated that MSI-1436 was well tolerated at high doses and did not harm patients. But because the drug comes as a liquid that needs to be injected every day, it was unlikely to be popular with patients who already had alternatives, such as pills, that were easier to take. Pharmaceutical companies did not pursue it.

But for regenerating damaged cells, there are currently not a lot of medical options. There have been many headlines about stem cells, unspecialized cells that can, with the right cues, give rise to the myriad highly differentiated cell types that make up the human body. In theory, they could repair damaged parts. Unfortunately, despite many years of clinical trials and other tests, stem cell transplants remain challenged by a lack of efficacy and other serious concerns. The only wide use now is in bone marrow transplants to treat blood cell diseases. But MSI-1436, which has a proved safety record, could become valuable regenerative medicine for repairing the destruction from heart attacks and potentially from other devastating diseases as well.

Credit: Campbell Medical Illustration

Restoration project

Many animals have startling regenerative capabilities. Salamanders regrow entire limbs after amputation. The lamprey, an eel-like fish, can repair a severed spinal cord. Zebra fish, a popular aquarium fish species that is also broadly used in biomedical research, can regenerate damaged hearts, kidneys, pancreases and appendages. Pick almost any tissue or organ, and there is probably an animal that can readily regenerate it.

Even humans are not out of this regrow-and-repair game completely. Our capability appears more limited, but our skin, blood and gut cells regenerate constantly. Muscle can add new cells after some small injuries. And like Prometheus of Greek legend perpetually regenerating his liver, ours, too, can regrow after limited injury. So our cells have these abilities, but they get dialed down and switched off, especially as we grow older. Yet because they exist in the first place, we thought it might be possible to turn them back on with the proper molecular signal. But of course, we first had to find that signal. The fast-healing animal world was the logical place to look.

Zasloff, in his prospecting for antibiotics in animals, had come across a class of molecules called aminosterols—MSI-1436 is one of them—that also had the potential to stimulate regeneration because they could regulate cell activities such as growth. We decided to test their capacities using zebra fish. As vertebrates, the fish have many of the same major organs that people do, and about 70 percent of their genes have human counterparts. They are transparent as embryos, making it easy to study changes in anatomy. We wanted to see if any of the aminosterols made the fish’s ability to regrow tissue happen faster and better.

We started with a simple amputation test, cutting off part of the tail and adding various aminosterols to the water in the fish tanks. Nothing happened. That changed, however, when we got some help from Helen Roberts, a recently graduated high school senior working as an intern in Yin’s laboratory. Roberts developed methods to inject substances directly into the zebra fish rather than adding them to the water. When she did this with MSI-1436, it stimulated the rate of tail fin regeneration by more than 300 percent. Instead of taking 10 to 12 days to regenerate, the fin took only three to four days, and there were no signs of abnormal growth. We had Roberts and a lab technician independently repeat the experiments, comparing different compounds, and made sure they did not know which one they were injecting into the fish. MSI-1436 worked in each situation; other compounds did not. This was stunning and prompted some exclamations of excitement in Strange’s office that are not appropriate to repeat here.

How did MSI-1436 stimulate regeneration in such a dramatic fashion? Some scientists had studied its effects on cells, and after we did more experiments, the answer seemed pretty clear: MSI-1436 hobbled an enzyme named protein tyrosine phosphatase 1B (PTP1B), which has several jobs in the body, one of which is to regulate the growth of new cells. That is an important occupation because widespread uncontrolled growth can make an organ malfunction or become cancerous. PTP1B is essentially a brake on cell regeneration. Our compound released that brake but only at injury sites, in a very local, focused and controlled way.

Zebra fish heart, damaged and then dosed with the compound MSI-1436, quickly regrew muscle and regained blood-pumping ability. Credit: Christina A. Dykeman and Yin Laboratory; MDI Biological Laboratory

When PTP1B brakes, it does so by interfering with a crucial class of cell proteins called receptor tyrosine kinases, or RTKs. RTKs are embedded in cell membranes and form parts of signaling pathways that start outside the cell and lead inside; the signals the path carries tell a cell to grow and divide. To become active and pass those signals along, RTKs need to be bound to another type of molecule, called a phosphate group. PTP1B gets in the way because it cuts phosphate groups away. No phosphate, no RTK signaling and no cell regeneration. But our compound, MSI-1436, disables PTP1B’s phosphate-cutting ability. And with these brakes disabled, RTKs and cell regeneration run happily along.

Heart disease and heart attacks

In addition to regrowth of the zebra fish tail fin, we found that our PTP1B blocker stimulates regeneration of the zebra fish heart. That is quite important because while humans may not have a tail fin, we do have a heart, and it often needs help. Cardiovascular disease is the leading cause of death worldwide, killing about 18 million people every year, and 85 percent of those deaths are caused by heart attack and stroke. Heart muscle cells that die in an attack do not regenerate but instead form a scar that increases the chances of another attack. A 45-year search for treatments, including stem cell transplants, to help the heart repair itself has failed.

So when we saw that MSI-1436 helped fish, we moved on to test it in mice, an animal model widely used in heart disease research. We induced heart attacks in the rodents and then injected them with MSI-1436 every three days over a span of four weeks. The blood-pumping ability of the organ improved by more than twofold, the amount of scar tissue was reduced by 50 percent, and heart muscle cells at the injury site proliferated by nearly 600 percent. MSI-1436 is the only small molecule known to have this effect.

Recently we began testing the compound in mice with a completely different kind of disease: a rodent version of Duchenne muscular dystrophy. This is a slow, degenerative muscle-wasting ailment, quite distinct from the sudden damage of a heart attack. Our preliminary data indicate that MSI-1436 prompts enough cell regeneration to keep skeletal and heart muscles ahead of the wasting. It does not stop the disease, but it may mitigate its effects.

On from animals

What bodes well for humans is that the compound stimulates tissue regeneration in both zebra fish and adult mice. These animal species are separated by approximately 450 million years of evolution. Because MSI-1436 works on such distinct creatures, the compound most likely targets cellular pathways that have been strongly conserved, or reused, by evolution in organism after organism. It increases the chances that such pathways exist in people and can be manipulated in a like fashion.

Testing drug candidates in a diverse population of humans, however, is very different from tightly controlled lab animal studies. The potential for failure in clinical trials is high. And although there are good reasons to be optimistic about MSI-1436, the reality is that we will not really know if it is effective in treating heart attacks until we try it in human patients. As a first step in that direction, we have begun National Institutes of Health–funded tests of this drug candidate in a pig heart attack model. The pig heart is remarkably similar to the human heart, and the size of the animal allows us to mimic a human heart attack and its early-stage treatment much better than we can in mice. If the pig trial results are positive, we will be well positioned to seek permission from the FDA to conduct clinical trials.

In our studies, we are also going to be watching out for signs of cancer. A concern in regenerative medicine is that treatments to stimulate tissue growth and repair may trigger uncontrolled cell proliferation, which is the biological hallmark of a cancerous cell. We believe that this concern is lower with MSI-1436. The extensive toxicity testing already done on the compound during its earlier incarnation as a diabetes and obesity drug was designed to identify problems such as cancer. None were found, and the FDA deemed MSI-1436 safe to use in studies of human patients. Limiting the presence of PTP1B also seems reasonably safe. The gene responsible for making it was first knocked out in mice in 1999. These mice have been studied extensively. They showed no signs of overt tumor growth, which suggests that even long-term inhibition of PTP1B does not cause cancer. Plus, treatment using MSI-1436 to stimulate tissue regeneration would likely last only a few weeks or months.

Finally, our own experiments indicate that MSI-1436 acts only at an injury site and does not send cells in normal tissue into a kind of cancerous overdrive. In zebra fish and mice, we did not observe tissue overgrowth or abnormalities in tissue and organ shapes (a sign of growing malignancy) when injured animals were treated with the compound. We tested this idea in one-cell zebra fish embryos, a highly sensitive point in the development of the fish. Embryos injected with the compound for 14 straight days developed into normal adult animals. Going from a single cell to a full-blown animal is, obviously, a complex process that requires tremendous cell proliferation and differentiation. Many drugs and environmental factors easily send it wildly off-kilter when given at such an early stage. It is reassuring that MSI-1436 does not.

The natural advantage

Perhaps animals respond well to the compound because it evolved in animals in the first place. It was not identified in a genetically engineered lab mouse or in cells grown in a dish at a medical center or from a screen of tens of thousands of synthetic chemicals at a drug company. Our findings came out of lessons we learned from dolphins, sharks and zebra fish. MDIBL, where we took advantage of those lessons, was founded to do exactly that. Our institution began as a marine research station on the coast of Maine in 1898, when biologists wanted an immediate connection to the natural world they were trying to understand.

This link is, unfortunately, something that the larger biomedical research enterprise, and the pharmaceutical industry in particular, has drifted away from. There is an important role for computer-designed molecules, of course. But regenerative medicine biologist Alejandro Sánchez Alvarado of the Stowers Institute for Medical Research, who is not involved in our research, has told us that MSI-1436 is “a great case study of what happens when scientists choose to walk away from the familiar and search nature for answers to vexing biomedical problems.”