Researchers identify new way skin stops tumor growth

‘Oncogene-induced differentiation’ joins two other strategies for counteracting potentially cancer-causing DNA changes
Three doors, three anti-tumor choices
Hutch scientists have discovered a third anti-tumor strategy, called oncogene-induced differentiation, that cells can use to stop tumor formation. Image by Kim Carney / Fred Hutch News Service

Cancer is so closely associated with DNA mutations that it seems almost inevitable that such changes must lead to cancer. But a square inch of skin can harbor tens of thousands of DNA changes and yet be tumor-free. How?

Cells, as it happens, aren’t passive victims of DNA damage. When faced with what could be cancer-causing DNA alterations, some cells sacrifice themselves. Others turn off their ability to divide. And there’s a third way, just discovered by scientists at Fred Hutchinson Cancer Research Center: they get specialized.

Skin is continually undergoing a cycle of shedding and creation of new skin. This process is sustained by skin stem cells, which balance two functions: creation of new specialized skin cells and renewal of themselves to ensure a never-ending source of new skin cells. In work published today in Nature Cell Biology, Hutch researchers show that skin stem cells in mice respond to what ought to have been a cancer-causing mutation by opting to differentiate, or specialize, instead of renewing themselves. Because differentiated skin cells eventually slough off, the strategy appears to allow skin to jettison dangerous mutations without disrupting its function.

“We’ve been uncovering ways that tissues ensure their function and health long term, despite getting these [cancer-associated] mutations,” said senior author Dr. Slobodan Beronja, who studies the genes and molecular pathways that underlie cancer. The new mechanism the team discovered for preventing a mutated cell from becoming cancerous “just sort of nudges it gently out of the tissue using a physiological process of differentiation,” he said.

Slobodan Beronja
Dr. Slobodan Beronja Fred Hutch file

An unexpected finding

Much of Beronja’s previous work has focused on specific genes that, when mutated, cause tumors to develop. The new findings sprang from what initially seemed to be a failed experiment, said first author Dr. Zhe Ying, a postdoctoral fellow in Beronja’s laboratory.

The two researchers wanted to focus on commonly mutated genes in squamous cell carcinomas that arise in areas such as the skin and lining of the mouth. The idea was to mutate many of these genes simultaneously and see how they affected tumor development in a mouse model of skin cancer.

But after months of prep work and a year of waiting, “There was not a single squamous cell carcinoma in the skin,” Ying said.

He was surprised to see that several genes thought to promote tumor development instead actually suppressed it. One of these was the most-mutated, cancer-promoting gene in squamous cell carcinoma, called Pik3ca. This gene codes for a component of PI3K, a protein that regulates a complex molecular pathway that is often disordered in skin cancer.

The finding “was totally unexpected,” Ying said, which meant they next had to figure out what it meant.

A tale of two choices

Skin is one of the fastest dividing tissues in the body. The stem cells that continually produce new skin cells divide every couple of days. At every cell division, each of the two new daughter cells must choose between staying a stem cell (and renewing itself) or becoming a specialized skin cell.

Cancer researchers often focus on cell division as a key cancer metric. It makes sense: More division usually equals more cells, and more cells generally equal a bigger tumor. However, when Beronja and Ying looked at the tumor-free mouse skin that they’d manipulated to carry Pik3ca mutations, they saw that these cells divided about twice as fast as normal cells. But the acceleration seemed to have no effect: No tumors developed, and the skin wasn’t any thicker than healthy skin.

When the team looked more closely, the mutated cells didn’t seem to be balancing their hectic division with death (called apoptosis, it's a well-known strategy to stop tumors before they start) or with going to sleep, also known as senescing.

Ying and Beronja realized that the skin’s ability to suppress tumors must rest on the stem cells' fate choice: to renew or not renew. Renewal — providing an endless source of new tumor cells — is exactly what cancer needs. If eternal renewal could sustain tumor growth, could differentiation, the opposite choice, be a road to tumor suppression?

To answer this question, the researchers needed to directly measure stem cells’ renewal and differentiation. Ying and graduate student Madeline Sandoval developed a method to track the daughter cells of stem cells that had divided. Using their method, they found that when Pik3ca mutates, it tips the balance toward differentiation and away from renewal.

In normal adult tissue, renewal and differentiation strike an even balance: Half the daughter cells from a pool of stem cells differentiate and half remain stem cells.

In contrast, cancer-associated mutations in Pik3ca cause about 55 of every hundred daughter cells to differentiate and only 45 to remain stem cells. Over time, this means that the mutated stem cells will slowly be depleted as more and more eventually differentiate.

Once they saw that mutated Pik3ca was associated with a lower renewal rate, the researchers knew they needed to trace dividing skin stem cells as they made the choice to renew or divide, Ying said.

Using a method developed by collaborators at Yale University that allows them to watch stem cells divide, renew and differentiate in living mice, Ying and Beronja were able to confirm their findings. They were also able to trace the molecular pathway from PI3K to a protein called SH3RF1, which appears to be a major player in steering the cells toward differentiation.

Because the phenomenon was triggered by cancer-causing, or oncogenic, changes to the Pik3ca gene, the team dubbed it “oncogene-induced differentiation.”

“The cool thing is, by maintaining that cell in the tissue, it’s not disrupting tissue architecture. … [It is] skin employing a natural mechanism to control its health,” Beronja said.

Can cancer cells be guided back to health?

The findings suggest that excessive cell division is a red herring when it comes to skin cancer, Beronja said.

“Our study says that proliferation doesn’t matter. Apoptosis [programmed cell death] and senescence don’t matter. You do or you don’t get cancer in skin depending on whether your progenitor renewal–differentiation process is affected,” he said.

Tumor cells often regress to a more stem-like state, which fits with Beronja’s theory that cancer occurs when stem cells’ fate choice gets unusually weighted toward renewal. Though their work focused on skin, Ying and Beronja suspect that other fast-dividing tissues, such as the lining of the intestine and the oral cavity, may combat squamous cell carcinoma using a similar process.

The big open question now is how squamous skin tumors overcome oncogene-induced differentiation. Ying is working on untangling this mystery. Perhaps the tumor cells accumulate more mutations in key genes that switch off this process. Or perhaps cells get stuck in what should be a temporary process, such as wound healing. In normal tissue wound healing is short term: skin stem cells sense that they need to increase their self-renewal to close the gap in skin, and then sense when healing has occurred and it’s time to return to normal. What if cancer occurs when stem cells can’t sense that the wound has healed?  

Looking ahead, the work may have implications for development of new therapeutics. It’s often assumed that the most commonly mutated genes must drive tumor formation and that targeting them therapeutically will hamstring the tumor’s ability to survive and grow. The current study suggests that it may be more complicated. Merely targeting the most frequently mutated genes may not be an effective drug-development strategy, Ying said.

It also suggests that it may be possible to take advantage of skin cells’ natural tumor-suppressive process to combat cancer. Cancer often co-opts natural processes for its own ends. What if scientists could turn this around on tumors?

“Can we actually go and awaken this [process] in an established tumor as a way to treat it — not by killing proliferative cells, but just guiding proliferative cells out of existence through differentiation?” Beronja said.

The National Institutes of Health and a Thomsen Family Fellowship supported this research.

Sabrina Richards, a staff writer at Fred Hutchinson Cancer Center, has written about scientific research and the environment for The Scientist and OnEarth Magazine. She has a PhD in immunology from the University of Washington, an MA in journalism and an advanced certificate from the Science, Health and Environmental Reporting Program at New York University. Reach her at

Related News

All news
Good News: Normal skin cells keep cancer at bay Celebrating faculty and staff achievements August 3, 2017
Skin cancer: 9 things to know to lower your risk From cover-ups to caffeine, here are some tips for protecting your skin during the summer and all year long July 2, 2015
Untangling the kinetochore Dr. Changkun Hu named Jane Coffin Childs – HHMI Fellow; Boosts research into critical and complex molecular machine November 29, 2023

Help Us Eliminate Cancer

Every dollar counts. Please support lifesaving research today.