George Daley and his lab may have
found a new way to connect the dots
between cancer and diabetes.
(michelle.gray/Flickr)
By Tom Ulrich
Most of us think about cancer as a disease of genes gone awry – of mutations, deletions, duplications, etc. causing unchecked cell growth.
But could you also view cancer as a metabolic disorder, like type 2 diabetes? George Daley and his lab in the Stem Cell Transplantation Program at Children’s have found some intriguing molecular links that make this a plausible idea.
While it’s not yet clear what this means for patients with either disease, the findings help untangle some very perplexing data about human genetics and diabetes risk, and could change doctors’ thinking about the treatment of both conditions down the road.

Scientists have long known that cancerous and healthy cells don’t use sugar in the same ways. In the 1930s German physiologist Otto Warburg observed that cancer cells metabolize glucose via different biochemical pathways than healthy cells, a phenomenon known now as the Warburg effect. Tumors also tend to burn up glucose more quickly, which is why they show up so brightly onpositron emission tomography (PET) scans, which measure tissues’ rate of glucose use.
Our story starts when Daley’s team set out to learn more about the relationship between cancer and a pair of molecules: a microRNA called let-7 and a protein called Lin28. Let-7 dampens the expression of genes related to cell growth and development. Lin28, which keeps stem cells from differentiating at the wrong time in the developing embryo, blocks let-7′s production – in essence disabling the growth check. It is also active in about 15 percent of all cancers, the Daley lab has shown.
“The relationship between Lin28 and let-7 is ancient, found in organisms as diverse as worms, mice, and humans,” Daley says. “This suggested to us that the Lin28/let-7 pathway was profoundly important, but we didn’t expect such exciting results.”
To learn more about the physiology of the pair, Daley’s team engineered a trio of mouse models: one in which Lin28 was overproduced, a second lacking Lin28 in the muscles, and a third in which let-7 production could be dialed up or down.
Here is where things took a turn. Instead of increased or decreased rates of cancer, they saw profound changes in how the mice processed glucose. For instance, when fed a high-fat diet, the mice that produced surplus Lin28 remained lean and continued to use glucose efficiently, while their normal counterparts became obese and diabetic. The mice that overproduced let-7, on the other hand, became diabetic – even on a normal diet.
“The results were startling,” Daley says. “Previously we had considered these molecules only as regulators of cell growth and cancer. But in these mice we discovered remarkable effects on sugar processing and diabetes.”
But does this mean anything for people? Daley’s team turned to data from genome-wide association studies (or GWAS) on type 2 diabetes. GWAS look for genetic variations within populations to find markers associated with disease. These studies are particularly good when trying to understand complex diseases like asthma, heart disease, or diabetes, where common variations in several genes contribute to the disease state.
Healthy and cancerous cells use
sugar in very different ways. Lin28
and let-7 may help us understand
why. (Uwe Hermann/Flickr)
Working with Harvard Medical School geneticist David Altshuler, they found that many diabetes-associated genes pinpointed in the GWAS data were either known or predicted to be acted on by let-7, connecting the genetic data with the physiology they saw in the mouse models.
“It can be difficult to link genetic association data to actual biochemical pathways in the cell,” Daley explains. “Let-7 has hundreds, maybe even thousands, of gene targets. The genetic data, coupled to the striking similarities in the way our mice and patients with diabetes handle glucose argues that regulation by Lin28/let-7 is a unifying feature of many genes associated with diabetes.”
So where is the connection? Why might a pair of molecules with clear metabolic effects also have such strong ties to cancer? Back to the Warburg effect, which in many ways reflects how cells in the developing embryo use sugar. “Cancer cells have a metabolism that is very embryo-like,” says Daley, “so we speculate that tumors take advantage of the Lin28/let-7 pathway to turn back the clock, in metabolic terms. It’s a avenue we’re going to study further.”
Tom Ulrich is a senior science writer in the Children’s Hospital Boston Department of Public Affairs, covering laboratory and clinical research innovations across the hospital. Over the last ten years, Tom has parlayed his curiosity about science and passion for science writing and communications into a number of roles, including development writer at Dana-Farber Cancer Institute, marketing writer at AIR Worldwide, and editorial & account director at Feinstein Kean Healthcare. Most recently, he was the communications manager at Harvard Catalyst | The Harvard Clinical and Translational Science Center. Tom earned a master’s degree in molecular microbiology and immunology from the Bloomberg School of Public Health at Johns Hopkins University, and is an amateur photographer.