Turning Sweet on Cancer

See allHide authors and affiliations

Science  11 Jul 2003:
Vol. 301, Issue 5630, pp. 159-160
DOI: 10.1126/science.301.5630.159

A focus on how cancer cells make the complex sugars that dot their surfaces suggests novel chemical tricks that can stop metastasis and perhaps tag cancer cells in the body

Call it sweet-talk. One of the key ways cells communicate with each other is by sticking sugars onto a variety of proteins on the cell surface. These carbohydrate groups help negotiate the cell's relationships with its external environment; they can determine, for example, whether an interaction with another cell or protein is standoffish or ends up in a tight embrace. Cancer cells speak a different sugar dialect than do normal cells, a difference noted 35 years ago and one that researchers have spent 2 decades trying to parlay into ways of selectively targeting cancer cells for destruction. But after some tantalizing results in the early 1990s failed to translate into clinical success, sugars went out of favor as drug targets du jour. Now, armed with new chemical tools for studying how cells construct and use the complex sugars on their surfaces, investigators are again exploring sugar-related processes in the war on cancer.

It's become clear over the past decade that glycosylation, the biochemical process of putting sugars onto proteins and other molecules, is “critically important to many of the signaling pathways involved in turning a normal cell into a cancer cell,” explains Harvard Medical School biochemist Norbert Perrimon, who studies the role that polysaccharides play in signal transduction. “If you were able to inhibit specific glycosylation reactions, you might be able to alter these pathways and turn off the cancer cell.” Adds Ken Irvine, a developmental biologist at Rutgers University in Piscataway, New Jersey, who's been studying a key cancer pathway that turns out to be regulated by a sugar-containing protein: “What [sugars] are providing is a completely different way of targeting cancer cells.”

One promising approach, says James Paulson, a glycobiologist at the Scripps Research Institute in La Jolla, California, involves diverting specific glycosylation pathways into a metabolic dead end. Jeffrey Esko, a glycobiologist at the University of California, San Diego, whom Paulson cites as the main innovator in this area, hit upon an idea for doing this after studying how cells put together multisugar chains, particularly those that contain a type of sugar called sialic acid.


Complex sugars on a cancer cell's surface may enable wayward malignant cells to stick to blood platelets and migrate to distant sites in the body.


A sialic acid-rich carbohydrate known as sialyl Lewis X juts out from many cells, especially cancer cells, and binds to molecules known as selectins that are found on the surfaces of platelets and endothelial cells. This binding enables cancer cells to spread, or metastasize, beyond their point of origin. Ten years of experimental data from numerous groups worldwide have shown that patients whose cancer cells express sialyl Lewis X—about 25% to 35% of patients with breast, colon, thyroid, and gastric cancers—have a much poorer prognosis for survival.

Esko and his co-workers established that specific two-sugar units, known as disaccharides, serve as primers for cells to start making sialyl Lewis X. By modifying these disaccharides with various chemical groups and adding the modified primers to cell cultures as decoys, the researchers found that they could shunt at least some of a cancer cell's carbohydrate-forming reactions away from the pathway that makes sialyl Lewis X on proteins. Although the cells still made some sialyl Lewis X, they bound less avidly to selectin-containing cells. The reason, explains Esko, is that the selectins must bind to multiple sialyl Lewis X chains simultaneously, “so you only have to knock its level down by a factor of 2 or so to have a big effect on binding.”

In subsequent experiments, published last month in Cancer Research, Esko showed that tumor cells treated with one of the decoys failed to form lung tumors when they were injected into immune-compromised mice. He obtained the same result when the decoys were administered by infusion pump directly into the animal, rather than as a pretreatment. Moreover, for reasons that are still unclear, the injected tumor cells were more susceptible to attack by immune system cells.

Sweet tricks.

UCSD's Jeff Esko and UC Berkeley's Carolyn Bertozzi are pioneering sugar trickery in the cancer field.


Additional experiments have not found any adverse immune system response in animals treated with the decoy, a concern because sialyl Lewis X is thought to play a role in various inflammatory responses. “So far, the data look good, but we need to improve the solubility and other physical properties of the decoys themselves in order for them to be potential anticancer drugs,” says Esko. Paulson, who co-directs the Consortium for Functional Glycomics—an academic collaboration funded by the National Institute of General Medical Sciences—has seen the data from Esko's lab and is surprised at his colleague's reticence. Although the approach could still fail for several reasons, “the fact is, it stops the cancers from metastasizing and it does so with relatively small amounts of compound. I'm impressed.”

Tagging cancer

Decoys are just one of the tricks that glycobiologists are developing in an attempt to fight cancer. Chemical biologist Carolyn Bertozzi of the University of California, Berkeley, is pioneering efforts to exploit the extraordinary turnover of sugar-coated molecules characteristic of cancer cells. Her goal is to slip modified sugars onto the surfaces of cancer cells. These modified sugars contain chemical tags that, with the right touch, could serve as a homing beacon for both diagnostic and therapeutic applications. Paulson and others call her work “elegant.”

Bertozzi and her colleagues start with modified mannosamine analogs, sugars that cells naturally turn into sialic acid. When injected into mice, these sugars wind up on the surface of cells. The modified sugars contain an activated nitrogen group called an azide, which can then react with various phosphorous-containing chemicals that the researchers administer several days later. “We can do this in animals with no apparent ill effects,” says Bertozzi.

Bertozzi's group is hoping to use this reaction to label tumor cells for in vivo imaging—a Holy Grail for cancer researchers. The idea is to administer the nitrogen-containing sugars and then inject a compound containing a reactive tag that is visible by a standard imaging technique, such as magnetic resonance imaging. The tag would then get incorporated on the surface of any cell with the modified sugar. “Since tumor cells have a higher metabolic flux, you get turnover of glycoproteins on the cell surface and the modified sialic acids are preferentially incorporated there, more so than on most healthy cells,” explains Bertozzi. If the approach works in animals, experiments that should begin shortly, it could provide an in vivo signpost singling out cancer cells over all others. “That's a big if,” says Perrimon of the planned studies, “But if Bertozzi can pull this off, that would be a breakthrough.”

Although Esko's and Bertozzi's results are still preliminary, pharmaceutical interest in tinkering with the ways in which cancer cells use sugars is once again heating up, says Paulson. One of his consortium colleagues, he reports, just raised $19 million to start a biotech company to capitalize on sugar-altering methods. The time may finally be ripe for sugars to bring some sweetness to cancer drug development.

Navigate This Article