Turning tumors against themselves to stop metastasis

Tumor cells need the right environment in order to spread. But a protein that some tumors themselves produce can make some environments inhospitable to metastasis. (Albh/Wikimedia Commons)
With cancer, it’s generally not the primary tumor that kills people, but metastasis—the spread of cancer to locations far from the original tumor.

Finding ways of stopping metastasis has proven immensely challenging. On some level, it’s a problem with the models that we use to study metastatic cancer. But it’s also a matter of understanding why particular tumor types spread where they do—like prostate tumors to the bones or breast cancer to the brain—and what about the microenvironment—the combination of cells, proteins and other factors—makes different sites in the body metastatically friendly to different tumors.

Randolph Watnick, PhD, and his research team in Boston Children’s Vascular Biology Program have been asking this question, and in the process have found that a protein called prosaposin can make sites unfriendly to metastasis. Interestingly, it’s a protein that some tumors actually make themselves.

But even better, Watnick has found that a tiny fragment of prosaposin—a peptide that is a mere five amino acids long—has the same anti-metastatic power of the full protein, making it highly attractive for drug development. He and his collaborators reported the full story in a recent paper in the journal Cancer Discovery.

Using models of breast, prostate and lung cancer, Watnick and his team found that tumors release signals that trick a particular kind of cell called a monocyte into migrating from the bone marrow to potential metastatic sites in the body. If a tumor is metastatic, it further fools the monocytes into helping prepare that site’s microenvironment, so that cancer cells can move in.

But tumors that cannot metastasize also produce prosaposin. And prosaposin tells those same monocytes to release another protein called thrombospondin-1. This protein, in turn, potently inhibits blood vessel growth, or angiogenesis—in effect denying nascent metastases access to the oxygen and nutrients they need to grow, thereby stopping metastasis in its tracks.

It would seem logical, then, to try to develop drugs based on thrombospondin-1. And this has been tried to no avail thus far. Thrombopondin-1 is a very large protein, too large to serve as a drug, and shortened versions of it have not worked well.

However, that doesn’t mean that thrombospondin-1 can’t be exploited indirectly. “If we can trigger monocytes recruited by pro-metastatic tumors to produce thrombospondin-1 like those recruited by non-metastatic tumors,” Watnick said in a press release, “we will be able to hijack the mechanism by which tumors create metastasis-permissive sites to close the door on those sites.”

A five-amino acid stretch of prosaposin contains all of the full protein’s thrombospondin-1-triggering activity.

So Watnick and his collaborators turned back to prosaposin, experimenting with shorter and shorter versions to see how small they could make it while still retaining the ability to trigger thrombospondin-1 production in monocytes. “We started with an 80-amino acid long peptide from prosaposin,” Watnick says, “and then made overlapping fragments to narrow down the active site.”

From there, they whittled prosaposin down to a 20-amino acid segment, then 13 amino acids, then six. They finally found a five-amino acid stretch that contained all of the full protein’s thrombospondin-1-triggering activity.

“We’ve tried the peptide in models of lung and breast cancer,” Watnick says. “It completely blocked metastasis in both, with no toxicity.”

At this point, we must ask the $64,000 question: Why would a tumor make a protein that could prevent its own spread?

(Clockwise from upper left) Crystal structures of saposin A, B, C and D, the four products of prosaposin. A five-amino acid portion of saposin A has all of the anti-angiogenic power of the full protein. (Lambk68/Wikimedia Commons)
“I can’t answer that question with any real certainty,” Watnick admits. “We know that prosaposin is usually split into four separate, smaller proteins called saposin A, B, C and D. These proteins help cells metabolize lipids, and could help give tumor cells more energy for important things like cell division, invasion, migration and production of growth factors.

“So it could be that prosaposin production has a growth advantages in early stages of tumor development,” he continues. “But if a tumor cell makes too much of it, it gets secreted from the cell, where it has a different set of interactions that affect the microenvironment.”

Watnick is hopeful that his prosaposin fragment might serve as the basis for a drug to block a variety of metastatic cancers. “The peptide is about the same size as many of the small molecule inhibitors like imatinib [Gleevec®] or gefitinib [Iressa®],” he explains. “Theoretically, it should be possible to formulate it in multiple ways for different tumors, like a topical ointment for melanoma, an inhalable form for lung cancer, an oral form and so forth.”

The Technology and Innovation Development Office (TIDO) at Boston Children’s has filed patent applications on these peptides, peptide derivatives and their uses. To learn more or inquire about licensing or investment opportunities, visit the TIDO website.