Slaying Cancer At Its Roots

James Inglese knew he had to choose carefully. It was 2004, and Inglese, deputy director of the National Institutes of Health’s Chemical Genomics Center (NCGC), had a plan that could save time and resources for scientists screening large chemical libraries for enzyme inhibitors or activators. But he needed just the right protein to test the idea. It had to be something familiar, something well understood, but something that wasn’t such a hot target that it would steal the spotlight from the screening method itself. A colleague, Douglas Auld, suggested pyruvate kinase, an enzyme that helps break down glucose. That seemed to fit the bill just fine, Inglese recalls.

Because of its screening efforts, Inglese’s team is poised to make some big contributions to a promising area of cancer research, one in which the altered metabolic state all cancer cells share—rather than gene mutations specific to certain cancers—could be the key to finding new treatments (C&EN, Oct. 26, 2009, page 20). Coincidentally, pyruvate kinase has emerged as a key player in maintaining the altered metabolism of cancer cells and is thus a potential target for cancer drugs; the enzyme is a scene stealer after all.

The inspiration for this research movement goes back nearly 100 years. In the 1920s, German biochemist Otto Heinrich Warburg learned that cancer cells and healthy cells metabolize glucose, and ultimately obtain energy, in different ways. Normal human cells use an oxygen-demanding pathway to break down glucose but can switch to an oxygen-free route when oxygen runs low. Tumor cells, in contrast, go the oxygen-free route no matter how much oxygen is around. Warburg, who would later become a Nobel Laureate, surmised that this difference gives cancer cells advantages in the growth department, a hypothesis that today bears his name.

In ensuing decades, the sugar biochemistry of the Warburg hypothesis took a backseat to other ideas. In the clinic, agents that interfered with nucleotide biochemistry, such as 5-fluorouracil, became popular chemotherapeutics. And the molecular biology revolution shifted the cancer research spotlight away from metabolism to the myriad gene mutations that occur in scores of different cancers. Today, though, advances in several areas are helping researchers understand the implications of Warburg’s hypothesis as never before and might lead to new cancer treatments.

In the 1920s, Warburg and his contemporaries answered many fundamental questions about biochemistry, but they never figured out how normal cells and cancer cells regulate their nutrient uptake, how the cells know that they have enough resources to grow and divide, explains Craig B. Thompson, a cancer researcher at the University of Pennsylvania. The renaissance of the Warburg effect and studies of cancer and metabolism are about answering those questions, about understanding the fundamental growth defects all cancer cells have in common. “This is about attacking cancer at its roots,” Thompson says. And it just so happens that one of those roots might be pyruvate kinase.

The body can make several forms of pyruvate kinase, and tumor cells harbor the same type that shows up in other rapidly dividing cells, such as those in an embryo, explains Lewis C. Cantley, a biochemist at Harvard Medical School. Cantley’s team has found that this rogue form, called pyruvate kinase M2, is critical for rapid growth in cancer cells (Nature 2008, 452, 181 and 230).

“The goal of a cancer cell is to grow and survive,” Cantley says. And one thing a growing cell needs is chemical building blocks such as nucleic acids, amino acids, and lipids. Pyruvate kinase M2 works more slowly than its counterpart in healthy cells, but it is better at helping to generate lots of those necessary metabolites.

“I like to picture a conveyor belt,” says Matthew B. Boxer, a chemistry team leader who works with Inglese at NCGC. If glycolysis, the process that breaks glucose down to form pyruvate, is a series of conveyor belts, pyruvate kinase sits at the end of the line. If the last belt in a series is slow, “you get a pileup in the middle,” Boxer explains. The buildup of glycolysis intermediates can then spill over into other metabolic pathways that build whatever the cancer cell needs, he says.

Access to building blocks might not be the only reason that cancer cells’ metabolism differs from normal cells’, says Benjamin F. Cravatt, a chemical biologist at Scripps Research Institute. The pileup of glycolysis intermediates might also be signaling molecules that direct cancer cells’ behavior and that could even instruct the cells to become more aggressive. Cantley’s work shows that “there is probably tremendous cross-talk between signaling pathways and metabolic pathways,” Cravatt says.

From a drug development perspective, Cantley’s results got other cancer researchers excited because the subversive pyruvate kinase M2 in tumors isn’t present in most healthy adult tissues. Thus, it might be possible to target the enzyme with a cancer drug, says Matthew G. Vander Heiden, a former postdoctoral fellow in the Cantley group who has since started his own laboratory studying cancer metabolism at Massachusetts Institute of Technology.

At a 2008 conference in Australia, after presenting his group’s pyruvate kinase findings, Cantley heard about a team that had just the type of screening expertise he’d need to target pyruvate kinase M2 with a drug. A few phone calls later, Cantley was in touch with Inglese and his NCGC colleagues, including Auld, who is the group leader for genomic assay technologies. “It all just came together for us,” Auld says, as the excitement around Cantley’s pyruvate kinase results melded with the NCGC team’s screening expertise, which had serendipitously been optimized for that very enzyme (Proc. Natl. Acad. Sci. USA 2006, 103, 11473).

The collaborative effort is well under way. By screening nearly 300,000 molecules in NIH’s Molecular Libraries Small Molecule Repository and optimizing the hits they found, the NCGC researchers identified a set of diarylsulfonamide activators that are highly selective for the rogue kinase in tumors (J. Med. Chem. 2010, 53, 1048). The molecules boost the sluggish kinase’s activity so it behaves more like the faster-acting form found in healthy adult cells. The team is now exploring other classes of molecules that do the same.

What’s really intriguing is how the new activators seem to work, says Hee-Won Park, a collaborator who belongs to both the Structural Genomics Consortium (SGC) and the University of Toronto’s pharmacology department. Park, SGC research associate Bum-Soo Hong, and colleagues determined the X-ray crystal structure of pyruvate kinase M2 bound to several of the diarylsulfonamides. They have deposited their data into the publicly available Protein Data Bank.

At the molecular level, the new activators coax pyruvate kinase M2 into its active state, which contains four copies of the enzyme locked together. The molecules complement fructose bisphosphate, a metabolite that is a natural activator of the enzyme, by binding to sites that the metabolite does not occupy. Together, the compounds confer superstability to the kinase’s active state, Park says, perhaps preventing the buildup of metabolic building blocks that sustain cancer cells.

As interesting as the structures seem, “we want to know if that’s how things really work in cancer cells,” Park says. Now, Vander Heiden’s lab at MIT is testing the best of the diarylsulfonamides to find out.

It’s likely that other enzymes throughout humans’ vast metabolic network also have rogue forms that might be tackled in a similar way, Vander Heiden says. Already in 2007, Cantley, Thompson, and fellow cancer researcher Tak W. Mak of the University of Toronto cofounded a biotechnology company inspired by that idea: Agios Pharmaceuticals, in Cambridge, Mass. Vander Heiden was one of several people who signed on as a consultant for the company, which derives its name from a Greek word meaning “holy” or “saint.” In mid-2008, privately held Agios landed $33 million in venture capital funding.

The pyruvate kinase M2 story is just one piece of the complicated puzzle that is metabolism and cancer, and it’s too early to tell how it will end. But as Inglese and his colleagues can attest, you never know where you’ll find more targets.