Bacteria unlock secrets that may aid cancer treatment

The murky flasks of bacteria growing in Ben Shen’s lab may change how we look at both chemistry and chemotherapy.

These strains of “designer bacteria” have been bioengineered to pump out powerful chemical compounds that offer new biochemical strategies in the battle against cancer.

Bacteria are rich sources of the naturally occurring chemical compounds known as natural products, many of which have garnered attention in the world of drug discovery for potent protective properties against natural threats like infection and cancer.

“If you look at the track record of all drugs, natural products remain the best source of drugs,” says Shen, the Charles M. Johnson chair in the School of Pharmacy and professor of chemistry and pharmaceutical sciences.

And when it comes to cancer, he adds, more than 70 percent of the roughly 170 current anticancer drugs are either natural products or based on natural products.

Since 2005, Shen’s blend of basic and applied chemistry has been instrumental in launching a National Cooperative Drug Discovery Group (NCDDG) on campus, one of nine centers nationwide funded by the National Cancer Institute to promote innovative research for new anticancer agents.

The UW–Madison group, which focuses on natural products, is based in the Paul P. Carbone Comprehensive Cancer Center and brings together campus experts in the many steps of drug discovery, from basic biochemistry to preliminary testing.

“The NCDDG brings all these resources together to establish the infrastructure so that we are one of the few academic research communities where not only can we do the discovery, but we also have the capacity to do enough pre-clinical studies to push the compound closer to clinical application,” says Shen. “That is something that not many campuses can do.”

Shen cultivates strains of Streptomyces, common soil bacteria, for their production of several natural products including an intriguing compound called enediyne C-1027 that he calls “the most potent anticancer agent ever known.” Though two related members of the enediyne chemical family are clinically used chemotherapy treatments, C-1027 has not made it anywhere near a hospital in the 20 years since it was first discovered — it’s simply too strong and too toxic to be tolerated by patients.

“As far as potency is concerned, many natural products are already very good, but they never made it into the clinic because there are side effects that conventional technology couldn’t easily overcome,” Shen says.

His group aims to improve on nature’s design, using novel technology to tailor promising natural products with an eye toward clinical use. They create sets of sibling compounds, each slightly different from its parent, then screen them for functional improvements — improved specificity or targeting, for example, or reduced side effects and toxicity.

While the idea is straightforward, execution is not so simple. For all their promise, natural products are notoriously difficult to work with. The compounds tend to be large and complex, bristling with unfamiliar structures beyond the scope of standard medicinal chemistry.

What makes these compounds difficult, though, is also what makes them so promising, Shen says. “Structural complexity is a great challenge, but by itself it’s also a wonderful opportunity to discover novel chemistry.”

For example, the core of enediyne compounds is a large carbon ring structure with unusual bonding patterns. “Probably the making of this compound involves something that cannot be explained by the collective knowledge of today — and presumably new science will be there,” he says.

Since natural product variants are too complicated to manipulate directly or make from scratch in the lab, Shen’s team instead harnesses the natural synthetic power and efficiency of the bacteria. They alter individual genes involved in the bacterial biosynthetic machinery — akin to switching out a single machine of an assembly line to result in a slightly different final product.

Called metabolic pathway engineering, this approach allows them to create and test new potential drug compounds without needing to know every step involved in their synthesis.

The approach has other advantages as well. Because Streptomyces bacteria produce enediynes through simple fermentation, the whole process is much more environmentally friendly than traditional synthetic chemistry. The bacteria are also easy to grow and large-scale fermentation can pump out compounds of interest in large quantities, overcoming the limited availability that has historically dogged natural products.

Metabolic pathway engineering capitalizes on recent advances in biotechnology that have rapidly generated vast amounts of biological information, such as complete genetic sequences of several organisms.

From such a sequence, Shen’s group has uncovered a wealth of novel biochemistry coded in the bacterial genes involved in C-1027 biosynthesis. In two papers published last month in the Proceedings of the National Academy of Sciences, they reported a brand new chemical pathway for the common biological molecule chorismate and a novel class of enzymes integral to the synthesis of the unusual but characteristic core of enediyne compounds.

In another surprising set of experiments, published last year, one of the synthesis enzymes presented the team with a perplexing set of clues. The expected function — predicted based on its genetic similarity to known enzymes in other species — did not match the process needed to generate the observed compound, a type of amino acid. Some sleuthing led them to an unexpected set of biochemical steps that revealed a novel function for the type of enzyme and a new way to build the observed amino acids.

“When you only have one anomaly you say, ‘oh, that’s an exception’,” Shen says. “But if you can create the exception once, twice, three times, then you start to question yourself: Is this really an exception or just something we don’t understand?”

Perhaps their most exciting anomaly is a variant of C-1027 that opens the door to a whole new way to tackle tumors.

Existing chemotherapy drugs and radiation therapy rely on oxygen to damage DNA and kill tumor cells. “The problem is that inside a solid tumor or mass, there is very little oxygen,” Shen says, with the result that cancer treatments’ effects are severely blunted exactly where they are needed most.

One of Shen’s C-1027 variants, called desmethyl C-1027, circumvents this problem. The small modification of a single chemical group allows the new compound to interact with DNA in a completely novel — and, crucially, oxygen-independent — manner. Rather than breaking DNA strands as the oxygen-dependent drugs do, desmethyl C-1027 acts as a bridge permanently linking DNA’s double helix into an inflexible and non-functional ladder, a process called interstrand cross-linking.

This new anticancer strategy, reported last November in the Proceedings of the National Academy of Sciences, has raised interest in desmethyl C-1027 as a novel type of chemotherapy agent. “This new mechanism can inspire new hypotheses to address issues that are faced today in chemotherapy,” Shen says.

Shen’s collaborators are now testing desmethyl C-1027’s performance in cancer model systems, the next step toward a potential clinical application.

“You have to understand the underlying genetics, biochemistry, chemistry, all of that, to take full advantage of this knowledge explosion and translate that into a tangible finding,” Shen says. “We are fortunate to have a chance to translate some of our fundamental research into drug discovery.”