A growing number of researchers in the Drosophila community are pioneering the use of Drosophila screens to identify potential drug candidates for cancer, cardiovascular disease and neurodegenerative diseases. (Source: Wikimedia Commons)Like most Drosophila researchers, Ross Cagan was a developmental biologist, studying eye development in flies at the Washington University Medical School in St. Louis (WUSTL). One day, a frustrated colleague was venting to Cagan about how even though researchers knew that a mutation in the RET gene caused medullary thyroid cancer (MTC), a successful therapeutic had yet to be developed. 

“Being an arrogant fly guy, I told him that he should just model it in flies,” says Cagan. “So, because flies don’t have thyroids, we threw oncogenetic RET in what we worked with, which was the eye, and the eye started to grow these tumors.”  Soon after, Cagan’s lab began using the Drosophila model to do what came naturally to them: studying the mechanisms and pathways in the disease. But then Cagan remembered his original conversation. It wasn’t about mechanisms and pathways; it was about patients.

So, he switched gears. His lab started screening compounds in the Drosophila MTC model in 96-well plates to discover a treatment for the disease. That resulted in the discovery of vandetanib, which in 2011 became the first drug approved by the U.S. Food and Drug Administration (FDA) for the treatment of late-stage MTC in adult patients not eligible for surgery.

“That worked out better than I had expected,” says Cagan. “I had literally one person working on it. Everyone else in my lab was working on something else. And at that point, I moved to Mount Sinai in New York, brought in a new group of people and focused just on doing this.”

And following Cagan’s lead, other fly guys and girls are also developing high-throughput screening assays that use Drosophila models to discover drug candidates for cancer as well as cardiovascular disease and neurodegenerative diseases. Right now, these researchers are outliers, questioned by their academic colleagues for moving into the translation space, while largely ignored by the pharmaceutical industry. And, in the end, their results will be how they prove to both the value of screening drugs in Drosophila.

Another model?

Following Cagan’s lead, other fly guys and girls are also developing high-throughput screening assays that use Drosophila models to discover drug candidates for cancer as well as cardiovascular disease and neurodegenerative diseases. (Source: Ross Cagan)To find a drug candidate that hits a drug target involves the trial-and-error process of screening libraries. Even with thousands of compounds, the chances of finding a drug candidate are pretty low, so making the process as high-throughput as possible is optimal. The go-to method for this process is to use cells, which are relatively easy to genetically manipulate and fit nicely in the wells of a microplate. And, indeed, you can find compounds that hit your drug target with high specificity, but those candidates are often toxic because the drivers of disease are likely drivers of other important cellular functions in the body as well. “And when you try those drugs in an animal, the animal would drop dead,” says Cagan.

The alternative is to use whole animals, which are more comprehensive models. In cancer, tumor cells are not isolated like they are in a Petri dish, but rather have a relationship with the surrounding cellular environment as well as other tissues throughout the body. So whole-animal screening allows researchers to find the effects that a compound has, not only on tumor cells themselves, but also on the cells that interact with them.

As a result, when Cagan’s lab screens for drugs in Drosophila, his team identifies a different class of drugs than they would in cell lines. The drugs they find have low specificity against their targets. Sometimes his team even finds drugs that have no specificity against the primary target but instead hit side targets that help push the diseased cell back to normalcy. Many of the cancer drug candidates that they identify hit several targets: some in the tumor, some in the local microenvironment and some in tissue elsewhere in the body that feeds back to the tumor. These types of drugs are impossible to predict and, for now, can only be found empirically through whole-animal screening.

“For example, our RET inhibitor is the world’s worst kinase inhibitor, but that’s the magic,” says Cagan. “It hits multiple targets in the right ratio that the fly can take that hit overall but the tumor has trouble dealing with it.” 

Of course, the workhorse of drug development is the mouse, but almost immediately, the idea of quick and inexpensive high-throughput drug screening begins to fall apart. First of all, mice don’t exactly fit nicely into a well on a microplate. Secondly, they are expensive to create and upkeep. Zebrafish can be grown in such a format, but their four- to six-month lifespan slows down the process of screening for most diseases– with the exception of heart development, as their hearts develop on day one. And although C. elegans have a relatively short lifespan of three days, it lacks some important signaling pathways present in humans.

On the other hand, Drosophila is able to be grown in a 96-well plate, has a relatively short lifespan of less than a month and has many of the important signaling pathways involved in human disease. And they’re cheap to create and feed. In addition, the tools to genetically modify a Drosophila are more mature than those for any other model organism. And Cagan’s lab is continuing to develop those tools and can now rapidly built 10-hit flies. “I don’t know another animal that you can do that with,” says Cagan.    

In a paper published last year in the journal Nature, Cagan and colleagues demonstrated how to build a new drug with fly genomics and medicinal chemistry. Their drug hit four targets while avoiding an important anti-target. “I’m told that there’s really buzz about that paper, no pun intended. We’re trying to give them a way forward to think about pharmacology,” says Cagan.


So, because flies don’t have thyroids, we threw oncogenetic RET in what we worked with, which was the eye, and the eye started to grow these tumors," says Cagan. (Source: Ross Cagan)When Cagan first asked the UW graduate student who had made the Drosophila model of MTC to develop a drug screen, she refused. “She thought it was a dumb idea,” says Cagan. 

And she’s not alone. Tin Tin Su, an associate professor in molecular, cell and developmental biology at the University of Colorado who also uses Drosophila for drug screening, has also heard strong opinions from her colleagues. “There is a bias against using flies for drug screening from Drosophila biologists. I get this sense that if I mention it, I’m a sellout,” says Su.

Overall, the academic and pharmaceutical industries have different mindsets. Where one is geared towards uncovering mechanisms and publishing papers, the latter is focused on discovering drug candidates and getting treatments approved. So would the pharmaceutical industry—so focused on developing better drugs while spending less—welcome Drosophila screening? So far, that hasn’t been the case. 

“If you never worked with Drosophila in school, you don’t really think about it,” says Su. “And a lot of people in drug development fall into that category. There are people—really smart people—who ask me seriously if flies have brains or contain cells. When you tell them more about it, they catch on, but if you don’t think about it...”

Right now, she doesn’t believe pharmaceutical companies will adopt this technique anytime soon because it’s a little too risky for them right now. And that opens a window of opportunity for Drosophila researchers. For instance, through their screens, Su and colleagues have identified two small molecule inhibitors that might have potential to improve cancer treatments, and she founded a company called Suvica Inc. in 2010 to do validation studies on those candidates. The idea is that eventually, if everything goes according to plan, a pharmaceutical company will license those molecules.  

“We’re in the early stages, but we’re quite happy that something we found in the fly screen that can decrease the effect of radiation,” says Su. “This work is being done so we can translate it into a better therapy, and being able to do that in my lifetime would be really satisfying.”

Learning curve

Vosshall's lab screened for drugs that would reduce the food (dyed blue) intake in Drosophila larva. (Source: Leslie Vosshall)Su never worked with a real organism before her postdoc. While it took her about six weeks to learn how to work with cells in culture, it took her closer to a year before she was comfortable working with flies. Even now, when her lab begins working with a new mutant, it takes time to figure out and adjust to its preferences for conditions such as food and temperature. And that type of information isn’t written down anywhere. You can only learn by talking with other researchers and from experience.

“The technical challenges are real, and you need to know the Drosophila biology to get success from them,” says Su. “If you never worked with flies, it’s a bit daunting.” And even with significant experience working with flies, the 96-well format remains a challenge for Su’s lab. It’s still a slow process that requires many hands that she has to manage. Right now, she’s looking to automate the process as much as possible.

For Leslie Vosshall, a Howard Hughes Medical Institute investigator and head of the Laboratory of Neurogenetics and Behavior at Rockefeller University, the impetus to do drug screening in Drosophila was driven by the technology available. Her lab has been interested in screening the genes and circuits involved with feeding behavior, so when the university purchased a high-throughput instrument for screening cell lines for small inhibitors, she thought, ”We have this high-throughput screening instrument and a good chemical library, so why don’t we just dispense the animals directly?” 

To automate the process, Vosshall’s team used a Union Biometrica COPAS flow cytometer. The system is a flow cytometer with a gate and internal plumbing big enough to accommodate whole animals such as C. elegans, Drosophila, or even zebrafish embryos. This allowed her team to dispense Drosophila flies into 96-well plates quickly and accurately. “This has been one of my favorite sets of experiments that we’ve done recently, and it worked very well,” says Vosshall. 

Even with that technology, it took Vosshall longer than expected to get the assay optimized. Their goal was to screen 100,000 to 200,000 molecules. In the end, they only were able to screen just under 4,000 molecules before the postdoc in charge of the study left the lab. But they did identify a receptor that acted as a feeding modulation target, and published their results in Scientific Reports. “It’s an assay that’s ready to go, and I would love it if someone picked this up,” says Vosshall.

Vosshall knows that pharma doesn’t see Drosophila as a useful model right now, but that may change as we learn more about these flies. For instance, the leptin gene, which regulates a pathway involved in obesity in mice and was thought not to exist in Drosophila, was discovered in the flies just last year. “The more I look at the fly, the more I see them as little teeny humans,” says Vosshall. “Right now, it’s up to the academics to show the pharma industry that these are useful disease models.”

Back at his lab at Mount Sinai Hospital, Cagan is hoping to do just that. Right now, his lab is using screens to map out the flaws of approved and developed drugs and trying to flush out those problems to make good drugs great. So, he is working with pharma companies and has already found instances where developers may have left potentially better drugs on the table in the drug discovery process.

“I’ve learned a lot about the advantages of flies since we started this,” says Cagan. “I don’t think I fully appreciated them as a model for disease before, but I do now.”