Mixed Media Messaging: Different Cell Culture Conditions Influence Drug Activity in Cancer Cells

Photo of Kyle Flickinger working in the lab.

Choosing the ideal cell culture medium makes a difference in the life or death of a cell line. This growth environment mimics the conditions that cells experience in a living organism, providing the nutrients that the cells need to thrive.

Biochemistry assistant professor and Morgridge Investigator Jason Cantor and colleagues are exploring how changes in the cell culture environment might reveal a better understanding of the biological mechanisms that lead to disease — and the treatments that can stop or prevent them.

In a new study published in Science Advances, the researchers used high-throughput chemical screening to demonstrate how different cell culture medium conditions impact the effectiveness of drugs.

“We’re trying to address the modeling capacity of the nutrient conditions that cells might see in the body,” says Cantor. “If you can understand a mechanism and the underlying causes of things, it is probably the most powerful way to inform how you approach what’s going on downstream.”

Photo of Jason Cantor
Assistant professor Jason Cantor.

The Cantor Lab uses Human Plasma-Like Medium (HPLM), a cell culture medium systematically designed to more closely simulate the nutrient environment in human blood, which Cantor developed as a postdoctoral fellow. In 2021, Thermo Fisher Scientific made HPLM commercially available to scientists worldwide.

In this study, with collaborators at the NIH National Center for Advancing Translational Sciences (NCATS), the team performed high-throughput chemical screens on leukemia cell lines grown in HPLM and two additional conditions that instead relied on RPMI, a cell culture medium commonly used to culture human blood cells.

They screened the cell lines against a comprehensive library of nearly 2,000 compounds across several drug classes, including antivirals, antibiotics, antihistamines, and others indicated for diseases like cancer, Alzheimer’s or diabetes.

“It uses drugs that are already approved, drugs in clinical trials, drugs that are investigational that haven’t even entered the pipeline yet — so it really spanned the gamut of possible drugs out there,” says Kyle Flickinger, a graduate student in the UW–Madison Integrated Program in Biochemistry and co-first author of the paper.

Their analysis revealed over 100 compounds with different treatment responses based on the growth conditions. Some of the strongest responses were identified for drugs linked to purine metabolism, the set of pathways that make some of the building blocks for DNA and RNA. These differential responses were linked to the availability of hypoxanthine, a molecule that can also support purine metabolism, in HPLM versus RPMI.

“The cells salvage the poison, basically, and they can die off,” says Flickinger. “But in conditions where you have hypoxanthine present, it competes with that drug for the ability of the cells to take it up, and so it protects them.”

This suggests that the available nutrients in the environment can either protect cells from drug action or make them more susceptible.

The researchers also identified a unique mechanism with brivudine, a drug approved as an antiviral in some countries. When this compound was added to leukemia cell lines cultured in HPLM, the cancer cells failed to grow.

“We basically see that it has anticancer activity, in our hands, that was virtually masked in conventional media, and we didn’t know why,” says Cantor. “This is where we can say based on what we know, we can try to identify an underlying drug-nutrient interaction, and maybe it will point us in the direction of where we need to look to understand the mechanism.”

Going back to the fundamentals, they used an integrative approach — including gene knockouts, CRISPR screening, and several distinct methods in mass spectrometry — to understand the mechanism responsible for this drug-nutrient interaction. They traced the conditional effect of brivudine to folic acid, an essential vitamin. Ultimately, the researchers found that brivudine impairs cell growth in low folic acid conditions by targeting two enzymes involved in folate metabolism.

Folic acid is found at comparably low levels in human and mouse blood. Building upon their data from in vitro experiments, they worked with Rebecca Richards and her lab at the Wisconsin Blood Cancer Research Institute to test the efficacy of brivudine treatment in vivo using mice with leukemia transplants. The treatment had an effect after 1 week, showing a 50% reduction in signal markers that track tumor growth.

Cantor says this collaborative research was integral to demonstrating that their workflow could offer a proof of concept for downstream applications.

“You get these two experimental examples: you have the confirmatory, that something is on target with what you already know, and then you have the discovery of something unexpected or non-canonical,” he adds.

This comparative approach — looking at conditional effects on cell metabolism, genetic dependencies, and drug sensitivity — opens the door for drug discovery, as well as finding ways to expand the therapeutic window of existing drugs, essentially improving the outcomes.

“To approve one drug, there are thousands of compounds that go through an initial screening phase, and that’s just time and money,” says Flickinger. “If we can improve that in vitro condition even just 1-2%, that’s still time, money, and lives that are saved.”

Different models in research will always have a balance of pros and cons, since it is not yet possible to fully capture the complexity of what is happening inside a human body. The Cantor Lab continues to pursue the unknown with future research expanding upon modification of cell culture conditions, as well as exploring other cell types.

“There are always ways we can improve our models and perhaps make them more complex,” Cantor says. “Considerations into integrating other aspects of the environment are always ripe for picking.”

Written by Mariel Mohns, Morgridge Institute for Research.