In 1957, not long after James Watson and Francis Crick discovered the structure of DNA, Charles Heidelberger, a cancer researcher at the University of Southern California, designed and synthesized a molecule to break it (1). Similar in structure to the RNA base uracil, 5-fluorouracil (5-FU) hijacks DNA synthesis machinery, which, in cancer cells, stops them from replicating uncontrollably (2). For decades, most scientists focused on the drug’s DNA-damaging activity, but results from a new study in Cell Medicine Reports suggest that DNA may not be 5-FU’s main cancer-killing target, potentially changing the way physicians use the drug in the future (3).

Heidelberger and his colleague first designed 5-FU after noticing that cancer cells used uracil as a nucleic acid building block more than normal cells (4). 5-FU and uracil look identical except that one of the 5-FU carbons has a fluorine instead of a hydrogen atom attached to it. Early clinical studies showed that the drug had modest efficacy towards several cancer types including breast, colon, and liver (5). But when used with other compounds, its antitumor activity seemed to improve (6). As a result, doctors frequently administer 5-FU as a cocktail with other DNA-damaging drugs. However, not all patients benefit from these combinations.

“Believe it or not, most of those cocktails have come about empirically,” said study coauthor Michael Yaffe, a physician and cancer researcher at the Massachusetts Institute of Technology. So long as the two drugs combined are not toxic to the patients, they are generally considered to be compatible, he said.

Yaffe and his team were interested in understanding how to rationally combine 5-FU with other anticancer drugs to enhance treatment efficacy. To do this, he needed to know how 5-FU induces cancer cell death.

On its own, 5-FU is inactive. The drug breaks down into 5-fluorouridine (5-FUR) and 5-fluorodeoxyuridine (5-FdUR) inside the body (7). These get further metabolized into compounds that incorporate into RNA and DNA respectively.

Scientists often attribute 5-FU’s anticancer activity to its DNA-acting metabolites, namely 5-fluorodeoxyuridine monophosphate (5-FdUMP) and 5-fluorodeoxyuridine triphosphate (5-FdUTP), largely ignoring the RNA-acting metabolites.

Inside cancer cells, 5-FdUMP inhibits thymidylate synthase, an enzyme that is crucial for DNA replication and repair. Whereas 5-FdUTP interacts with DNA polymerase and misincorporates into new strands of DNA (7).

Yaffe and his colleagues hypothesized that if 5-FU primarily acts by damaging DNA, then adding other DNA-damaging drugs such as oxaliplatin or irinotecan should lead to more cell death.

But when he and his team compared the efficacy of 5-FU alone and in combination with irinotecan or oxaliplatin against 11 colorectal cancer cell lines, they found that treating the cells with two drugs only resulted in sub-additive cell death.

“Essentially in almost all cases, two plus two was less than four,” Yaffe said. In other words, there is no evidence that the drugs acted synergistically.

Even more surprising, in certain colorectal cancer cell lines, the combination yielded a worse outcome than giving the drug alone.

Essentially in almost all cases, two plus two was less than four.
- Michael Yaffe, Massachusetts Institute of Technology

“That was very perplexing. It didn't make any sense with what we understood should be the mechanisms for how these drugs worked,” Yaffe said.

The team searched through publicly available clinical data and found that the lack of synergy was also apparent in patients who have taken the drugs either in combination or individually. This result led Yaffe to suspect that 5-FU might act via a different pathway.

To investigate its mechanism of action, Yaffe and his team followed the fate of radioactively labeled ¹⁴C-5-FU inside colorectal cancer cells. First, they observed that the drug accumulated more in RNA than DNA. Then, when they knocked down the activity of thymidylate synthase, the cells became more sensitive to 5-FU treatment, which confirmed that 5-FU does, in fact, work by disrupting DNA synthesis. But there was a twist. When the scientists added a ribosomal RNA inhibitor, the effects of the drug were completely reversed. This suggested that the 5-FU’s efficacy depended more on its interaction with RNA than DNA. Moreover, adding oxaliplatin reduced the ¹⁴C-5-FU accumulation in RNA, which may explain the drugs’ lack of synergy in patients.

For the first time, Yaffe and his team showed that 5-FU can induce cancer cell death by acting primarily on ribosomal RNA. It is a “daring [pursuit] to go against the long-held belief that thymidylate synthase inhibition and DNA incorporation are the primary mechanisms of 5-FU cytotoxicity,” Henrik Pettersen, a molecular biologist at the Norwegian University of Science and Technology who was not associated with the study, wrote in an email. More than ten years ago, Pettersen and his group showed that 5-fluorouracil mainly incorporates into RNA, but their study did not identify ribosomal RNA as the main target (8). “This is thankfully and excellently addressed in this new Cell paper,” he added. 

According to John Knight, a cancer biologist at the University of Manchester who was not involved in the study, RNA has historically been much more difficult to work with than other macromolecules due to its unstable molecular structure. But the tools available to study RNA biology have improved and enabled more research over the years. “It's not new that [5-FU] damages RNA,” Knight said. “But it is new that that is an important part of the mechanism of the drug. That’s the step [Yaffe and his team] have taken here.”

Knight’s group recently showed that 5-FU treatment triggers a stress response in colorectal cancer cells that stalls ribosomes, leading to cell death (9). The next big question is whether this phenomenon also happens in patients, he said.

5-FU’s RNA-damaging effect was not shared across all tumors. When Yaffe and his team compared the dose-response data of 60 cancer cell lines from a National Cancer Institute database, they found that tumors that were historically responsive to 5-FU, like colorectal cancer, were almost always more sensitive to its RNA-damaging activity. But those cancers that did not respond well to 5-FU showed mixed sensitivity to both its RNA and DNA-damaging effects.  

“Now that we have a different understanding of how fluorouracil works, it does change how we think about rational efforts to make better drug combinations in the future,” said Adam Palmer, a systems pharmacologist at the University of North Carolina at Chapel Hill who collaborated with Yaffe on this project.

Yaffe and his team hope to collaborate with other clinicians to find the best ways to use 5-FU in the clinic. “There is still a lot left to learn about this drug,” he said. “[This study] really suggests that this old drug, which we have known forever, could probably be repurposed, or at least be used in a more intelligent manner to treat different types of tumors.”

References

1. Duschinsky R. et al. The synthesis of 5-fluoropyrimidines. J Am Chem Soc  79, 4559 (1957).
2. Heidelberger, C. et al. Fluorinated Pyrimidines, A New Class of Tumour-Inhibitory Compounds. Nature   179, 663–666 (1957).
3. Chen, J-K. et al. An RNA damage response network mediates the lethality of 5-FU in colorectal cancer. Cell Rep Med  5, 101778 (2024).
4. Heidelberger, C. et al. The Comparative Utilization of Uracil-2-C¹⁴ by Liver, Intestinal Mucosa, and Flexner-Jobling Carcinoma in the Rat. Cancer Res  17, 399–404 (1957).
5. Curreri A.R. et al. Clinical studies with 5-fluorouracil. Cancer Res  18, 478–484 (1958).
6. Peters, G.J. & van Groeninge, C.J. Clinical relevance of biochemical modulation of 5-fluorouracil Ann Oncol  2, 469-480 (1991).
7. Chalabi-Dchar, M. et al. A novel view on an old drug, 5-fluorouracil: an unexpected RNA modifier with intriguing impact on cancer cell fate NAR Cancer  3, zcab032 (2021).
8. Pettersen, H.S. et al. UNG-initiated base excision repair is the major repair route for 5-fluorouracil in DNA, but 5-fluorouracil cytotoxicity depends mainly on RNA incorporation Nucleic Acids Res  39, 8430-8444 (2011).
9. Chatterjee, S. et al. Ribosome Quality Control mitigates the cytotoxicity of ribosome collisions induced by 5-Fluorouracil, Nucleic Acids Res  52, 12534-12548 (2024).