Identifying new targets for cancer drug 5′-fluorouracil

This article refers to:

Chromosome segregation and organization are targets of 5′-Fluorouracil in eukaryotic cells

Improved targeting of drugs is essential in personalizing the fight against cancer. Historically, drug repurposing or retargeting has arisen through serendipity or fortuitous clinical observations. In a study published in this issue of Cell Cycle,¹ Mojardín and colleagues have used a novel systematic approach to identify new targets of 5′-Fluorouracil (5-FU).

5-FU, a uracil analog, was first developed in 1957 as an anticancer agent on the basis that radioactive uracil was found to be absorbed by liver cancers more readily than normal cells.² Synthesized by reacting fluorine with uracil, 5-FU has been used as part of the clinician's armament in the treatment of a range of cancer types including colorectal, anal, head and neck, esophageal, and gastric cancers, either alone or in combination with other drugs and ionizing radiation, for nearly 60 years.

5-FU, a member of the antimetabolite drug family, is thought to work by irreversibly blocking the activity of thymidylate synthase, which methylates deoxyuridine monophosphate (dUMP) to form thymidine monophosphate (dTMP). Subsequent dTMP depletion causes bacteria and cancerous cells to die a ‘thymineless’ death.³ While the precise mechanism of such death is unknown, this pyrimidine analog inhibits DNA synthesis, and can also be incorporated into RNA, thus promoting apoptosis. However, perhaps because of its lack of specificity this drug has earned the epithet ‘5-feet under’ among its detractors, reflecting the urgent need for better drug targeting. So can the mechanism of 5-FU action be better understood and thus used to target cancer cells more effectively? Enter the fission yeast FU fighters.

In a new study, Laura Mojardín and colleagues (see Cell Cycle this issue¹) used a systematic genome-wide approach to screen a fission yeast haploid deletion library of ~3,300 strains to identify genes which when deleted conferred sensitivity to 5-FU. The screen identified 270 sensitive deletion mutants and validated previously characterized drug targets, including genes involved in RNA metabolism. Surprisingly, this approach also revealed a significant enrichment for genes involved in chromosome organization and chromosome segregation.

A number of histone modifiers were identified within the chromosome organization group, including genes encoding components of the SAGA and NuA4 histone acetyltransferase (HAT) complexes, which promote transcription; the Rpd3 complex, which deacetylates histones during transcription; and the Set1C/COMPASS complex, which methylates histone H3 lysine 4 to promote transcriptional activation. Many heterochromatin components were also identified including members of the histone H2B ubiquitylation ligase complex (HULC) which promotes centromeric transcriptional derepression; the RNA-induced transcriptional silencing (RITS) complex; the Clr4 methyltransferase complex (CLRC) which methylates histone H3 lysine 9 thereby facilitating heterochromatin spreading; and the Stn1-Ten1 complex, which is required for telomere protection in yeast and humans. Accordingly, the authors demonstrated that 5-FU reduces the levels of histone H3K9 methylation and reduces heterochromatic transcription. These findings suggest a novel way to target chromatin defects using 5-FU, which is of clinical significance as chromatin regulatory factors and histone modifiers are frequently mutated in human cancer.⁴

This study also identified striking connections between 5-FU and chromosome segregation. 5-FU treatment induced the transcription of genes encoding components of the Mis6-Sim4 complex, which mediates loading of Cnp1/CENP-A onto centromeres; mitotic cohesins; and the NMS complex, required for attachment of spindle microtubules to the kinetochore. Further, deletion of genes encoding proteins involved in centromere cohesion (Pds5), heterochromatin formation (Swi6, the yeast homolog of human HP1) and kinetochore function (Alp14, Mhf1 and Mhf2) exhibited 5-FU sensitivity. Moreover, 5-FU treatment significantly increased the rate of loss of a non-essential minichromosome, consistent with 5-FU targeting chromosome segregation.

This study also provides insights into how 5-FU functions in relation to other drugs. A considerable overlap in sensitivity to both 5-FU and valproic acid (an anticancer drug targeting histone deacetylase) was revealed in the chromosome organization and RNA metabolism groups, suggesting these drugs have overlapping modes of action. Further, thiabendazole, a microtubule-destabilizing agent was found to synergistically enhance the cytotoxic effects of 5-FU.

Understanding how 5-FU sensitizes these novel targets will be of both biological and clinical interest. Indeed, as many of these genes and protein complexes identified in this screen have human counterparts, this study provides important insights into potential targets and drug combinations that could enhance and personalize the use of this old drug in cancer therapy.