Persister cells in cancerous tumours are a major contributor to treatment resistance. Unfortunately, these rare cells are not well understood. A new study, published in Nature, has developed a new method to characterise these persister cells. Their findings could lead to improved therapeutic approaches.
What are persister cells?
Some cancer cells are able to survive treatment by entering a reversible, drug-tolerant ‘persister’ state. Most of these persister cells stop dividing in the presence of drugs, but a rare sub-set of them are able to re-enter the cell cycle. These are known as cycling persister cells. Once they are back in the cell cycle, they are able to proliferate, resulting in high numbers of resistant cancer cells.
Until recently, research on the resistance of persister cells to treatment has been focused on genetic mechanisms. However, emerging data suggests that non-genetic mechanisms may also play a role.
EGFR mutated cancer cells
In this study, researchers analysed cell divisions in vitro in human lung cancer cells that have mutations in the gene encoding the epidermal growth factor receptor (EGFR).
The cancer cells were treated with osimertinib, which is an EGFR inhibitor, and the number and timing of division events were then quantified. It was found that only 8% of cell lineages gave rise to persister cells. This was defined as cells that were alive at day 14 of drug treatment. Of these persisters, 13% re-entered the cell cycle and gave rise to multi-cellular persister colonies. These findings show that although they are rare, both cycling and non-cycling persistent cells can emerge early in the course of drug treatment.
Identifying cell properties with Watermelon
In order to identify mechanisms that allow cancer cells to develop into cycling persisters, the cellular and molecular properties of the cells must be measured before and during drug treatment. To achieve this, the researchers developed a DNA barcoding system called Watermelon. Watermelon was able to simultaneously trace the lineage, transcriptional state and proliferative state of each cell.
The team used the Watermelon system to test whether cycling persister cells were intrinsically less sensitive to osimertinib. Both cycling and non-cycling persister cells were re-exposed to osimertinib after a break in treatment where they were found to have reacquired drug sensitivity. This suggests that a non-genetic, reversible mechanism underlies the ability to cycle under continuous drug treatment.
The role of antioxidants
Single-cell RNA sequencing (scRNA-seq) was also used to profile the gene expression of cancer cells during 14 days of osimertinib treatment. The profiles indicated that cycling and non-cycling persister cells followed distinct transcriptional trajectories and arose from different cell lineages. In addition, the expression of genes upregulated by cycling persisters were not found to be upregulated before drug treatment. This suggests that cycling persisters arise from cells poised to induce these programs, rather than cells that already express them.
Cycling persisters exhibited greater expression of antioxidant gene programs, including NRF2, a transcription factor induced in response to oxidative stress. NRF2 gene targets were also correlated with larger persister clone sizes. These results support a role for antioxidant defence programs in cycling persisters.
To test the role of antioxidants further, the production of reactive oxygen species (ROS) in drug treated cells was measured. It was found that osimertinib induced ROS production, which caused oxidative stress in the cancer cells. However, by day 14 of the treatment, cycling persister cells had significantly lower levels of ROS than non-cycling persisters. In addition, when the levels of ROS were alleviated, the percentage of cycling persisters in the population increased. These results suggest that redox balance plays a role in the regulation of cycling persisters.
Metabolism shifts in persister cells
Redox balance is tightly linked to metabolism. Therefore, the metabolic profiles between persister subpopulations were compared. In total, 56 metabolites were found to be significantly different in abundance between cycling and non-cycling persisters. In particular, the level of carnitine-linked fatty acids, which are substrates of mitochondrial β-oxidation, was higher in cycling persisters. When fatty acid oxidation was increased, the percentage of cycling persisters increased and vice versa. These results support the theory that fatty acid oxidation contributes to the cycling persister phenotype.
Next, the researchers asked whether their findings were also observed in other cancer types. Using the Watermelon system, they generated models of various human cancers, including breast and colorectal cancer. In five models, more than 50 cycling persister cells were observed. In four out of the five models, the cycling persisters showed elevated fatty acid metabolism compared to non-cyclers. The cycling persister cells also exhibited significantly lower ROS levels than non-cycling persisters, matching the results already gathered from the in vitro lung cancer cells.
In vivo investigation
To validate the in vitro results in vivo, EGFR-mutated lung cancer tumours were established in mice. These mice were then treated with osimertinib. The characteristics of the persister cells were consistent with the in vitro findings.
Finally, the team analysed scRNA-seq profiles from human EGFR-driven lung adenocarcinoma tumour samples. Once again, the findings were in line with the in vitro results.
Conclusion
The Watermelon system developed in this study allowed the researchers to identify adaptations that may facilitate the cell cycle re-entry of persister cells. Cell lineages with these features were found to be preferentially poised to proliferate under drug pressure. Their results highlight potential pathways and vulnerabilities that can be targeted in cycling persister cells.
The findings have also greatly advanced our understanding of cycling persisters and will be extremely valuable to the development of future cancer treatments.
Image credit: Background photo created by kjpargeter – www.freepik.com
