CPT1A Mediates Radiation Sensitivity in Colorectal Cancer

Reviewer #1 (Public Review):

Summary:

Fats and lipids serve many important roles in cancers, including serving as important fuels for energy metabolism in cancer cells by being oxidized in the mitochondria. The process of fatty acid oxidation is initiated by the enzyme carnitine palmitoyltransferase 1A (CPT1A), and the function and targetability of CPT1A in cancer metabolism and biology have been heavily investigated. This includes studies that have found important roles for CPT1A in colorectal cancer growth and metastasis.

In this study, Chen and colleagues use analysis of patient samples and functional interrogation in animal models to examine the role CPT1A plays in colorectal cancer (CRC). The authors find that CPT1A expression is decreased in CRC compared to paired healthy tissue and that lower expression correlates with decreased patient survival over time, suggesting that CPT1A may suppress tumor progression. To functionally interrogate this hypothesis, the authors both use CRISPR to knockout CPT1A in a CRC cell line that expresses CPT1A and overexpress CPT1A in a CRC cell line with low expression. In both systems, increased CPT1A expression decreased cell survival and DNA repair in response to radiation in culture. Further, in xenograft models, CPT1A decreased tumor growth basally and radiotherapy could further decrease tumor growth in CPT1A-expressing tumors. As CRC is often treated with radiotherapy, the authors argue this radiosensitization driven by CPT1A could explain why CPT1A expression correlates with increased patient survival.

Lastly, Chen and colleagues sought to understand why CPT1A suppresses CRC tumor growth and sensitizes the tumors to radiotherapy in culture. The antioxidant capacity of cells can increase cell survival, so the authors examine antioxidant gene expression and levels in CPT1A-expressing and non-expressing cells. CPT1A expression suppresses the expression of antioxidant metabolism genes and lowers levels of antioxidants. Antioxidant metabolism genes can be regulated by the FOXM1 transcription factor, and the authors find that CPT1A expression regulates FOXM1 levels and that antioxidant gene expression can be partially rescued in CPT1A-expressing CRC cells. This leads the authors to propose the following model: CPT1A expression downregulates FOXM1 (via some yet undescribed mechanism) which then leads to decreased antioxidant capacity in CRC cells, thus suppressing tumor progression and increasing radiosensitivity. This is an interesting model that could explain the suppression of CPT1A expression in CRC, but key tenets of the model are untested and speculative.

Strengths:

• Analysis of CPT1A in paired CRC tumors and non-tumor tissue using multiple modalities combined with analysis of independent datasets rigorously show that CPT1A is downregulated in CRC tumors at the RNA and protein level.

• The authors use paired cell line model systems where CPT1A is both knocked out and overexpressed in cell lines that endogenously express or repress CPT1A respectively. These complementary model systems increase the rigor of the study.

• The finding that a metabolic enzyme generally thought to support tumor energetics actually is a tumor suppressor in some settings is theoretically quite interesting.

Weaknesses:

• The authors propose that CPT1A expression modulates antioxidant capacity in cells by suppressing FOXM1 and that this pathway alters CRC growth and radiotherapy response. However, key aspects of this model are not tested. The authors do not show that FOXM1 contributes to the regulation of antioxidant levels in CRC cells and tumors or if FOXM1 suppression is key to the inhibition of CRC tumor growth and radiosensitization by CPT1A. Thus, the model the authors propose is speculative and not supported by the existing data.

• The authors propose two mechanisms by which CPT1A expression triggers radiosensitization: decreasing DNA repair capacity (Figure 3) and decreasing antioxidant capacity (Figure 5). However, while CPT1A expression does alter these capacities in CRC cells, neither is functionally tested to determine if altered DNA repair or antioxidant capacity (or both) are the reason why CRC cells are more sensitive to radiotherapy or are delayed in causing tumors in vivo. Thus, this aspect of the proposed model is also speculative.

• The authors find that CPT1A affects radiosensitization in cell culture and assess this in vivo. In vivo, CPT1A expression slows tumor growth even in the absence of radiotherapy, and radiotherapy only proportionally decreases tumor growth to the same extent as it does in CPT1A non-expressing CRC tumors. The authors propose from this data that CPT1A expression also sensitizes tumors to radiotherapy in vivo. However, it is unclear whether CPT1A expression causes radiosensitization in vivo or if CPT1A expression acts as an independent tumor suppressor to which radiotherapy has an additive effect. Additional experiments would be necessary to differentiate between these possibilities.

• The authors propose in Figure 3 that DNA repair capacity is inhibited in CRC cells by CPT1A expression. However, the gH2AX immunoblots performed in Figure 3H-I that measure DNA repair kinetics are not convincing that CPT1A expression impairs DNA repair kinetics. Separate blots are shown for CPT1A expressing and non-expressing cell lines, not allowing for rigorous comparison of gH2AX levels and resolution as CPT1A expression is modulated.

• There are conflicting studies (PMID: 37977042, 29995871) that suggest that CPT1A is overexpressed in CRC and contributes to tumor progression rather than acting as a tumor suppressor as the authors propose. It would be helpful for readers for the authors to discuss these studies and why there is a discrepancy between them.

Reviewer #2 (Public Review):

The manuscript by Chen et al. describes how low levels of CPT1A in colorectal cancer (CRC) confer radioresistance by expediting radiation-induced ROS clearance. The authors propose that this mechanism of ROS homeostasis is regulated through FOXM1. CPT1A is known for its role in fatty acid metabolism via beta-oxidation of long-chain fatty acids, making it important in many metabolic disorders and cancers.

Previous studies have suggested that the upregulation of CPT1A is essential for the tumor-promoting effect in colorectal cancers (CRC) (PMID: 32913185). For example, CPT1A-mediated fatty acid oxidation promotes colorectal cancer cell metastasis (PMID: 2999587), and repression of CPT1A activity renders cancer cells more susceptible to killing by cytotoxic T lymphocytes (PMID: 37722058). Additionally, inhibition of CPT1A-mediated fatty-acid oxidation (FAO) sensitizes nasopharyngeal carcinomas to radiation therapy (PMID: 29721083). While this suggests a tumor-promoting effect for CPT1A, the work by Chen et al. suggests instead a tumor-suppressive function for CPT1A in CRC, specifically that loss or low expression of CPT1A confers radioresistance in CRC. This makes the findings important given that they oppose the previously proposed tumorigenic function of CPT1A. However, the data presented in the manuscript is limited in scope and analysis.

Major Limitations:

(1) Analysis of Patient Samples

- Figure 1D shows that CPT1A levels are significantly lower in COAD and READ compared to normal tissues. It would be beneficial to show whether CPT1A levels are also significantly lower in CRC compared to other tumor types using TCGA data.
- The analysis should include a comparison of closely related CPT1 isoforms (CPT1B and CPT1C) to emphasize the specific importance of CPT1A silencing in CRC.
- Figure 2 lacks a clear description of how IHC scores were determined and the criteria used to categorize patients into CPT1A-high and CPT1A-low groups. This should be detailed in the text and figure legend.
- None of Figure 2B or 2C show how many patients were assigned to the CPT1A-low and CPT1A-high groups.

(2) Model Selection and Experimental Approaches

- The authors primarily use CPT1A knockout (KO) HCT116 cells and CPT1A overexpression (OE) SW480 cells for their experiments, which poses major limitations.
- The genetic backgrounds of the cell lines (e.g., HCT116 being microsatellite instable (MSI) and SW480 not) should be considered as they might influence treatment outcomes. This should be acknowledged as a major limitation.
- Regardless of their CPT1A expression levels, for the experiments with HCT116 and SW480 cells in Figure 3C-F, it would be useful to see whether HCT116 cells can be further sensitized to radiotherapy in overexpression and whether SW480 cells can be desensitized through CPT1A KO.
- The use of only two CRC cell lines is insufficient to draw broad conclusions. Additional CRC cell lines should be used to validate the findings and account for genetic heterogeneity. The authors should repeat key experiments with additional CRC cell lines to strengthen their conclusions.

(3) Pharmacological Inhibition

Several studies have reported beneficial outcomes of using CPT1 pharmacological inhibition to limit cancer progression (e.g., PMID: 33528867, PMID: 32198139), including its application in sensitization to radiation therapy (PMID: 30175155). Since the authors argue for the opposite case in CRC, they should show this through pharmacological means such as etomoxir and whether CPT1A inhibition phenocopies their observed genetic KO effect, which would have important implications for using this inhibitor in CRC patients.

(4) Data Representation and Statistical Analysis

- The relative mRNA expression levels across the seven cell lines (Supplementary Figure 1C) differ greatly from those reported in the DepMap (https://depmap.org/portal/). This discrepancy should be addressed.
- The statistical significance of differences in mRNA and protein levels between RT-sensitive and RT-resistant cells should be shown (Supplementary Figure 1C, D).

Conclusion

The study offers significant insights into the role of CPT1A in CRC radioresistance, proposing a tumor-suppressive function. However, the scope and depth of the analysis need to be expanded to fully validate these claims. Additional CRC cell lines, pharmacological inhibition studies, and a more detailed analysis of patient samples are essential to strengthen the conclusions.

Reviewer #3 (Public Review):

Summary:

The study aims to elucidate the role of CPT1A in developing resistance to radiotherapy in colorectal cancer (CRC). The manuscript is a collection of assays and analyses to identify the mechanism by which CPT1A leads to treatment resistance through increased expression of ROS-scavenging genes facilitated by FOXM1 and provides an argument to counter this role, leading to a reversal of treatment resistance.

Strengths:

The article is well written with sound scientific methodology and results. The assays performed are well within the scope of the hypothesis of the study and provide ample evidence for the role of CPT1A in the development of treatment resistance in colorectal cancer. While providing compelling evidence for their argument, the authors have also rightfully provided limitations of their work.

Weaknesses:

The primary weakness of the study is acknowledged by the authors at the end of the Discussion section of the manuscript. The work heavily relies on bioinformatics and in vitro work with little backing of in vivo and patient data. In terms of animal studies, it is to be noted that the model they have used is nude mice with non-orthotopic, subcutaneous xenograft, which may not be the best recreation of the patient tumor.