Inhibiting glioblastoma stem cells by targeting pyruvate carboxylase: A novel therapeutic strategy with both opportunities and challenges

We appreciate the opportunity to stimulate a scientific discussion on our recent study by Renoult et al.,¹ revealing pyruvate carboxylase as a critical survival factor in glioblastoma stem cells, brought by the comments of Dr Han and colleagues. GBM tumors are characterized by intrinsic and dynamic heterogeneity, which complicates both fundamental experimental studies and the development of clinically effective treatments. While GBM stem cells (GSC) play a crucial role in GBM aggressiveness and relapse, specific universally informative biomarkers of GSC are still missing, owing to their genetic and molecular heterogeneity. Common specific GSC biomarkers are CD133, CD90, CD44, L1CAM, A2B5, and GPD.² Thus, to determine whether pyruvate carboxylase (PC) expression is enriched in GSC, Han et al. should extend their single-cell RNA sequencing analysis to these reliable GSC markers. We agreed that using GBM cell lines is suboptimal, even if they are easier to culture and manipulate, and share molecular and phenotypic features with patient-derived cell cultures (PDC). PDC are significantly more valuable as they provide better clinical relevance and better mimic human pathology, which is why all experiments in the Renoult et al.’s study have been performed using both the human U251-MG cell line and PDC models.

Like all metabolic enzymes, PC is expressed in normal cells, including astrocytes and neurons.³ In the brain, an essential metabolic coupling exists between astrocytes and neurons for maintaining neurotransmitter pools and regulating excitatory and inhibitory signaling in the brain. Astrocytes typically do not consume glutamine, but instead scavenge glutamate from the extracellular space and convert it to glutamine using glutamine synthetase. Thus, the extracellular pool of glutamate remains low, reducing the risk of chaotic excitatory neurotransmission.⁴ Cancer cells exhibit high metabolic flexibility, enabling them to survive in nutrient-deprived microenvironments.⁵ Since anaplerosis significantly contributes to biomass production, anaplerotic pathways are compelling therapeutic targets. Interestingly, in agreement with our study, Deberardinis and colleagues have reported that PC activity was essential for tumor growth in low-glutamine conditions, and even predicts resistance to glutamine withdrawal.⁶ The mechanism of PC induction appears to be complex since, in both studies, increased PC protein did not require increased mRNA abundance. These results highlight a general limitation of relying solely on RNA sequencing, which assumes a direct correlation between mRNA abundance and protein levels. Such limitation is even greater when considering PC enzymatic activity. Thus, studies only relying on transcriptomic analyses may potentially lead to an incomplete understanding of critical signaling and/or metabolic pathways.

Finally, we agree that any potential clinical application targeting PC definitively requires further experimental studies. In our study, both 3MCPD and the genetic silencing of PC were used as proof-of-principle approaches, and have never been considered as potential therapeutic strategies. First, we fully agree that therapeutic inhibitors need to cross the Blood-Brain barrier (BBB), despite its GBM-driven alterations. Second, inhibitors have to be uptaken by the cells and reach the mitochondria since PC is located within the mitochondrial matrix. Third, PC inhibition is not sufficient, as shown in our study, but needs to be combined with other treatments. Recently, some substituted imidazolidinetriones have been identified as potential PC inhibitors through an in silico screening approach.⁷ However, further experiments are required to demonstrate their ability to cross the BBB, reach the mitochondrial matrix, and directly inhibit PC activity using ¹³C tracing approaches.

Conflict of interest statement

The authors declare no conflict of interest.

References

Author notes