https://orcid.org/0000-0001-8357-5974
Glioblastoma stem cells (GSCs) are a tumor cell subpopulation underlying the properties of tumor recurrence and resistance to therapy in glioblastoma (GBM). Renoult et al. recently used molecular profiles of GSCs to identify genes critical to the maintenance of GSC viability and resistance to therapy. They found that these critical characteristics depend in part on an enzyme involved in glycometabolism and anaplerosis, pyruvate carboxylase (PC),1 which catalyzes the conversion of pyruvate to oxaloacetate (OAA), a pivotal intermediate in the tricarboxylic acid cycle. This work raises the possibility of PC as a therapeutic target for GBM but has the following potential caveats.
First, although this study demonstrated that PC is preferentially expressed in GSCs, the expression patterns and cellular distribution of PC at mRNA and protein levels have not been systematically investigated in GBM and normal brain. We, therefore, retrieved immunohistochemistry staining data from the Human Protein Atlas, and found that PC is highly expressed in GBM cells, and localized within the cytoplasm and cell membrane (Figure 1A, left). However, PC protein is also high in normal brain cells, especially in neurons and glial cells (Figure 1A, right). Furthermore, analysis of single-cell RNA sequencing data revealed that cells in the tumor microenvironment (TME) all exhibit some level of PC expression and no specific enrichment of PC expression in GSCs was found (Figure 1B and 1C). Given the importance of PC in metabolism and its abundance in the normal brain, inhibiting PC may cause neurotoxic side effects. Therefore, the feasibility of PC-targeted therapy for GBM patients requires further examination.
Expression characteristics of PC in GBM and a summary of features of several PC inhibitors. (A) Representative image of IHC staining of PC from the Human Protein Atlas in GBM and normal brain tissue samples (https://www.proteinatlas.org/). (B) Annotated TME cell clusters and the expression distribution of PC visualized in a uniform manifold approximation and projection (UMAP) representation derived from single-cell RNA-seq data GSE131928. (C) Bar chart analyses of single-cell RNA-seq of GBM showing no significant difference in the overlap of stem markers (SOX2 or NESTIN) and PC-positive cells compared to PC-negative cells. (D) Summary of several features of PC-specific inhibitors. GBM, glioblastoma; IHC, immunohistochemistry; PC, pyruvate carboxylase; TME, tumor microenvironment.
Second, although the authors investigated GSC viability under nutrient-restrictive conditions by limiting glutamine (Gln), another intermediate used in the synthesis of OAA, various sources of Gln exist in the TME. Tardito et al. discovered that astrocytes under Gln restriction provide an alternative source of Gln in the TME for GBM cells.2 Thus, introducing co-culture models mimicking the TME, such as transwell assays with astrocytes and GSCs in different chambers, will further illuminate the contribution of PC to GSC viability during nutrient restriction.
Third, the PC inhibitor 3-MCPD, a pyruvate analog the authors used to suppress PC activity in GSCs, is not a specific PC inhibitor. 3-MCPD, a well-known food-borne contaminant produced during food processing, also directly binds and inhibits other enzymes vital for cell viability, including glyceraldehyde-3-phosphate dehydrogenase (GAPDH)3 and glutathione reductase.4 Moreover, 3-MCPD has not been assessed for its blood–brain barrier (BBB) permeability, a property critical for clinical use. Other PC inhibitors are available (Figure 1D), such as IN-5, a molecule developed by Nicholas et al.,5 which has been validated to specifically inhibit PC. More such PC-specific inhibitors could be further tested for BBB permeability with in vitro BBB chips or in vivo animal models and ultimately delivered in target-specific nanoparticles.
Finally, the U251 cell line in these experiments is a weak GSC model. The genome of U251 does not harbor hallmark genetic features of primary GBM, such as EGFR amplification, and further genetic drift has been observed after long-term culture to obtain subclones.6 An alternative is to analyze the response to inhibitors in the authors’ patient-derived primary cultures (PDCs) based on specific genetic features, such as developing PDC groups based on PC high and low expression. This will help to understand the response to PC inhibitor therapy in model systems and possibly interpret the results of future clinical trials.
Renoult et al. have employed innovative methods to identify a correlation between PC and GSCs. We believe that taking the issues discussed above into consideration will yield a more detailed exploration of the putative importance of PC in the metabolism of GBM.
Funding
This work was supported by the Shandong Excellent Young Scientists Fund Program (2022HWYQ-035), the Shandong Provincial Natural Science Foundation (ZR2021QH030), the Special Foundation for Taishan Young Scholars (tsqn202211041), the Department of Science & Technology of Shandong Province (SYS202202), and the Qilu Young Scholar Program of Shandong University, China.
Conflict of interest statement
The authors declare that there are no potential conflicts of interest.
Author contributions
M.C., R.B., and M.H. designed the project. M.C. and G.M. wrote the manuscript. M.H., W.D., and W.H. supervised the project. All authors contributed to the article and approved the submitted version.
Data availability
All data generated or analyzed during this study are included in this article.
Ethics statement
All research procedures were approved by the institutional research ethics committee of Shandong University.
Consent for publication
All authors have agreed to publish this manuscript.
References
Author notes
Co-first author.
