27 mars 2025
Buj et al. (BioRxiv)
Keywords
CDKN2A alterations
Macrophages
Immune checkpoint blockade therapy
Main Findings
In this preprint, the authors investigate whether varying CDKN2A expression levels in cancer cells modulates tumor microenvironment (TME) immune subsets and immune checkpoint blockade (ICB) therapy response. CDKN2A encodes for 2 proteins, p16INK4A and p14ARF, both of which have well-documented roles in cell cycle regulation and cellular senescence. In the context of cancer, CDKN2A loss of function mutations are common across numerous cancer types and have been associated with ICB resistance. More recently, p16INK4A loss has also been shown to alter cancer cell metabolism, enhancing cholesterol and nucleotide synthesis. However, the impact of these CDKN2A loss-driven intracellular changes on downstream TME metabolite composition as well as its potential cascading effects on immune cell response and ICB resistance remains unexplored. Using isogenic YUMM lines with low and high expression of Cdkn2a (denoted as Cdkn2alowand Cdkn2ahigh respectively), the authors demonstrate Cdkn2alowtumours are resistant to anti-PD1 and anti-PD-L1 therapy, recapitulating trends observed in human melanoma and other cancer types. Mass spectrometry on tumour interstitial fluid reveals significantly lower zinc levels in tumours that are Cdkn2alowcompared to Cdkn2ahigh. Correspondingly, results from an in vivo CRISPR screen focused on negatively enriched plasma membrane solute carrier transporters in Cdkn2alow and Cdkn2ahightumours treated with or without anti-PD1 blockade, identify the zinc importer SLC39A9 as a top hit for mediating anti-PD1 resistance. The authors show that p16INK4Aloss enhances plasma membrane expression of SLC39A9 in vitro, increasing intracellular zinc and conversely, decreasing extracellular zinc. Critically, Slc39a9knockdown sensitizes Cdkn2alow tumours to anti-PD1 blockade in vivo. Application of high concentration zinc diet enhances zinc levels in the TME, also sensitizing Cdkn2alow tumours to ICB, similar to Slc39a9knockdown results. These findings suggest that zinc repletion within the TME restores anti-PD1 sensitivity in Cdkn2alow tumours. Using Rag1-/- mice or pharmacologic inhibition of CSF1R signalling, the authors show that macrophages, but not T and B cells, mediate the zinc-dependent sensitivity to anti-PD1 blockade. Analysis of a human melanoma single cell dataset highlight that zinc and phagocytosis gene signatures are lower in macrophages from CDKN2Alow tumours compared to CDKN2Ahigh tumours. Furthermore, these zinc and phagocytosis signatures were found to positively correlate with each other. The authors then experimentally validate that macrophages with higher phagocytic capabilities also have higher zinc levels in vitro and in vivo. Additionally, high zinc diet enhances tumour macrophage zinc levels and promotes phagocytosis upon anti-PD1 blockade. The authors conclude that CDKN2Alow cancer cells deprive macrophages of environmental zinc, impairing their phagocytic capacity while promoting resistance to ICB therapy.
Limitations
Inhibition of CSF1R signalling, while a popular method of depleting macrophages, has been found to result in enhanced recruitment of pro-tumorigenic polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs) in certain tumour models (https://doi.org/10.1016/j.ccell.2017.10.005). Flow cytometric analysis to immunophenotype tumours post-Pexidartinib treatment to assess cell numbers of various immune subsets such as MDSCs (using markers Ly6G and Ly6C) may be helpful to verify if the resistance to anti-PD1 blockade observed is solely due to the absence of macrophages or if the microenvironment has been remodelled with other cell types that further contribute to resistance.
While the authors highlight a relationship between zinc levels, macrophage phagocytosis, and ICB sensitivity in their system, it remains elusive if enhanced macrophage phagocytosis is the sole driver of anti-tumour immunity in their model or part of an arsenal of other macrophage-directed mechanisms. For example, other mechanisms such as increased macrophage release of nitric oxide through enhanced iNOS expression, could also play roles for mediating anti-tumour immunity. A better characterization of alternative anti-tumour immune mechanisms would be helpful to provide a more mechanistic insight on the actions of zinc on macrophages.
A recent study has shown that obesity can result in increased PD1 expression on macrophages and thus provides rationale to explain why obese patients are better responders to anti-PD1 blockade (https://doi.org/10.1038/s41586-024-07529-3). In a similar vein, it would be interesting to see whether PD1 expression changes on macrophages in tumour-bearing mice on a high zinc diet. For example, is the ICB sensitivity-inducing effect of zinc supplementation acting by increasing PD1 expression on macrophages, or are alternate pathways at play that ameliorate the inhibitory effects of PD1 signalling (perhaps both)?
The authors establish that macrophages with greater intracellular zinc staining intensity are associated with enhanced phagocytosis in vitro and in vivo. However, it would be nice to show in an isolated system if cancer cells directly influence macrophage phagocytosis in a manner that depends on cancer cell zinc uptake. To support this, a co-culture experiment with macrophages and Cdkn2alow YUMM lines with and without Slc39a9 knockdown that uses macrophage phagocytosis as a readout (e.g. with pHrodo), may be helpful to further support the authors’ claims. An even greater benefit (despite an added layer of difficulty) would be the inclusion of different combinations of anti-PD1 and recombinant mouse PDL1 to this co-culture system.
Significance/Novelty
The authors demonstrate a critical link between low CDKN2A expression in cancer cells, availability of microenvironmental zinc, and macrophage phenotypes in mediating ICB resistance. While there have been numerous lines of evidence associating CDKN2A loss with intracellular metabolic changes in cancer cells, the consequences of these changes on the tumour immune microenvironment and subsequent impact on ICB response has not been previously investigated. Thus, this study by Buj et al., which intersects cancer genetics, metabolism, immunity, and therapeutic outcome, is very novel and an excellent advancement in our understanding of tumour biology.
Credit
Reviewed by Robbie Jin as part of a cross-institutional journal club between the Icahn School of Medicine at Mount Sinai, the University of Oxford, the Karolinska Institute and the University of Toronto.
The author declares no conflict of interests in relation to their involvement in the review.