Acod1 expression in cancer cells promotes immune evasion through the generation of inhibitory peptides

Abstract

Targeting programmed cell death protein 1 (PD-1) is an important component of many immune checkpoint blockade (ICB) therapeutic approaches. However, ICB is not an efficacious strategy in a variety of cancer types, in part due to immunosuppressive metabolites in the tumor microenvironment. Here, we find that αPD-1-resistant cancer cells produce abundant itaconate (ITA) due to enhanced levels of aconitate decarboxylase (Acod1). Acod1 has an important role in the resistance to αPD-1, as decreasing Acod1 levels in αPD-1-resistant cancer cells can sensitize tumors to αPD-1 therapy. Mechanistically, cancer cells with high Acod1 inhibit the proliferation of naive CD8⁺ T cells through the secretion of inhibitory factors. Surprisingly, inhibition of CD8⁺ T cell proliferation is not dependent on the secretion of ITA but is instead a consequence of the release of small inhibitory peptides. Our study suggests that strategies to counter the activity of Acod1 in cancer cells may sensitize tumors to ICB therapy.

Keywords: Acod1; CP: Cancer; CP: Immunology; TCA cycle; immune evasion; itaconate.

Figure 1.. Itaconate (ITA) is produced at high levels in ICB-resistant prostate cancer cells

PRs or PRp cells were grown in the indicated conditions for 24 h. (A) Representative confocal images of cells stained with MitoTracker Red (red) and Hoechst 33342 (blue). Scale bars, 20 μm. (B and C) Flow cytometry of cells stained with MitoTracker Red. Representative histograms (B) and mean fluorescence intensity (MFI) expressed relative to the corresponding attached (Att) culture condition (C). n = 3 independent biological samples. (D) Volcano plot for intracellular metabolites in PRs and PRp cells after 24 h of detachment (Det). p < 0.05 and fold change (FC) > 1.2 used for cutoffs. (E) Fractional enrichment of labeled ITA (left) and total pool (right) from [U-¹³C]glucose tracing. (F) Fractional enrichment of labeled ITA (left) and total pool (right) from [U-¹³C]glutamine tracing. Labeling done for 4 h in detached cells. Graphs represent data collected from a minimum of 3 biological replicates. p values are calculated by two-tailed Student’s t test. Data are mean ± SEM.

Figure 2.. ACOD1 levels are elevated in ICB-resistant cells

(A) Schematic of ITA generation by cis-aconitate decarboxylase (Acod1). (B and C) PRs or PRp cells were grown in the indicated conditions for 24 h. (B) Lysates were collected and immunoblotted as noted. (C) Gene expression of Acod1 by quantitative real-time PCR calculated as FC relative to Att PRs. (D) PRp cells were grown in the indicated conditions in the presence of DMSO or carbonyl cyanide m-chlorophenyl hydrazone (CCCP) (20 μM) for 48 h. Lysates were collected and immunoblotted as noted. (E) PRs or PRp cells were grown in the indicated conditions in the presence of DMSO or the nuclear factor κB (NF-κB) inhibitor BAY1170-82 (2.5 μM) for 9 h. Lysates were collected and immunoblotted as noted. Graphs represent data collected from a minimum of 3 biological replicates, and all western blotting experiments were independently repeated a minimum of three times with similar results. p values are calculated by one-way ANOVA followed by Tukey’s test in (C). Data are mean ± SEM.

Figure 3.. ACOD1 dictates sensitivity to αPD-1 ICB in vivo

(A) Experimental design of in vivo experiment. (B) Western blot for Acod1 of cells injected for tumor experiments. (C) PRp Scr or PRp Acod1 KO cells were subcutaneously injected in mice and received either isotype control (IgG) or αPD-1 monoclonal antibody (mAb) when tumors reached ≥50 mm³. Data are mean ± SEM (n = 6–10 per group), and p values were calculated by two-way ANOVA. (D and E) Percentage of CD4⁺ (D) or CD8⁺ (E) T cells within each tumor. Data represent the mean ± SEM (n = 4 mice/group), and p values were calculated by two-tailed Student’s t test. n.s., not significant. (F) Representative images of immunofluorescence (IF) staining of GzmB⁺ (red) and TUNEL⁺ (green) cells in tumors from the indicated conditions described in (A). Scale bars of low and high magnification represent 200 and 50 μm, respectively. DAPI, 4′,6-diamidino-2-phenylindole. (G) Quantification of relative GzmB intensity (top) or average number of TUNEL⁺ cells (bottom) with SD from six random fields of view (FOVs). The p values were calculated by one-way ANOVA. n.s., not significant.

Figure 4.. Secreted factors from ECM-Det ACOD1-high cells restrict T cell activation and proliferation

(A) Percentage of EdU⁺ CD8⁺ T cells following 48 h of activation with αCD3/CD28 in indicated conditioned medium (CM). (B) Representative histograms of violet proliferation dye 450 (VPD450) dilution in CD8⁺ T cells following 72 h of activation with αCD3/CD28 in indicated CM. (C) Histograms of CD44 (left) and CD25 (right) expression in CD8⁺ T cells following 72 h of activation with αCD3/CD28 in indicated CM (n = 3). (D) Percentage of CD44⁺ (left) and CD25⁺ (right) CD8⁺ T cells following 72 h of activation with αCD3/CD28 in indicated CM. (E) Cells grown for 24 h in Det conditions. Lysates were collected and immunoblotted as noted. (F) Percentage of EdU⁺ CD8⁺ T cells as described in (A). (G) Histograms of CD44 (left) and CD25 (right) expression as described in (C). Scr, PRp Scr CM; KO, PRp Acod1 KO CM (n = 3). (H) Percentage of CD44⁺ (left) and CD25⁺ (right) CD8⁺ T cells as described in (D). Data from EdU experiments (A and F) represent the means ± SEM of triplicate wells, and p values were calculated by one-way ANOVA analysis. Data are representative of three independent experiments. The p values for the T cell activation experiments (D and H) were calculated using a paired, two-tailed t test (n = 4 mice/group). n.s., not significant. Western blotting and VPD450 dilution experiments were independently repeated a minimum of three times with similar results.

Figure 5.. ACODI-mediated effect on CD8⁺ T cells is independent of extracellular ITA

(A) Percentage of EdU+ CD8⁺ T cells following 48 h of activation with αCD3/CD28 in the presence of the indicated concentrations of ITA. (B) Histograms of CD44 (left) and CD25 (right) expression in CD8⁺ T cells following 72 h of activation with αCD3/CD28 in the presence of the indicated concentrations of ITA. (C) Bar graphs showing the percentage of CD44⁺ (top) or CD25⁺ (bottom) CD8⁺ T cells activated as described in (B). (D) IFN-γ production by CD8⁺ T cells activated as in (B). Representative histograms of IFN-γ expression in CD8⁺ T cells (left) and relative MFI of IFN-γ in IFN-γ⁺ CD8⁺ T cells (right). (E) Concentration of ITA in the CM from PRs or PRp cells. Data in EdU experiment (A) represent the means ± SEM of four replicate wells. Data are representative of three independent experiments. Data in (C) and (D) represent the means ± SEM (n = 4 mice/group). One-way ANOVA analysis. Data in (E) are analyzed by two-tailed Student’s t test. n.s., not significant.

Figure 6.. ACOD1 regulates the secretion of immunomodulatory peptides

(A) VPD450 dilution in CD8⁺ T cells following 72 h of activation with αCD3/C28 in indicated CM that was either not boiled (top) or boiled (bottom) prior to activating the T cells. (B) Percentage of EdU⁺ CD8⁺ T cells following 48 h of activation with αCD3/C28 in full PRp CM or <3 kDa fraction of PRp CM. (C and D) Percentage of EdU⁺ CD8⁺ T cells in indicated conditions following activation as noted in (B). (E) Bar graph showing the relative abundance of a 518 Da molecular weight compound in the listed CM. (F and G) Percentage of EdU⁺CD8⁺T cells following 48 h of activation in the presence of either NGTID peptide (F) or ASNDL peptide (G). Peptides were used at 0.1, 10, or 1,000 nM. Data in (A) represent the mean ± SEM (n = 4 mice/group). Data in EdU experiments (B–D, F, and G) represent the means ± SEM of four replicate wells. Data are representative of three independent experiments. p values were calculated by two-tailed Student’s t tests (A) or one-way ANOVA analysis (B–G). n.s., not significant.