Leveraging autophagy and pyrimidine metabolism to target pancreatic cancer

Abstract

Autophagy inhibitors are promising compounds to treat pancreatic ductal adenocarcinoma (PDA) but their efficacy in patients is unclear, highlighting a need to understand mechanisms of resistance. We used a novel approach to uncover metabolic adaptations that bypass autophagy inhibition. Utilizing PDA cells with acquired resistance to different autophagy inhibitors, we found that severe autophagy depletion induces metabolic rewiring to sustain TCA intermediates and nucleotides for biosynthesis. Long-term autophagy inhibition results in altered pyruvate metabolism likely regulated by lower pyrimidine pools. Cells adapting to loss of autophagy preferentially salvage pyrimidines to replenish these pools instead of synthesizing them de novo. Exploiting this metabolic vulnerability, we found that acquired resistance to autophagy inhibition promotes increased salvage and therefore sensitivity to pyrimidine analogues, including gemcitabine and trifluridine/tipiracil leading to combinatory effects with autophagy inhibitors and pyrimidine analogs. These studies provide mechanistic insight defining how autophagy inhibition can be leveraged to treat pancreatic cancer.

Keywords: Autophagy; ULK1/2; drug resistance; gemcitabine; hydroxychloroquine; lysosomes; nucleotides; pancreatic cancer; pyrimidine metabolism; pyruvate metabolism.

Figure 1.. PDA cells that acquire resistance to autophagy inhibition have blocked autophagy

(A) Schematic overview of the process used to make resistant cells. (B-C) MTT cell viability assay in FC1199 CTL and HCQ-R cells treated with HCQ for 72h (B) and calculated IC50 values (C). (D-E) MTT cell viability assay in FC1199 CTL and MRT-R cells treated with MRT68921 for 72h (D) and calculated IC50 values (E). (F-G) Incucyte live cell imaging in FC1199 CTL, HCQ-R or MRT-R cells treated with HCQ or MRT68921 as indicated for 3 days. (H-I) FC1199 CTL, HCQ-R and MRT-R cells transfected with the mCherry-EGFP-LC3 tandem, starved for 24hrs in EBSS, or treated with Baf A1 (20nM) for 24hrs. mCherry/eGFP fluorescence was measured by flow cytometry (H) and autophagy induced under EBSS starvation was determined (I). (J-K) MTT cell viability assay in FC1199 CTL and HCQ-R cells treated with MRT68921 for 72h (J) and calculated IC50 values (K). (L-M) MTT cell viability assay in FC1199 CTL and MRT-R cells treated with HCQ for 72h (L) and calculated IC50 values (M). (N-Q) MTT cell viability assay in FC1199 CTL and HCQ-R cells treated with DC661 (N-O) or Baf A1 (P-Q) for 72h to determine the IC50 values. (R-U) MTT cell viability assay in FC1199 CTL and MRT-R cells treated with DC661 (N-O) or Baf A1 (P-Q) for 72h to determine the IC50 values. Data are represented as a mean ± SEM. Data on panels B, D, F-G, J, L, N, P, R and T show n=3 technical replicates representative of n=3 biological replicates. Data on panels C, E, I, K, M, O, Q, S and U represent n=3 biological replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001 determined by two-tailed Student’ t test (C, E, K, M, O, Q, S and U), two-way repeated measures ANOVA with Tukey’s post hoc test (F-G) and one-way ANOVA with Tukey’s post hoc test (I). MRT68921 abbreviated as MRT.

Figure 2.. PDA cells resistant to autophagy inhibition rewire their metabolism and decrease pyruvate decarboxylation

(A) Overlapping GO terms from RNA sequencing performed in FC1199 HCQ-R and MRT-R cells compared to CTL, and from FC1199 Atg7 KO compared to NT controls. (B) Top metabolism-related GO terms pathway significantly upregulated (left) or downregulated (right) in FC1199 HCQ-R, MRT-R and Atg7 KO cells. (C-D) Pyruvate (C) and citrate (D) isotopologue abundances in FC1199 CTL, HCQ-R and MRT-R cells after a 24h [U-¹³C₆] glucose trace. (E) Tracing map for [U-¹³C₆] glucose trace into pyruvate decarboxylation and TCA cycle. (F) Pyruvate decarboxylation measured as M2 Citrate to M3 pyruvate ratio from the [U-¹³C₆] glucose trace in FC1199 CTL, HCQ-R and MRT-R cells. (G) Correlation GO macroautophagy gene signature and the reactome regulation of PDH signature in PDA cells determined by single linear regression using Depmap data. (H-J) Seahorse assay performed in FC1199 CTL, HCQ-R and MRT-R cells (H) with calculated basal (I) and maximal (J) respiration rates displayed as the fold change relative to CTL. Data are represented as a mean ± SEM. Data on panels C, D F and H represent n=3–6 technical replicates representative of n=3 biological replicates. Data on panels I and J are representative of n=3 biological replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001, determined by two-tailed Student’ t test (C-D and F) and one-way ANOVA with Tukey’s post hoc test (H-J). MRT68921 abbreviated as MRT.

Figure 3.. PDA cells resistant to autophagy inhibition undergo anaplerosis to limit TCA dysfunctions

(A) Tracing map for [U-¹³C₆] glucose trace into pyruvate anaplerosis and TCA cycle round 1 (left) and round 2 (left). (B-D) Pyruvate anaplerosis measured as M3 Fumarate/M3 succinate (B), M3 Malate/M3 succinate (C) or M3 Aspartate/M3 succinate (D) from the [U-¹³C₆] glucose trace in FC1199 CTL, HCQ-R and MRT-R cells. Reference dotted line at 1.0 represents no pyruvate anaplerosis. (E) Tracing map for [3-¹³C₁] glucose trace into pyruvate anaplerosis. (F-G) Pyruvate anaplerosis measured as M1 aspartate/M1 pyruvate ratio (F) and M1 malate/M1 pyruvate (G) from the [3-¹³C₁] glucose trace in FC1199 CTL and HCQ-R cells. (H) Pcx normalized mRNA expression in FC1199 CTL, HCQ-R and MRT-R cells. (I) Western blots for PC and Tubulin in FC1199 CTL, HCQ-R and MRT-R cells. Images are representative of n=3 biological experiments. Data are represented as a mean ± SEM. Panels B-D and F-G represent n=3 technical replicates. Panel H represents n=3 biological replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001, determined by two-tailed Student’ t test (B-D and F-G) and one-way ANOVA with Tukey’s post hoc test (H). MRT68921 abbreviated as MRT.

Figure 4:. PDA cells resistant to autophagy inhibition have decreased nucleotide pools and altered pyrimidine metabolism

(A) Tpk1 normalized mRNA expression in FC1199 CTL, HCQ-R and MRT-R cells. (B) GSEA analysis for pyrimidine metabolism related processes in FC1199 CTL and HCQ-R cells. (C-D) CMP (D) and UMP (D) levels in FC1199 CTL, HCQ-R and MRT-R cells. (E) Western blot for DHODH, CAD, DCK, TK1, SLC29A1 and Tubulin in FC1199 CTL, HCQ-R and MRT-R cells. Images are representative of n=3 biological experiments. (F) Tracing map for [α−¹⁵N] glutamine trace into de novo pyrimidine synthesis. (G) CTP isotopologue abundance in FC1199 CTL, HCQ-R and MRT-R cells after a 24h [α−¹⁵N] glutamine trace. Data are represented as a mean ± SEM. Panels A and C-D represent n=3 biological replicates. Panel G represents n=3 technical replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001, determined by one-way ANOVA with Tukey’s post hoc test (A and G) and two-tailed Student’ t test (C-D). MRT68921 abbreviated as MRT.

Figure 5.. PDA cells resistant to autophagy inhibition have increased sensitivity to pyrimidine analogs

(A) Log2 fold change between the IC50 of FC1199 HCQ-R and CTL cells or FC1199 MRT-R and CTL cells for the indicated autophagy inhibitors, pyrimidine analogs and other PDAC standard of care determined with MTT cell viability assays. (B) Structure of deoxycytidine compared to gemcitabine and thymidine compared to trifluridine. (C-F) MTT cell viability assay and the calculated IC50 for FC1199 CTL and HCQ-R cells treated with Gemcitabine (C-D) and FTD/TPI (E-F) (G-J) MTT cell viability assay and the calculated IC50 for FC1199 CTL and MRT-R cells treated with Gemcitabine (G-H) and FTD/TPI (I-J) (K-L) Incucyte live cell imaging in FC1199 CTL and HCQ-R cells treated with gemcitabine (K) or FC1199 CTL and MRT-R cells treated with FTD/TPI (L) as indicated for 3 days. (M) ) Incucyte live cell imaging in FC1199 CTL and HCQ-R cells treated with gemcitabine or vehicle control with or without the addition of deoxycytidine as indicated for 3 days. (N) Incucyte live cell imaging in FC1199 CTL and MRT-R cells treated with FTD/TPI or vehicle control with or without the addition of thymidine as indicated for 3 days. (O) MTT cell viability assay in FC1199 CTL (NT or DCK KO) and HCQ-R cells (NT or DCK KO) treated with gemcitabine for 72h. (P-Q) Picture (P) and weight (Q) of the orthotopic tumors resected from mice injected with FC1245 CTL or HCQ-R cells treated with gemcitabine or saline control (n=9–10/group). (R) Schematic overview of the mechanisms identified or hypothesized in PDA cells resistant to autophagy inhibitors. Data are represented as a mean ± SEM. Panels A, D, F, H, and J represent n=3 biological replicates. Panel D-G represents n=3 technical replicates representative of n=3 biological replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001, determined by two-tailed Student’ t test (A, D, F, H, and J), two-way repeated measures ANOVA with Tukey’s post hoc test (K-N) and one-way ANOVA with Tukey’s post hoc test (Q). MRT68921 abbreviated as MRT.

Figure 6:. Autophagy inhibitors and pyrimidine analogues have a combinatory effect in PDA

(A) Correlation between autophagy gene signature and gemcitabine sensitivity in hPDOs determined by single linear regression. (B-I) Incucyte live cell imaging in FC1199 cells treated as indicated for 3 days and graphed as percent confluence (B,D,G,H) or caspase 3/7 object count/mm² normalized to confluence (E,H) with (F) showing representative phase imaging and cas3/7 object count from (C and E) and (I) showing representative phase imaging from (G and H). (J-L) Tumor volumes (J) from mice injected orthotopically with FC1199 cells and treated with gemcitabine and/or SBP-1750 or vehicle control with (K) showing representative images of tumors and (L) changes in body weight starting 7 post orthotopic injection on day 1 of treatment (K). n = 9–10 per group. Data are represented as a mean ± SEM. Data on panels B-E and G-H represent n=3 technical replicates representative of n=3 biological replicates. * indicates p<0.05, ** p<0.001, *** p <0.0001, **** p<0.00001, determined by two-way repeated measures ANOVA with Tukey’s post hoc test (B-E, G-H and L) and one-way ANOVA with Tukey’s post hoc test (J). MRT68921 abbreviated as MRT.