eIF4A supports an oncogenic translation program in pancreatic ductal adenocarcinoma

Abstract

Pancreatic ductal adenocarcinoma (PDA) is a lethal malignancy with limited treatment options. Although metabolic reprogramming is a hallmark of many cancers, including PDA, previous attempts to target metabolic changes therapeutically have been stymied by drug toxicity and tumour cell plasticity. Here, we show that PDA cells engage an eIF4F-dependent translation program that supports redox and central carbon metabolism. Inhibition of the eIF4F subunit, eIF4A, using the synthetic rocaglate CR-1-31-B (CR-31) reduced the viability of PDA organoids relative to their normal counterparts. In vivo, CR-31 suppresses tumour growth and extends survival of genetically-engineered murine models of PDA. Surprisingly, inhibition of eIF4A also induces glutamine reductive carboxylation. As a consequence, combined targeting of eIF4A and glutaminase activity more effectively inhibits PDA cell growth both in vitro and in vivo. Overall, our work demonstrates the importance of eIF4A in translational control of pancreatic tumour metabolism and as a therapeutic target against PDA.

Fig. 1

The eIF4A inhibitor CR-31 potently targets pancreatic ductal adenocarcinoma. a Normal (N), and Kras^G12D;p53^R172H (KP) organoids were treated with 10 nM CR-31 for 60 min and compared to vehicle (DMSO). During the last 30 min, 10 μM O-propargyl-puromycin (OP-Puro) was supplemented into the media. Dissociated single cells were fixed, stained by CuAAC with Alexa488-azide and the level of OP-Puro conjugation within polypeptide chains was quantified by flow cytometry. b Cell viability of N and KP organoids upon a 72 h treatment with increasing concentrations of CR-31. Dotted lines indicate 95% confidence intervals (n = 5). Inset shows representative images of organoids treated with 10 nM CR-31 for 72 h. Scale bars, 1000 μm. c SUnSET assay of representative PDA tumours from mice following 7 daily treatments with vehicle (5.2% PEG-400/5.2% Tween80) or 0.2 mg kg⁻¹ CR-31. Mice were euthanized for tissue collection 2 h after the final treatment with vehicle or CR-31 and 30 min after 40 nmol g⁻¹ puromycin injection. Scale bars, 50 μm. d Relative tumour volume of PDA-bearing mice after 7 daily treatments with 0.2 mg kg⁻¹ CR-31 or vehicle. Relative growth was estimated by comparison of tumour volume 7 days post injection to that at pre-enrolment (day 0) by ultrasound imaging. Data are mean +/– S.D. Student’s t-test. (n = 11, vehicle; n = 7, CR-31). e Immunohistochemistry of cleaved caspase 3 (left) in representative tumours from mice treated daily for 7 days with vehicle or 0.2 mg kg⁻¹ CR-31 together with quantification of cleaved caspase 3 (right) from n ≥ 7 fields of view. Data are mean +/–S.D. Student’s t-test. Tissues were harvested 2 h post final treatment. Scale bars, 50 μm. f Kaplan–Meier survival analysis for Kras^G12D;p53^R172H;PdxCre (KPC) mice treated daily with vehicle (n = 13) or 0.2 mg kg⁻¹ CR-31 (n = 16). Mantel-Cox (log rank) test. Source data are provided as a Source Data file

Fig. 2

Tumour-selective modulation of translational efficiencies by CR-31. a Scatterplot of polysome-associated mRNA to total mRNA log2 fold-changes comparing vehicle-treated Kras^G12D;p53^R172H (KP) to normal (N) organoids. The numbers of mRNAs with a change in translation efficiency (light red and dark red) or mRNA abundance (light green and dark green) are indicated (n = 3; positive log2 fold-change indicates higher translation/abundance in KP organoids). b Scatterplot of polysome-associated mRNA to total mRNA log2 fold-changes comparing N organoids treated with vehicle or 10 nM CR-31 for 1 h. The numbers of mRNAs with a change in translation efficiency (light red and dark red) are indicated (negative log2 fold-change indicates suppressed translation upon CR-31 treatment). c Scatterplot of polysome-associated mRNA to total mRNA log2 fold-changes comparing KP organoids treated with vehicle or 10 nM CR-31 for 1 h. The numbers of mRNAs with a change in translation efficiency (light red and dark red) are indicated (negative log2 fold-changes indicates suppressed translation upon CR-31 treatment). d Scatterplot of log2 fold-changes for polysome-associated mRNA and total mRNA (left), an empirical cumulative distribution (ecdf) of log2 fold-changes for polysome-associated mRNA (middle) and total mRNA (right) comparing KP and N organoids. Transcripts whose translation was activated (light red) or suppressed (dark red) upon CR-31 treatment in KP organoids (i.e. from (c)) are indicated for each plot. Wilcoxon rank-sum test p-values comparing regulation of subsets to background are indicated together with the log2 fold-changes difference between these subsets to the background at the 50th percentile (q50) of the ecdf. e Same representation as in panel d but examining differences in gene expression in N organoids upon CR-31 treatment. Source data are provided as a Source Data file

Fig. 3

CR-31 translationally targets glutathione metabolism in PDA. a Heatmap of glutathione metabolism pathway proteins (KEGG database) showing fold-changes of polysome-associated mRNA. b Relative levels of NADPH normalized to viability based on CellTitre-Glo (CTG) assay (left) and ratio of NADP⁺ to NADPH (right) in organoids and patient-derived PDA lines upon treatment with vehicle (DMSO) or 100 nM CR-31 for 16 h. Data are mean +/–S.D. Student’s t-test, n = 4 (organoids), n = 5 (patient-derived lines). NS = not significant. c Relative levels of glutathione in patient-derived PDA lines upon treatment with vehicle or CR-31 for 24 h. Data are mean +/–S.D., Student’s t-test. n = 5, NS = not significant. *p < 0.05 compared to DMSO treatment. d Ratio of GSH to GSSG in patient-derived PDA lines upon treatment with vehicle or the indicated concentrations of CR-31 for 24 h. Data are mean +/–S.D., Student’s t-test, n = 5. *p < 0.05 compared to DMSO treatment. e ROS levels in KP organoids upon treatment with vehicle or 10 nM CR-31 for 16 h. Data are representative from three biological replicates. f NRF2 protein levels in KP organoids treated with CR-31 for 16 h. Actin, loading control. Representative image from 3 biological replicates. g mRNA expression of NRF2 target genes in 3 patient-derived PDA lines upon treatment with CR-31 for 5 or 24 h. Data are mean +/–S.D. Each data point plotted is average from every biological replicate. h Cell viability of KP and Nrf2-deficient KP (KPn) organoids upon a 72 h treatment with increasing concentrations of CR-31. Dotted lines indicate 95% confidence intervals, n = 5. i Cell viability of patient-derived PDA lines upon treatment with increasing concentrations of CR-31 in the presence or absence of 100 μM buthionine sulfoximine (BSO) for 72 h. Data are mean +/–S.D., n = 5. j Cell viability of primary murine PDA lines upon treatment with increasing concentrations of CR-31 in the presence or absence of 1 mM of the ROS scavenger N-acetylcysteine (NAC) for 72 h. Data are mean +/–S.D., n = 3. Source data are provided as a Source Data file

Fig. 4

CR-31 translationally targets oxidative phosphorylation and glycolysis in PDA. a Heatmap of oxidative phosphorylation pathway proteins showing fold-changes of polysome-associated mRNA. b Boxplots of total and polysome-associated mRNA levels for transcripts encoding oxidative phosphorylation components. Horizontal line indicates the median; lower and upper box indicate the first and third quartiles. Whiskers extend from the box to the most extreme value but not further than +/−1.5 * inter-quartile range (IQR) from the box. Data exceeding the whiskers are plotted as points. b ¹³C₆-glucose tracing of N (n = 2) and KP organoids (n = 4). M +2 labelled citrate is derived directly from ¹³C₆-glucose. Data are mean +/– S.D., Student’s t-test. c Mitochondrial respiration reflected by oxygen consumption rate (OCR) was measured in KP organoids treated with vehicle or CR-31 for 5 h and following the addition of Oligomycin (1 μM), the uncoupler FCCP (0.5 μM), and the electron transport inhibitor Rotenone/Antimycin A (0.5 μM). Data are mean +/– S.D, n = 3. Data are representative from three biological replicates. d OCR was measured in N organoids treated with vehicle CR-31 for 5 h and following the addition of Oligomycin, FCCP, and Rotenone/Antimycin A. Data are mean +/– S.D. n = 5. Data are representative from three biological replicates. e Top, flow of heavy carbons through the glycolytic pathway. Bottom, ¹³C₆-glucose tracing of two sets of N and KP organoids (left) and 4 patient-derived PDA cell lines (middle and right) upon treatment with vehicle or 100 nM CR-31 for 5 h. Data are mean +/– S.D., Student’s paired t-test. f Glycolysis reflected by extracellular acidification rate (ECAR) was measured in KP organoids treated with vehicle or CR-31 for 5 h and following addition of glucose (10 mM), Oligomycin (1 μM), and the glucose analog, 2-deoxyglucose, 2DG (50 mM). Data are mean +/– S.D., n = 3. Data are representative of three biological replicates. g ECAR was measured in N organoids treated with vehicle or CR-31 for 5 h and following addition of glucose, Oligomycin, and 2DG. Data are mean +/– S.D., n = 5. Data are representative of three biological replicates. Source data are provided as a Source Data file

Fig. 5

CR-31 translationally targets glucose uptake in PDA. a Levels of transcripts encoding glycolytic proteins in light (<3 ribosomes) or heavy (≥3 ribosomes) polysome fractions from N or KP organoids treated with 10 nM CR-31 for 1 h relative to vehicle. Data are mean +/− S.D. n = 6. b Immunoblot analysis of SLC2A6, ENOLASE 1, ENOLASE2, ENOLASE 3, and GAPDH in N and KP organoids upon vehicle or CR-31 treatment for 5 h. Vinculin, loading control. c Immunoblot analysis of SLC2A6, ENOLASE1, ENOLASE2, ENOLASE3, and GAPDH in patient-derived PDA lines upon vehicle or CR-31 treatment. VINCULIN, loading control. d Glucose concentration in media spent from KP organoids upon treatment with vehicle or CR-31. Data are mean +/− S.D., Student’s t-test, n = 3. *p < 0.05, compared to DMSO treatment at the same timepoint. e Representative ¹⁸F-FDG-PET images of pancreatic tumours from mice after 24 h of treatment with vehicle or CR-31. Top, coronal view; bottom, transverse view. f Tumour uptake plots comparing initial metabolic activity before and after treatment with vehicle or CR-31. Left, maximum standardized uptake value (SUV)_max of tumour. Right, ratio of (SUV)_max in tumour compared to muscles. Data are mean +/− S.D., Student’s t-test. n = 7, Vehicle; n = 8, CR-31. g Glucose uptake comparing Cas9-expressing murine PDA cell lines transduced with sgRNAs targeting the ROSA locus (sgROSA) or against different regions of the Slc2a6 gene (sgSlc2a6). Data are mean +/− S.D., normalized to cell numbers. Student’s t-test, n = 4. h ECAR was measured in equal numbers of Cas9-expressing primary murine PDA cell lines transduced with sgRNAs targeting the ROSA locus (sgROSA) or different regions of the Slc2a6 gene (sgSlc2a6). ECAR was measured following the addition of glucose, Oligomycin, and 2DG. Data are mean +/− S.D., n = 5. i ECAR was measured in equal numbers of Cas9-expressing KP organoids transduced with sgRNAs targeting the ROSA locus (sgROSA) or different regions of the Slc2a6 gene (sgSlc2a6). ECAR was measured following the addition of glucose, Oligomycin, and 2DG. Data are mean +/− S.D., n = 6. Source data are provided as a Source Data file

Fig. 6

CR-31 treatment increases reverse glutamine metabolism in pancreatic cancer cells. a Levels of transcripts encoding proteins involved in glutamine metabolism in light (<3 ribosomes) or heavy (≥3 ribosomes) polysome fractions from N or KP organoids treated with 10 nM CR-31 for 1 h relative to vehicle (DMSO). Data are mean +/– S.D., n = 6. Data from two biological replicates are shown. b Immunoblot analysis of GLS in patient-derived PDA cell lines upon treatment with vehicle or CR-31 for 5 h. VINCULIN, TUBULIN, loading controls. c Intracellular levels of glutamine at steady state from 3 patient-derived PDA cell lines (2 technical replicates each) treated with vehicle or 100 nM CR-31 for 5 h. Data are mean +/– S.D., Student’s paired t-test. d ¹³C₅-glutamine tracing of 4 patient-derived PDA cell lines upon treatment with vehicle or 100 nM CR-31 for 5 h. Data are mean +/– S.D., Student’s paired t-test. Right, schematic illustrating the flow of heavy carbons through glutamine metabolic pathways. e ¹³C₅-glutamine tracing of 4 patient-derived PDA cell lines upon treatment with vehicle or 100 nM CR-31 for 5 h. M + 4 indicates metabolites of oxidative glutaminolysis and M + 5 indicates metabolites of reductive glutaminolysis. Data are mean +/– S.D., Student’s paired t-test. f ¹³C₅-glutamine tracing of N (n = 2) and KP (n = 2) organoids upon treatment with vehicle or 100 nM CR-31 for 5 h. Data are mean +/– S.D., Student’s t-test. g Cell viability of patient-derived PDA lines upon a 72 h treatment with increasing concentrations of CR-31 and the indicated concentrations of the glutaminase inhibitor BPTES (n = 5) or CB839 (n = 3). Data are mean +/– S.D., Student’s t-test, *p < 0.05 compared to CR-31 only treatment. h Cell viability of KP organoids upon a 72 h treatment with increasing concentrations of CR-31 and the indicated concentrations of CB839. Dotted lines indicate 95% confidence intervals, n = 5. i KP organoids (GFP-labelled) in co-culture with pancreatic cancer-associated fibroblasts (mCherry-labelled) upon a 72 h treatment with CR-31 in the presence or absence of CB839. Representative images from two biological replicates. Scale bars, 1000 μm. Source data are provided as a Source Data file