Obesogenic High-Fat Diet and MYC Cooperate to Promote Lactate Accumulation and Tumor Microenvironment Remodeling in Prostate Cancer

Abstract

Cancer cells exhibit metabolic plasticity to meet oncogene-driven dependencies while coping with nutrient availability. A better understanding of how systemic metabolism impacts the accumulation of metabolites that reprogram the tumor microenvironment (TME) and drive cancer could facilitate development of precision nutrition approaches. Using the Hi-MYC prostate cancer mouse model, we demonstrated that an obesogenic high-fat diet (HFD) rich in saturated fats accelerates the development of c-MYC-driven invasive prostate cancer through metabolic rewiring. Although c-MYC modulated key metabolic pathways, interaction with an obesogenic HFD was necessary to induce glycolysis and lactate accumulation in tumors. These metabolic changes were associated with augmented infiltration of CD206+ and PD-L1+ tumor-associated macrophages (TAM) and FOXP3+ regulatory T cells, as well as with the activation of transcriptional programs linked to disease progression and therapy resistance. Lactate itself also stimulated neoangiogenesis and prostate cancer cell migration, which were significantly reduced following treatment with the lactate dehydrogenase inhibitor FX11. In patients with prostate cancer, high saturated fat intake and increased body mass index were associated with tumor glycolytic features that promote the infiltration of M2-like TAMs. Finally, upregulation of lactate dehydrogenase, indicative of a lactagenic phenotype, was associated with a shorter time to biochemical recurrence in independent clinical cohorts. This work identifies cooperation between genetic drivers and systemic metabolism to hijack the TME and promote prostate cancer progression through oncometabolite accumulation. This sets the stage for the assessment of lactate as a prognostic biomarker and supports strategies of dietary intervention and direct lactagenesis blockade in treating advanced prostate cancer.

Significance: Lactate accumulation driven by high-fat diet and MYC reprograms the tumor microenvironment and promotes prostate cancer progression, supporting the potential of lactate as a biomarker and therapeutic target in prostate cancer. See related commentary by Frigo, p. 1742.

Figure 1.

Obesogenic high-fat diet accelerates the development of invasive adenocarcinoma. A, Experimental overview. B, Representative images of H&E (top), segmented invasive and outgrowing areas (middle), and a-smooth muscle actin (αSMA) IHC (bottom). The inset is a magnification of an area of invasive adenocarcinoma with complete negativity for αSMA. Black arrow, αSMA staining in vascular smooth muscle cells, which were used as internal positive control. Scale bar is shown. C, Dot plot showing the percentage of PIN in the whole lobe epithelium (benign + atypical). PIN percentage in anterior (AP), ventral (VP), and dorsolateral (DLP) prostate lobes was quantified using digital pathology (***, P = 0.0001, two-sided Mann–Whitney U test; n, biologically independent samples; mean ± SD). D, Bar plot showing the incidence of invasive adenocarcinoma. Data are presented as the percentage of MYC mice in each condition (CTD_MYC, n = 21; HFD_MYC, n = 24). A P value is shown (two-sided Fisher exact test). E, Dot plot comparing the size of outgrowing areas in DLP (***, P = 0.0002, two-sided unpaired t test; n, biologically independent samples; mean ± SD). ELISA, enzyme-linked immunoassay; ISH, in situ hybridization; WB, Western blotting.

Figure 2.

Obesogenic high-fat diet amplifies critical MYC-dependent metabolic vulnerabilities. A, Principal component analysis of metabolomics data from DLP. Four diet/genotype conditions are represented (n = 6 biologically independent samples/group, 603 metabolites analyzed). B, Pie charts showing the proportion of metabolites significantly altered by MYC [MYC effect (independent of diet); P < 0.05, FDR < 0.15, two-way ANOVA] or by HFD in MYC-transformed or WT DLP (HFD effect; P < 0.05, FDR < 0.15, two-sided unpaired t test). C, Dot plots showing relevant metabolites of key pathways enhanced by HFD in MYC-transformed DLP (P < 0.05; FDR < 0.15, two-sided unpaired t test; n = 6 biologically independent samples/group; mean ± SD). For each metabolite, comparison between HFD_MYC versus CTD_MYC is shown. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Exact P and FDR values are summarized in Supplementary Table S8. αKG, α-ketoglutarate; BCAA, branched chain amino acids; Gln, glutamine; Glu, glutamate; Gly, glycine; Ile, isoleucine; Leu, leucine; 5-MTA, 5′-methylthioadenosine; Met, methionine; PC, principal component; SAM, s-adenosylmethionine; Ser, serine; Val, valine; WT, wild-type.

Figure 3.

Obesogenic high-fat diet promotes aerobic glycolysis in MYC-driven prostate cancer. A, Heat map showing the 46 metabolites that are uniquely altered by HFD in MYC-transformed DLP (P < 0.05; FDR < 0.15, two-sided unpaired t test; n = 6 biologically independent samples/group). Metabolite relative concentration, exact P and FDR values are summarized in Supplementary Table S8. Metabolites of aerobic glycolysis are highlighted in bold red. Metabolites were measured using LC/MS-MS. B, Relative quantification of glucose 6-phosphate (*, P = 0.0211) and lactate (*, P = 0.0247), two-sided unpaired t test; n = 6 biologically independent samples/group; mean ± SD. C, Association between mouse weight and lactate levels (Pearson correlation). Statistics are indicated. D, Lactate concentration in DLP using NMR (*, P = 0.0227, two-sided unpaired t test, n = 3–5 biologically independent samples). E, Western blot densitometric analysis of LDHA (EXP 1, **, P = 0.0079; EXP 2 and *, P = 0.0358, two-sided unpaired t test), MCT-1 (P = 0.3939, two-sided Mann–Whitney U test), and MCT-4 (P = 0.1803, two-sided unpaired t test); n = 3–6 biologically independent samples/group; mean ± SD. Protein levels are normalized to β-actin or vinculin and expressed as arbitrary units (AU). F, Heat map showing enriched/depleted gene sets in DLP (GSEA_Hallmark, P < 0.01; FDR < 0.01). Critical gene sets enriched by MYC and further enhanced by HFD are shown in green. MYC transcriptional activity is emphasized in bold green. Critical gene sets exclusively enriched in MYC-transformed DLP from mice fed HFD are marked in red. Glycolysis and related gene sets are further highlighted in bold red. Normalized enrichment scores (NES) are reported in Supplementary Table S10. WT, wild-type.

Figure 4.

Obesogenic high-fat diet boosts glucose uptake. A, Imaging experiment workflow. B, Mouse weight before the start of CTD/HFD feeding (P = 0.7317) and at the end of the study. **, P = 0.0023, two-sided unpaired t test; n = 9 mice/group; mean ± SD. C, Representative PET images (axial, coronal, and sagittal sections) of mice placed in prone position. D, Bar plots showing difference in mean (*, P = 0.0355) and maximum (*, P = 0.0279) SUV. Two-sided unpaired t test; n = 4–5 mice/group; mean ± SD. One mouse (HFD_MYC) with tumor volume unsuitable for PET analysis was excluded (see Materials and Methods). E, Western blot of GLUT-1, LDHA, MCT-1, and MCT-4 proteins. F, Densitometric analysis. Protein levels are normalized to β-actin or vinculin and expressed as arbitrary units (AU; GLUT-1, P = 0.0540; LDHA, P = 0.3137; MCT-1, P = 0.8445; MCT-4, P = 0.9609, two-sided unpaired t tests; n = 4–5 biologically independent samples/group; mean ± SD). ¹⁸FDG-PET/CT, 2-deoxy-2-[18F]fluoro-D-glucose-PET/CT; GLUT-1, glucose transporter 1.

Figure 5.

Obesogenic high-fat diet promotes TAM and Treg infiltration in invasive prostate cancer. A, Box plots showing macrophage proportion using ImmuCC and mMCP deconvolution models (ImmuCC, *, P = 0.016; mMCP, **, P = 0.0079, two-sided Wilcoxon rank-sum test). Box plots show median value, box boundaries: 25th and 75th percentiles; interquartile range (IQR), whiskers: max or min value before the 1.5x IQR fence. The dot plotted in the IQR represents the mean value. B and C, Enrichment plot of TAM (B) and tumor-infiltrating Treg (TITR; C) signatures in MYC-transformed DLP. P value and normalized enrichment score (NES) are indicated. D, Left, representative images of F4/80 (teal) and CD206 (yellow) dual IHC staining in DLP stroma. Images show segmentation of double positive (F4/80⁺CD206⁺; green) and single positive (F4/80⁺; blue) macrophages. Scale bar, 20 μm. Right, quantification (dot plot) of F4/80⁺CD206⁺ macrophages/stromal area in each case (P = 0.0140, two-sided unpaired t test). One case (HFD_MYC) was removed from the analysis due to a staining issue. E, Left, representative images of dual ISH with RNAscope probes for Mm-Adgre 1 (F4/80; teal) and Mm-Cd274-C2 (PD-L1; red). Right, quantification (dot plot) of F4/80⁺PD-L1⁺ macrophages/area in each case (P = 0.0260, two-sided Mann–Whitney U test). Dual ISH was performed in a subset of cases. One case (CTD_MYC) was removed from the analysis due to a technical issue. F, Representative images of FOXP3 immunohistochemical staining in DLP stroma. FOXP3⁺ lymphocytes are highlighted in green. Scale bar is shown. Black arrows, invasive glands. G, Quantification of FOXP3⁺ lymphocytes using digital pathology. Data are expressed as number of FOXP3⁺ lymphocytes/stromal area (left, P = 0.0036; right, P = 0.0026, two-sided Mann–Whitney U test). H, Dot plot showing kynurenine levels measured in DLP by LC-MS-MS (*, P = 0.04, FDR < 0.15, two-sided unpaired t test). In D, E, G, and H, data are expressed as mean ± SD; n = biological independent mice/group.

Figure 6.

Lactate promotes tubulogenesis and prostate cancer cell migration. A, Enrichment plot of an endothelial cell signature derived from scRNA-seq analysis of prostate lobes from 24-week-old mice. P value and NES are indicated. B, Tube formation assay. Left, representative pictures of HUVEC cells grown on Matrigel and stained with calcein AM green-fluorescent dye. Cells were treated for 12 hours with lactate or PBS. Treatment with VEGF was used as positive control. Scale bar, 100 μm. Right, quantification of tube segments and nodes (tube segments: *, P = 0.0333; **, P = 0.0060; VEGF, ***, P = 0.0006; tube nodes: *, P = 0.0160; **, P = 0.0073; VEGF, **, P = 0.0014, Kruskal–Wallis test). Number of segments and nodes in each well was normalized to the mean value of PBS condition in each experiment. Two independent experiments with three biological replicates; each was performed (two pictures/well were taken in EXP 1; one picture/well was taken in EXP 2, n = 9 pictures/group). C, Left, representative pictures of HUVEC cells grown on Matrigel in the presence of the LDHA inhibitor FX11 or DMSO for 10 hours. Scale bar, 100 μm. Right, quantification of tube segments and nodes (tube segments: **, P = 0.0057; tube nodes: **, P = 0.0037, two-sided unpaired t test). Experiment was performed with three biological replicates (each biological replicate was the average of 5 technical replicates). In B and C, mean values ± SD are shown. D, Left, representative pictures of HUVEC cells grown on Matrigel treated with MCT-1 / MCT-4 dual inhibitor syrosingopine (Syro, 10 μmol/L) or DMSO in the presence of lactate 10 mmol/L for 12 hours. Scale bar, 100 μm. Magnifications, ×40 (right). Experiment was performed with two biological replicates (three technical replicates/each) and repeated twice. E, Bar plot showing the fraction of motile and non-motile cells under lactate or PBS treatment for 3 hours. Cells were previously pretreated for 72 hours. A P value is shown (two-sided Fisher exact test). Three independent experiments with three biological replicates were performed. F, Mean square displacement (MSD) overtime and matching nonlinear fits (95% CI; 301 cells/condition; mean ± SEM). G, MYC-CaP wound-healing assay. Bar plots showing the percentage of wound closure after 48 hours of treatment with lactate/PBS (**, P = 0.0047) or with FX11/DMSO (**, P = 0.0058), two-sided unpaired t test; mean ± SD. Three independent experiments with three biological replicates; each was performed. Two pictures/well were taken. H, Representative traction maps of MYC-CaP cells treated with lactate or PBS. Cell contour is shown (black line). I, Box plot showing average force in MYC-CaP cells treated with 10 mmol/L lactate (n = 103) or PBS (n = 95; *, P = 0.0139; two-sided unpaired t test with Welch correction). Box-plot shows median value, box boundaries: 25th and 75th percentiles; interquartile range, whiskers: min to max value. La, lactate; Pa, Pascal.

Figure 7.

High saturated fat intake and BMI are associated with prostate cancer tumors characterized by glycolytic features and an immunosuppressive TME. A, Comparison of BMI in prostate cancer patient with high and low SFI. B and C, Enrichment plot of “Hallmark_glycolysis” gene set in patients with prostate cancer stratified for SFI (B) and BMI (C). Normalized enrichment score (NES), FDR value, and number of cases are indicated. D, Box plots showing the proportion of tumor-infiltrating macrophages using QuanTIseq deconvolution model. Number of cases and P values are indicated. E and F, Box plots comparing LDHA mRNA levels in patients with prostate cancer with different Gleason score (GS; E) and BCR status (F). P values and number of patients are indicated; two-sided Mann–Whitney U test. Box plots show median value, box boundaries: 25th and 75th percentiles; interquartile range, whiskers: Min to max value. G, Kaplan–Meier curves of disease-free survival. A P value is indicated (log-rank test). H, Box plot comparing LDHA mRNA levels in patients with prostate cancer with/without BCR from the META855 dataset. For D and H, Wilcoxon rank-sum test was used; P value and number of patients are indicated. Box plots show median value, box boundaries: 25th and 75th percentiles; interquartile range (IQR), whiskers: max or min value before the 1.5x IQR fence. Dot plotted in the IQR represents the mean value. ns, not significant; PCa, prostate cancer.

Figure 8.

Tumor acquisition of glycolytic features is associated with worse prognosis. A, Multivariable analysis (MVA) using the proportional hazards regression model. Hazard ratio (HR) ± 95% confidence interval (CI) is shown. The P value was calculated with the Wald test. B, Graphical summary. The oncogene MYC promotes a broad metabolic reprogramming. Obesogenic HFD not only enhances MYC-driven metabolic vulnerabilities (bold black) but also induces aerobic glycolysis and lactate accumulation (bold red). The latter boosts angiogenesis, ECM remodeling, prostate cancer cell migration, immune evasion, generating a TME permissive of prostate cancer progression. Red outline in hexagons/circles indicates metabolites that are increased with HFD in MYC-transformed DLP. αKG, α-ketoglutarate; FADH₂, flavin adenine dinucleotide; FA, fatty acid; Glc, glucose; Glu, glutamate; Gln, glutamine; LPL, lysophospholipid; OXPHOS, oxidative phosphorylation; NADH, nicotinamide adenine dinucleotide; PL, phospholipid; PPP, pentose phosphate pathway; Pyr, pyruvate; Ser/Gly: serine/glycine SPL, sphingolipid; TG, triglyceride. Of note, food images are not a direct translation of the murine diets used in this study.