Cancer cells with high-metastatic potential promote a glycolytic shift in activated fibroblasts

Abstract

Cancer-associated fibroblasts (CAFs) are activated fibroblasts and are the major stromal component in various types of malignancies. CAFs often undergo metabolic reprogramming to create an appropriate microenvironment for cancer progression. However, it remains unclear whether the metastatic properties of cancer cells affect aerobic glycolysis in stromal cells. Here, we show that gastric cancer (GC) cells with high metastatic potential strongly promote the metabolic switch from oxidative phosphorylation to aerobic glycolysis in fibroblasts. Transcriptome analysis showed that the expression of glycolysis-related genes, such as LDHA and ENO2, significantly changed in fibroblasts when they were cocultured with cancer cells with high metastatic potential compared to fibroblasts incubated with cancer cells with low metastatic potential. Glucose uptake, lactate production and oxygen consumption in fibroblasts were changed by coculture with GC cells with high metastatic potential. Thus, metabolic reprogramming in CAFs may reflect the metastatic properties of GC cells.

Fig 1. Gastric cancer cells with high metastatic potential increased the expression level of glycolysis-related genes in the cocultured fibroblasts.

(A) Schematic protocol of coculture and gene expression analysis. (B) GSEA of the fibroblasts cocultured with 44As3 versus those cocultured with HSC-44PE (upper and lower left), the fibroblasts cocultured with 44As3 versus mono-cultured fibroblasts (upper and lower middle), and the fibroblasts cocultured with HSC-44PE versus mono-cultured (upper and lower right). NES: normalized enrichment score. The p-value was calculated by GSEA. (C) A heat map shows glycolysis-related gene expression in each culture condition. n = 2 technical replicates. (D) qRT-PCR analysis of ENO1, ENO2, LDHA, PDK1, and PDK3 expression in iNF-58 cells cocultured with DGC cells with high metastatic potential compared to the expression in mono-cultured iNF-58 cells. n = 3 biological replicates. Error bars represent s.d. *, p < 0.05 from ANOVA followed by Tukey’s HSD post hoc comparisons. (E) Western blot analysis of glycolysis-related proteins, ENO1, ENO2, LDHA, PDK1, PDK3, and β-actin in iNF58 cells in mono-culture or coculture with DGC cells (left). Densitometric analysis of Western blot on ENO2, LDHA and PDK3 normalized to the level of β-actin (right). n = 3 biological replicates. Error bars represent s.d. *, p < 0.05, **, p < 0.01 from ANOVA followed by Tukey’s HSD post hoc comparisons.

Fig 2. DGC cells with high metastatic potential enhanced the metabolic switch to aerobic glycolysis in the fibroblasts.

(A) Quantification of lactate production and glucose consumption in cocultured or mono-cultured iNF-58 cells. n = 3 biological replicates. Error bars represent s.d. *, p < 0.05 from ANOVA followed by Tukey’s HSD post hoc comparisons. (B) The pH of medium in which cocultured or mono-cultured iNF-58 cells and iNF-60 cells were maintained. n = 4 technical replicates in each fibroblast. Error bars represent s.d. *, p < 0.05 from ANOVA followed by Tukey’s HSD post hoc comparisons. (C) The cell proliferation rate of iNF-58 cells (left) and DGC cell lines (right) in the mono-culture and coculture. n = 3 technical replicates. Error bars represent s.d. *, p < 0.05 from ANOVA followed by Tukey’s HSD post hoc comparisons. (D) GSEA of 44As3 cells cultured with fibroblasts (As3 with NF) versus HSC-44PE cells cultured with fibroblasts (PE with NF), highlighting cell proliferation-related phenotypes. NES: a normalized enrichment score. The p-value was calculated by GSEA.

Fig 3. DGC cells with high metastatic potential promoted the metabolic switch from oxidative phosphorylation to aerobic glycolysis in fibroblasts.

(A) The lifetimes of oxygen consumption of iNF-58 cells in mono-culture and coculture is indicated by changes in fluorescence. n = 3 technical replicates. (B) OCR and ECAR in iNF-58 cells cocultured with DGC cells with high metastatic potential and mono-cultured iNF-58 cells. n = 8 technical replicates. (C) The histogram shows the basal oxygen consumption level in iNF-58 cells at three different time points (18, 30, and 41 min). (D) The histogram shows the ratio of ECAR:OCR at baseline (41 min).