Figure 1. Glucose restriction promotes a liver tumor-initiating cell phenotype.

Huh7 and CLC13 HCC cells were cultured in high (4.5 g/L) or restricted/low (0.45 g/L) glucose. (A) In vitro limiting-dilution assays showed that the frequency of tumor-initiating cells increased 4-fold after culturing in low glucose (pairwise tests for differences in stem cell frequencies). (B) In vivo limiting-dilution assays found that the cells cultured in low glucose displayed an enhanced tumor incidence, expedited tumor latency, and a higher frequency of tumor-initiating cells (primary implantation, n = 15 per group; secondary implantation, n = 5 per group). (C) Strategy for mRNA profiling to identify altered transcriptomes of HCC cells grown in high- or low-glucose conditions. (D) Gene set enrichment analysis (GSEA) of differentially expressed genes identified by RNA-seq found that the PERK-mediated unfolded protein response was highly enriched in HCC cells cultured under low-glucose conditions. (E) Western blot analysis also found that GRP78, p-PERK, p-eIF2α, ATF4, and FUT1 levels were enhanced in glucose-restricted conditions and that the addition of 1 μM PERK inhibitor (PERKi) for 48 hours reversed the expression of p-PERK, p-eIF2α, ATF4, and FUT1. HG, high glucose; LG, low glucose. (F) In vitro limiting-dilution assays showed that the frequency of tumor-initiating cells increased after culture in low glucose and decreased when HCC cells cultured in low glucose were treated with PERKi (1 μM) (pairwise tests for differences in stem cell frequencies). The data shown in A, E, and F are representative of 3 independent experiments. CSC, cancer stem cell; FDR, false discovery rate. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.