The Effects of Hypoxia on the Immune-Metabolic Interplay in Liver Cancer

Abstract

M2-like macrophages promote tumor growth and cancer immune evasion. This study used an in vitro model to investigate how hypoxia and tumor metabolism affect macrophage polarization. Liver cancer cells (HepG2 and VX2) and macrophages (THP1) were cultured under hypoxic (0.1% O₂) and normoxic (21% O₂) conditions with varying glucose levels (2 g/L or 4.5 g/L). Viability assays and extracellular pH (pHe) measurements were conducted over 96 hours. Macrophages were exposed to the tumor-conditioned medium (TCM) from the cancer cells, and polarization was assessed using arginase and nitrite assays. GC-MS-based metabolic profiling quantified TCM meta-bolites and correlated them with M2 polarization. The results showed that pHe in TCMs decreased more under hypoxia than normoxia (p < 0.0001), independent of glucose levels. The arginase assay showed hypoxia significantly induced the M2 polarization of macrophages (control group: p = 0.0120,^0.1%VX2-TCM group: p = 0.0149, ^0.1%HepG2-TCM group: p < 0.0001, ^0.1%VX2-TCM_HG group: p = 0.0001, and ^0.1%HepG2-TCM_HG group: p < 0.0001). TCMs also induced M2 polarization under normoxic conditions, but the strongest M2 polarization occurred when both tumor cells and macrophages were incubated under hypoxia with high glucose levels. Metabolomics revealed that several metabolites, particularly lactate, were correlated with hypoxia and M2 polarization. Under normoxia, elevated 2-amino-butanoic acid (2A-BA) strongly correlated with M2 polarization. These findings suggest that targeting tumor hypoxia could mitigate immune evasion in liver tumors. Lactate drives acidity in hypoxic tumors, while 2A-BA could be a therapeutic target for overcoming immunosuppression in normoxic conditions.

Keywords: 2-amino-butanoic acid (2A-BA); GC-MS-based metabolic profiling; M2-like polarization; hepatocellular carcinoma (HCC); hypoxia.

Figure 1

Flowchart summarizing the study design.

Figure 2

Viability assay for liver cancer cells (HepG2, VX2) and macrophages (THP-MΦ) incubated under normoxia or hypoxia with complete medium (blue box) or high-glucose complete medium (red box); all cell lines share the same legend. (a) HepG2 cell (b) VX2, and (c) THP-MΦ proliferation is displayed over 96 h. Three-way ANOVA was used for this analysis. ****, p < 0.0001 represents comparisons of the cell number changes over time within the groups (normoxic or hypoxic conditions) and ####, p < 0.0001 represents comparisons of the two groups for normoxia and hypoxia.

Figure 3

Longitudinal extracellular pH (pHe) measurements in the media of liver cancer cells (HepG2, VX2) and macrophages (THP-MΦ) incubated under normoxia (gray line) or hypoxia (orange line) with complete or high-glucose complete medium. (a) HepG2, VX2 (b), and THP-MΦ (c) keep the same design and share the legend. Three-way ANOVA was used for this analysis. ***, p < 0.001 and ****, p < 0.0001 are indicative of the difference between different time points; ####, p < 0.0001 is indicative of the difference between hypoxia and normoxia.

Figure 4

Peak area of lactic acid in the liver cancer tumor microenvironment (TME) based on GC-MS-based metabolic profiling. The left Y-axis shows the different types of tumor-conditioned mediums (TCMs). The direction of the arrow indicates an increase in peak area. Briefly, lactate concentrations were higher in the TCMs compared to the complete medium, and higher in the TCMs derived from hypoxic cancer cells compared to the TCMs derived from normoxic cancer cells, independent of glucose availability.

Figure 5

Heat map of the Spearman correlation matrix of the liver cancer tumor microenvironment (TME) GC-MS-based metabolic profiling. Two liver cancer cell lines were included in this cohort.

Figure 6

Violin plots showing arginase assay results for macrophages (THP-MΦ cells) incubated with different tumor-conditioned medium (TCM) types. The dotted line indicates the mean value. “Hypoxia (orange) or Normoxia (gray)” in the legend indicates if macrophages were cultured under 0.1% O₂ (hypoxia) or 21% O₂ (normoxia) in combination with different TCMs (see X-axis). Two-way ANOVA was used for this analysis. #, p < 0.05 and ##, p < 0.01: comparisons between two groups; *, p < 0.05 and ****, p < 0.0001: comparisons within a group.

Figure 7

(a) Violin plots showing viability assay results for macrophages (THP-MΦ) incubated with increasing concentrations of 2-amino-butanoic acid (2A-BA) for 24 and 48 h. Two-way ANOVA was used for this analysis. ####, p < 0.0001: comparisons between two incubation time groups (24 h vs. 48 h). (b) Arginase assay for macrophages (THP-MΦ) incubated with increasing concentrations of 2-amino-butanoic acid (2A-BA). Kruskal–Wallis test was used for this analysis. p = 0.02: comparisons within a group of different concentrations of 2A-BA. (a,b) Plots share a legend and the gradual deepening of the blue color represents a higher concentration of 2A-BA.