Extracellular adenosine oppositely regulates the purinome machinery in glioblastoma and mesenchymal stem cells

Abstract

Glioblastoma (GB) is a lethal brain tumor that rapidly adapts to the dynamic changes of the tumor microenvironment (TME). Mesenchymal stem/stromal cells (MSCs) are one of the stromal components of the TME playing multiple roles in tumor progression. GB progression is prompted by the immunosuppressive microenvironment characterized by high concentrations of the nucleoside adenosine (ADO). ADO acts as a signaling molecule through adenosine receptors (ARs) but also as a genetic and metabolic regulator. Herein, the effects of high extracellular ADO concentrations were investigated in a human glioblastoma cellular model (U343MG) and MSCs. The modulation of the purinome machinery, i.e., the ADO production (CD39, CD73, and adenosine kinase [ADK]), transport (equilibrative nucleoside transporters 1 (ENT1) and 2 (ENT2)), and degradation (adenosine deaminase [ADA]) were investigated in both cell lines to evaluate if ADO could affect its cell management in a positive or negative feed-back loop. Results evidenced a different behavior of GB and MSC cells upon exposure to high extracellular ADO levels: U343MG were less sensitive to the ADO concentration and only a slight increase in ADK and ENT1 was evidenced. Conversely, in MSCs, the high extracellular ADO levels reduced the ADK, ENT1, and ENT2 expression, which further sustained the increase of extracellular ADO. Of note, MSCs primed with the GB-conditioned medium or co-cultured with U343MG cells were not affected by the increase of extracellular ADO. These results evidenced how long exposure to ADO could produce different effects on cancer cells with respect to MSCs, revealing a negative feedback loop that can support the GB immunosuppressive microenvironment. These results improve the knowledge of the ADO role in the maintenance of TME, which should be considered in the development of therapeutic strategies targeting adenosine pathways as well as cell-based strategies using MSCs.

Keywords: adenosine; glioblastoma; mesenchymal stem cells (MSC); purine metabolism; tumor microenvironment (TME).

FIGURE 1

Effects of ADO on the purinome machinery gene expression in U343MG and MSCs. (A) Schematic representation of the enzymes and transporters implicated in the ADO metabolism; in brackets, the gene name is reported. U343MG (left) or MSCs (right) were treated with ADO (100 nM or 10 μM) or with NECA (100 nM) and real‐time PCR was performed after 48 h for (B) ADA, (C) ENTDP1, (D) NT5E, (E) ADK, (F) SCL29A1, and (G) SCL29A2. Data are expressed as the fold of change versus control set to 1 and are the mean values ± SD of three different independent experiments. For the statistical analysis, one‐way ANOVA was performed, followed by Bonferroni's post hoc test: *p < 0.05, **p < 0.01 vs CTRL.

FIGURE 2

ADO increased the protein expression of ADK and ENT1 in U343MG. U343MG cells were treated with ADO (100 nM or 10 μM) or with NECA (100 nM) for 48 h. Then, Western Blot was performed for ADK, ENT1, or ENT2: (A) a representative western blot image was reported and (B) the densitometric analysis was performed. Data are normalized to the total protein (Figure S2) and are expressed as the percentage of optical density versus control set to 100% ± SD of three different experiments. (C) Representative images of the immunofluorescence analysis. Scale bar: 200 μm. (D) Cells treated as above were fixed and incubated with ADK, ENT1, or ENT2 primary antibodies overnight; the day after Alexa Flu‐or 488‐conjugated secondary antibodies were used and the fluorescence was quantified. Data are expressed as the mean fluorescence intensity (AU) and are the mean values ± SD of two different experiments, each performed in triplicate. The significance was determined by one‐way ANOVA, followed by Bonferroni's post hoc test: *p < 0.05, **p < 0.01, ***p < 0.001 vs CTRL.

FIGURE 3

ADO decreased the protein expression of ADK and ENT1 in MSCs. MSCs cells were treated with ADO (100 nM or 10 μM) or with NECA (100 nM) for 48 h. Then, Western Blot was performed for ADK, ENT1, or ENT2: (A) a representative western blot image was reported and (B) the densitometric analysis was performed. Data are normalized to the total protein (Figure S4) and expressed as the percentage of optical density versus control set to 100% ± SD of three different experiments. (C) Representative images of the immunofluorescence analysis. Scale bar: 200 μm. (D) Cells treated as above were fixed and incubated with ADK, ENT1, or ENT2 primary antibodies overnight; the day after Alexa Flu‐or 488‐conjugated secondary antibodies were used and the fluorescence was quantified. Data are expressed as the mean fluorescence intensity (AU) and the mean values ± SD of two different experiments, each performed in triplicate. The significance was determined by one‐way ANOVA, followed by Bonferroni's post hoc test: *p < 0.05, **p < 0.01, ***p < 0.001 vs CTRL.

FIGURE 4

Time‐course analysis of ERK and JNK phosphorylation in U343MG and MSCs. (A,B) U343MG or (C,D) MSCs were treated for 1, 5, 10, 30, or 60 min with ADO (100 nM or 10 μM) or NECA (100 nM). Phosphorylation levels were assessed with the immune‐enzymatic assay. Data are normalized to the cell number and reported as percentages versus the CTRL set to 100% ± SD of three different experiments. The significance was determined by a two‐way ANOVA, followed by Dunnett's post hoc test: *p < 0.05, **p < 0.01, ***p < 0.001 vs basal levels for ADO 100 nM; #p < 0.05, ##p < 0.01, ###p < 0.001 vs basal levels for ADO 10 μM; $p < 0.05, $$p < 0.01 vs basal levels for NECA 100 nM.

FIGURE 5

Effect of ADO on MSC primed with U343MG‐conditioned medium. (A) Schematic representation of the treatment workflow. U343MG cells were starved for 48 h, and the conditioned medium was used to treat MSCs for 72 h (100% or 50% of the total medium). Then, MSCs were treated for a further 48 h with complete medium (CTRL), or with different percentages of CM in the absence (CTRL‐CM) or presence of ADO (100 nM or 10 μM) or NECA (100 nM). The immunofluorescence assay was used to evaluate the expression of (B) ADK, (C) ENT1, or (D) ENT2. Data are the mean ± SD of two experiments each performed in triplicate and are expressed as the mean fluorescence intensity (AU). The significance was determined by one‐way ANOVA, followed by Bonferroni's post hoc test: *p < 0.05, **p < 0.01, ***p < 0.001 vs CTRL.

FIGURE 6

Effect of MSC‐U343MG interplay in co‐culture on ADO response. (A) Schematic representation of the treatment workflow. (B) MSCs were co‐cultured for 72 h with U343MG in a 1:3 ratio or (C) for a further 48 h. At the end, the real‐time PCR on ADK, SCL29A1, and SCL29A2. Data are expressed as the fold of change versus control set to 1 and are the mean values ± SD of two independent experiments. For the statistical analysis, a parametric t‐test was performed: *p < 0.05, ***p < 0.001 vs CTRL. (D) The immunofluorescence assay was used to evaluate the expression of ADK, ENT1, or ENT2. Data are the mean ± SD of two experiments each performed in triplicate and are expressed as the relative mean fluorescence intensity with respect to CTRL set to 1. The significance was determined by one‐way ANOVA, followed by Bonferroni's post hoc test: *p < 0.05, **p < 0.01 vs CTRL.