β-Catenin/c-Myc Axis Modulates Autophagy Response to Different Ammonia Concentrations

Abstract

Ammonia a by-product of nitrogen containing molecules is detoxified by liver into non-toxic urea and glutamine. Impaired ammonia detoxification leads to hyperammonemia. Ammonia has a dual role on autophagy, it acts as inducer at low concentrations and as inhibitor at high concentrations. However, little is known about the mechanisms responsible for this switch. Wnt/β-catenin signalling is emerging for its role in the regulation of ammonia metabolizing enzymes and autophagosome synthesis through c-Myc. Here, using Huh7 cell line, we show a modulation in c-Myc expression under different ammonia concentrations. An increase in c-Myc expression and in its transcriptional regulator β-catenin was detected at low concentrations of ammonia, when autophagy is active, whereas these modifications were lost under high ammonia concentrations. These observations were also recapitulated in the livers of spf-ash mice, a model of constitutive hyperammonaemia due to deficiency in ornithine transcarbamylase enzyme. Moreover, c-Myc-mediated activation of autophagy plays a cytoprotective role in cells under ammonia stress conditions as confirmed through the pharmacological inhibition of c-Myc in Huh7 cells treated with low ammonia concentrations. In conclusion, the unravelled role of c-Myc in modulating ammonia induced autophagy opens new landscapes for the development of novel strategies for the treatment of hyperammonemia.

Keywords: autophagy; c‐Myc; hyperammonemia; liver.

Figure 1

A) (a–d) Huh7 cells were cultured in presence or absence of the indicated drugs (NH₄Cl 0.5 × 10⁻³ m; NH₄Cl 5 × 10⁻³ m; Rapamycin 1 × 10⁻⁶ m) for 48 h and then stained with monodansylcadaverine to mark autophagosomes (in blue) and lysotracker red for lysosomes (in red). Scale bar is 25 µm. (e–h) Merge images from both channels are shown at the bottom where autolysosomes are shown in pink. Scale bar is 25 µm. (e′–h′) Zoom images, scale bar is 10 µm. Image correlation analysis was performed through ImageJ Jacop. PCC = Pearson's correlation coefficient. −1 < PCC < 0 = negative correlation; PCC = 0 no correlation; 0 < PCC ≤ 1 = positive correlation. The mean correlation coefficient value ± s.d. of n = 3 images is shown on the merge images. B) Expression levels of autophagy markers SQSTM1/p62, Beclin1, and LC3 in Huh7 cells, after 48 h of treatment with NH₄Cl 0.5/5 × 10⁻³ m and Rapamycin 1 × 10⁻⁶ m. GADPH is used as loading control. Samples for the detection of SQSTM1/p62 and LC3II were run on the same gel but were not contiguous (Figure S1, Supporting Information). C) Densitometric quantification for SQSTM1/p62, LC3II, and SQSTMI:BECL1 (n = 3). The results were presented as means ± standard deviation; values were compared to CTR by one‐way ANOVA following Tukey test. *p < 0.0332, **p < 0.0021, ***p < 0.0002 in comparison to CTR; ^# p < 0.0332, ^## p < 0.0021, ^### p < 0.0002 in comparison to NH₄Cl 0.5 × 10⁻³ m; ⁺ p < 0.0332, ⁺⁺ p < 0.0021, ⁺⁺⁺ p < 0.0002 in comparison to NH₄Cl 5 × 10⁻³ m.

Figure 2

Huh7 cells were treated for 48 h with the indicated concentration of ammonia and rapamycin. Bafilomycin A1 (100 × 10⁻⁹ m) was added 4 h before to perform the staining with mondansylcadaverine for autophagosomes (in blue) and lysotraker red for lysosomes (in red). Scale bar is 10 µm. In panel b′, e′, h′, and m′ color scale has a maximum value of 154 A.U. against the 255 A.U. used in the other panels, to better display Lysotraker red fluorescence signal that could be affected by Bafilomycin A1 treatment. See Figure S2 (Supporting Information) for original images. Image correlation analysis was performed through ImageJ Jacop. PCC = Pearson's correlation coefficient. −1 < PCC < 0 = negative correlation; PCC = 0 no correlation; 0 < PCC ≤ 1 = positive correlation. The mean correlation coefficient value ± s.d. of n = 3 images is shown on the merge images.

Figure 3

A,B) Western blot analyses and densitometric quantification of c‐Myc in Huh7 cells treated with increased concentration of NH₄Cl for 24 h. GADPH is used as loading control (n = 3); C,D) Western blot analyses and densitometric quantification of c‐Myc, SQSTM1/P62, CyclinD1, LC3I/II in Huh7 cells treated with NH₄Cl 0.5 × 10⁻³ m and/or c‐Myc inhibitor 100 × 10⁻⁶ m (10058‐F4) for 48 h. GADPH, actin, and α‐tubulin are used as loading control (n = 3). The results were presented as means ± standard deviation; values were compared to CTR by one‐way ANOVA following Tukey test, *p < 0.0332, **p < 0.0021, ***p < 0.0002 in comparison to CTR; ^# p < 0.0332, ^## p < 0.0021, ^### p < 0.0002 in comparison to NH₄Cl 0.5 × 10⁻³ m; ⁺ p < 0.0332, ⁺⁺ p < 0.0021, ⁺⁺⁺ p < 0.0002 in comparison to NH₄Cl 5 × 10⁻³ m.

Figure 4

A) (a–n) Huh7 cells were transduced with 40 particles per cell of the Premo Autophagy Tandem Sensor and incubated overnight ≥16 h. Cells were then incubated with NH₄Cl 0.5 × 10⁻³ m and/or c‐Myc I for 48 h. Cells were imaged using standard FITC/TRITC filters sets. Scale bar: 25 µm. f′, i′, n′ are zoom images. B) Rational of Premo Autophagy Tandem Sensor RFP‐GFP‐LC3B kit for monitoring autophagic flux. By combining an acid‐sensitive GFP with an acid‐insensitive RFP, the change from autophagosome (neutral pH) to autolysosome (with an acid pH) can be visualized by imaging the specific loss of the GFP fluorescence, leaving only red fluorescence. Image shows in yellow autophagosomesand in red autolysosomes. C) Puncta of autophagosomes (yellow) and autolysosomes (red) per cell were measured using ImageJ software. The mean value ± s.d. of at least five cells is shown on the plots (n = 3). The results were presented as means ± standard deviation; values were compared to CTR by one‐way ANOVA following Tukey test, *p < 0.0332, **p < 0.0021, ***p < 0.0002 in comparison to CTR; ^# p < 0.0332, ^## p < 0.0021, ^### p < 0.0002 in comparison to NH₄Cl 0.5 × 10⁻³ m. Image correlation analysis was performed through ImageJ Jacop. PCC = Pearson's correlation coefficient. −1 < PCC < 0 = negative correlation; PCC = 0 no correlation; 0 < PCC ≤ 1 = positive correlation. The mean correlation coefficient value ± s.d. of n = 3 images is shown on the merge images.

Figure 5

A,B) Western blot analyses and densitometric quantification of non‐phospho‐β‐catenin, phospho‐β‐catenin, c‐Myc, phospho‐GSK3β (ser9), GSK3β, APC, TCF4 in Huh7 cells after 48 h of treatment with NH₄Cl 0.5 × 10⁻³ m and 5 × 10⁻³ m. GADPH is used as loading control (n = 3). The results were presented as means ± standard deviation; values were compared to CTR by one‐way ANOVA following Tukey test, *p < 0.0332, **p < 0.0021, ***p < 0.0002 in comparison to CTR; ^# p < 0.0332, ^## p < 0.0021, ^### p < 0.0002 in comparison to NH₄Cl 0.5 × 10⁻³ m; ⁺ p < 0.0332, ⁺⁺ p < 0.0021, ⁺⁺⁺ p < 0.0002 in comparison to NH₄Cl 5 × 10⁻³ m. C,D) Western blotting and densitometric analysis of phospho‐β‐catenin, non‐phospho β‐catenin, c‐Myc and of the autophagy markers SQSTM1/p62, LC3 in liver of wild type or spf‐ash mice (n = 3). The results were presented as means ± standard deviation; values were compared to CTR through Student's t‐test, *p < 0.05, **p < 0.01.

Figure 6

A (a–i) Huh7 cells were cultured in presence or absence of the indicated drugs (NH₄Cl 0.5 × 10⁻³ m; 3‐methyladenine 2 × 10⁻³ m; chloroquine 10 × 10⁻⁶ m, 10058‐F4 100 × 10⁻⁶ m) for 24 h and then stained with the LIVE/DEAD kit and observed at the fluorescence microscope. Image shows in blue live cells, and in green dead cells. Scale bar is 100 µm. B) Live/dead cell viability quantification: number of dead cells was calculated using ImageJ, data shown are mean ± SD, n = 3. The results were presented as means ± standard deviation; values were compared to CTR by one‐way ANOVA following Tukey test, *p < 0.0332, **p < 0.0021, ***p < 0.0002 in comparison to CTR; ^# p < 0.0332, ^## p < 0.0021, ^### p < 0.0002 in comparison to NH₄Cl 0.5 × 10⁻³ m; ⁺ p < 0.0332, ⁺⁺ p < 0.0021, ⁺⁺⁺ p < 0.0002 in comparison to NH₄Cl 5 × 10⁻³ m. C) Illustration of autophagy inhibitors mechanisms of action: 3MA (3‐methyladenine) inhibits autophagosomes maturation by inhibiting phosphatidylinositol‐4,5‐bisphosphate 3‐kinase (PI3K) activity, while CQ (chloroquine) could block autolysosomes fusion, instead c‐Myc inhibition impairs autophagosomes formation, while NH4Cl 5 × 10⁻³ m blocks autophagosomes degradation.