Upregulation of capacity for glutathione synthesis in response to amino acid deprivation: regulation of glutamate-cysteine ligase subunits

Abstract

Using HepG2/C3A cells and MEFs, we investigated whether induction of GSH synthesis in response to sulfur amino acid deficiency is mediated by the decrease in cysteine levels or whether it requires a decrease in GSH levels per se. Both the glutamate-cysteine ligase catalytic (GCLC) and modifier (GCLM) subunit mRNA levels were upregulated in response to a lack of cysteine or other essential amino acids, independent of GSH levels. This upregulation did not occur in MEFs lacking GCN2 (general control non-derepressible 2, also known as eIF2α kinase 4) or in cells expressing mutant eIF2α lacking the eIF2α kinase Ser(51) phosphorylation site, indicating that expression of both GCLC and GCLM was mediated by the GCN2/ATF4 stress response pathway. Only the increase in GCLM mRNA level, however, was accompanied by a parallel increase in protein expression, suggesting that the enhanced capacity for GSH synthesis depended largely on increased association of GCLC with its regulatory subunit. Upregulation of both GCLC and GLCM mRNA levels in response to cysteine deprivation was dependent on new protein synthesis, which is consistent with expression of GCLC and GCLM being mediated by proteins whose synthesis depends on activation of the GCN2/ATF4 pathway. Our data suggest that the regulation of GCLC expression may be mediated by changes in the abundance of transcriptional regulators, whereas the regulation of GCLM expression may be mediated by changes in the abundance of mRNA stabilizing or destabilizing proteins. Upregulation of GCLM levels in response to low cysteine levels may serve to protect the cell in the face of a future stress requiring GSH as an antioxidant or conjugating/detoxifying agent.

Figure 1

Effect of sulfur amino acid deficiency on GCLC and GCLM expression in rat liver and in HepG2/C3A cells. A. GCLC and GCLM protein and mRNA abundance in liver of rats fed a diet that contained 100 g soy protein isolate and no supplemental sulfur amino acids for 7 days. B. GCLC and GCLM protein and mRNA abundance in cells cultured in medium with 0.1 mM L-methionine and no cyst(e)ine for 30 h. Values are expressed as fold of control values for rats fed a methionine-supplemented diet or for cells cultured in cysteine-supplemented medium. GCLC and GCLM protein levels were obtained by western blotting, and mRNA levels were obtained by quantitative PCR. Bars indicated by an asterisk represent values significantly greater than control values (P≤0.05).

Figure 2

Thiol and GCLM mRNA levels in cells null for GCLC. GCLC-null MEFs were cultured in cysteine-deficient medium (0.1 mM L-methionine) with or without addition of 0.3 mM L-cysteine and/or 0.5 mM glutathione. Bars not labeled with the same letter represent values that are significantly different at P≤0.05 by ANOVA and Tukey’s comparison test.

Figure 3

Thiols (A) and GCLC and GCLM mRNA (B) in HepG2/C3A cells cultured for 20 h in cysteine-deficient medium (basal containing 0.1 mM methionine) with/without addition of 0.3 mM cysteine or 0.3 mM other thiol compounds. Bars labeled with an asterisk are significantly different than values for the cysteine-deficient medium at P≤0.05 by ANOVA and Dunnett’s comparison test.

Figure 4

Comparison of effects of various amino acid deficiencies on eIF2α phosphorylation and GCLC and GCLM expression in HepG2/C3A cells. A. Western blots showing eIF2α phosphorylation in cells cultured in medium deficient in leucine, cysteine or methionine and an increase in GCLM protein abundance in cells cultured in cysteine-deficient medium. B. GCLC and GCLM mRNA abundance in cells cultured for 12, 24, or 36 h in medium lacking cysteine, leucine, or histidine + arginine. mRNA abundance was determined by quantitative PCR and is shown as fold of the control (complete medium) value. Values significantly different than the control value for a particular time point are indicated by an asterisk.

Figure 5

Effect of deprivation of wildtype, Gcn2 knockout, and eIF2α(Ala/Ala) mutant MEFs of histidine or methionine for 12 h on GCLC and GCLM mRNA levels. An asterisk above the bar indicates the value is different (P ≤ 0.05) compared to the value (set as 1.0) for the same cells cultured in complete medium.

Figure 6

Effect of actinomycin D on GCLC mRNA (A) and GCLM mRNA (B) abundance in HepG2/C3A cells. Actinomycin D was added with treatment media at time 0, and 2 h was allowed for actinomycin D to inhibit mRNA synthesis. All media contained 0.1 mM methionine. +Cys medium contained 0.3 mM cysteine. +ActD medium contained 10 µg/mL actinomycin D. Changes in mRNA levels were measured between 2 h and 6 h.

Figure 7

Effect of cycloheximide on GCLC mRNA (A) and GCLM mRNA (B) abundance in HepG2/C3A cells. Cycloheximide was added with treatment media at time 0 and 2 h was allowed for cycloheximide to inhibit protein synthesis. All media contained 0.1 mM methionine. +Cys medium contained 0.3 mM cysteine. +CHx medium contained 10 µg/mL cycloheximide. Changes in mRNA levels were measured between 2 h and 6 h.