Role of GLUT1 in regulation of reactive oxygen species

Abstract

In skeletal muscle cells, GLUT1 is responsible for a large portion of basal uptake of glucose and dehydroascorbic acid, both of which play roles in antioxidant defense. We hypothesized that conditions that would decrease GLUT1-mediated transport would cause increased reactive oxygen species (ROS) levels in L6 myoblasts, while conditions that would increase GLUT1-mediated transport would result in decreased ROS levels. We found that the GLUT1 inhibitors fasentin and phloretin increased the ROS levels induced by antimycin A and the superoxide generator pyrogallol. However, indinavir, which inhibits GLUT4 but not GLUT1, had no effect on ROS levels. Ataxia telangiectasia mutated (ATM) inhibitors and activators, previously shown to inhibit and augment GLUT1-mediated transport, increased and decreased ROS levels, respectively. Mutation of an ATM target site on GLUT1 (GLUT1-S490A) increased ROS levels and prevented the ROS-lowering effect of the ATM activator doxorubicin. In contrast, expression of GLUT1-S490D lowered ROS levels during challenge with pyrogallol, prevented an increase in ROS when ATM was inhibited, and prevented the pyrogallol-induced decrease in insulin signaling and insulin-stimulated glucose transport. Taken together, the data suggest that GLUT1 plays a role in regulation of ROS and could contribute to maintenance of insulin action in the presence of ROS.

Keywords: Antioxidant; Ataxia telangiectasia mutated (ATM); Glucose transport; Insulin resistance; Oxidative stress.

Fig. 1

Inhibition of GLUT1 increases ROS levels. L6 myoblasts were subjected to a DCFDA ROS assay in the presence or absence of (A) GLUT1 inhibitor, 80 α/4M fasentin with induction of ROS by 100 α/4M antimycin A (AntA), (B) 80 α/4M fasentin with 50 α/4M pyrogallol (PG), a superoxide generator, (C) GLUT1 inhibitor, 100 α/4M phloretin, with induction of ROS by 100 α/4M AntA, (D) 100 α/4M phloretin with 50 α/4M PG, and (E) 100 α/4M indinavir, which inhibits GLUT4, with induction of ROS by Ant-A or (F) by PG. (n = 6/group. Differences are statistically significant for all time points after time 0, p < 0.05. Subpanels E and F do not display significant differences, except 1F at 2 min, p < 0.05).

Fig. 2

GLUT1-S490 mutations affect ROS levels. L6 myoblasts transiently transfected with FLAG-GLUT1 constructs or FLAG-GLUT1-S490 point mutation constructs were loaded with the general ROS fluorescent probe, DCFDA (for all data except the right panel of A), or a purported superoxide probe that accumulates in mitochondria, MitoSox Red (right panel of A only), and fluorescent readings were taken. (A) Cells expressing wild-type GLUT1 (WT) vs. GLUT1-S490A under basal conditions, (B) with 100 α/4M Ant-A induction of ROS or (C) with induction of ROS by 50 α/4M PG. (D) WT vs. GLUT1-S490D mutants under basal conditions, (E) with 100 α/4M Ant-A induction of ROS, or (F) with induction of ROS by 50 α/4M PG. (n = 6/group, p < 0.05 for all panels. Differences are statistically significant for all time points after time 0, p < 0.05).

Fig. 3

GLUT1-S490D prevents increases in ROS when ATM is inhibited. L6 myoblasts were subjected to a DCFDA ROS assay in the presence or absence of (A) ATM inhibitor, 6 α/4M CP466722 (CP), (B) ATM inhibitor, 10 α/4M KU55933 (KU), (C) 6 α/4M CP with 100 α/4M Ant-A induction of ROS, and (D) 6 α/4M CP with induction of ROS by 50 α/4M PG. (E) L6 myoblasts transiently transfected with FLAG-GLUT1 (WT) or FLAG-GLUT1-S490D (S490D) constructs were treated plus and minus 10 α/4M KU. (n = 6/group, ^*p < 0.05 for all panels. Differences are statistically significant for all time points after time 0, p < 0.05, except for lines that visually overlap).

Fig. 4

GLUT1-S490A prevents the ROS-lowering effect of doxorubicin. L6 myoblasts underwent a DCFDA ROS assay in the presence or absence of ATM activator, 1 α/4M doxorubicin (DXR), and fluorescent readings were taken (A) under basal conditions and (B) after exposure to PG. (C) Myoblasts expressing FLAG-GLUT1(WT) and FLAG-GLUT1-S490A (S490A) were treated plus and minus 1 α/4M DXR, an ATM activator. (n = 6/group, ^*p < 0.05 for all panels. Differences are statistically significant for all time points after time 0, p < 0.05, except for lines that visually overlap).

Fig. 5

GLUT1-S490D prevents ROS-induced insulin resistance. L6 myoblasts transiently transfected with FLAG-GLUT1 (WT) constructs or FLAG-GLUT1-S490D (SD) point mutation constructs were treated plus and minus insulin (INS) and pyrogallol (PG), a superoxide generator, then subjected to (A) Western blot analysis probing for phosphorylated Akt S473 (p-Akt S473) and Akt as a loading control or (B) phosphorylated insulin receptor substrate 1 (p-IRS-1) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a loading control, and (C) a glucose transport assay (n = 3/group for the Western analysis and 6/group for the transport assay, ^*p < 0.05 compared to WT, ^†p < 0.05 compared to INS, and ^‡p< 0.05 compared to S490D for all panels).