Antioxidant Protection of NADPH-Depleted Oligodendrocyte Precursor Cells Is Dependent on Supply of Reduced Glutathione

Abstract

The pentose phosphate pathway is the main source of NADPH, which by reducing oxidized glutathione, contributes to antioxidant defenses. Although oxidative stress plays a major role in white matter injury, significance of NADPH for oligodendrocyte survival has not been yet investigated. It is reported here that the NADPH antimetabolite 6-amino-NADP (6AN) was cytotoxic to cultured adult rat spinal cord oligodendrocyte precursor cells (OPCs) as well as OPC-derived oligodendrocytes. The 6AN-induced necrosis was preceded by increased production of superoxide, NADPH depletion, and lower supply of reduced glutathione. Moreover, survival of NADPH-depleted OPCs was improved by the antioxidant drug trolox. Such cells were also protected by physiological concentrations of the neurosteroid dehydroepiandrosterone (10(-8) M). The protection by dehydroepiandrosterone was associated with restoration of reduced glutathione, but not NADPH, and was sensitive to inhibition of glutathione synthesis. A similar protective mechanism was engaged by the cAMP activator forskolin or the G protein-coupled estrogen receptor (GPER/GPR30) ligand G1. Finally, treatment with the glutathione precursor N-acetyl cysteine reduced cytotoxicity of 6AN. Taken together, NADPH is critical for survival of OPCs by supporting their antioxidant defenses. Consequently, injury-associated inhibition of the pentose phosphate pathway may be detrimental for the myelination or remyelination potential of the white matter. Conversely, steroid hormones and cAMP activators may promote survival of NADPH-deprived OPCs by increasing a NADPH-independent supply of reduced glutathione. Therefore, maintenance of glutathione homeostasis appears as a critical effector mechanism for OPC protection against NADPH depletion and preservation of the regenerative potential of the injured white matter.

Keywords: demyelination; metabolism; oligodendrocytes; oxidative stress; pentose phosphate pathway; white matter.

Figure 1.

The NADPH antimetabolite 6-amino-NADP (6AN) is toxic to oligodendrocyte precursor cells and oligodendrocyte precursor cell-derived oligodendrocytes. Undifferentiated OPCs (a–f), oligodendrocytes (g), and Schwann cells (h) were treated with 6AN as indicated. (a) Representative phase contrast micrographs depict declining density of OPC cultures that were exposed to 6AN for 24 h. (b) MTT cell survival assays revealed declining number of viable OPCs in response to such treatment. (c) LDH release assay revealed extensive plasma membrane permeabilization as early as 6 h after adding 6AN. Such a response suggests that necrosis is the major cause of reduced viability in 6AN-treated OPCs. (d) Western blot with an antibody specific for the activated form of the apoptotic protease caspase-3 (cleaved caspase-3, CC3) revealed caspase activation in response to the DNA damaging drug etoposide (1 μM) but not 50 μM 6AN. Equal loading was confirmed by reprobing the membrane with an antibody against GAPDH. (e) and (f) Cells that were treated with 6AN for 6 h were loaded with the mitochondrial potential sensor TMRM. (e) In vehicle-treated control cells, red fluorescence of TMRM in the perikaryal region reflects distribution of functional mitochondria as confirmed by no signal in cells that were treated with the mitochondria uncoupling chemical FCCP (1 μM, data not shown). TMRM fluorescence was reduced in 6AN-treated cells. (f) Quantification of TMRM fluorescence intensity revealed a 60% decline following 6AN exposure. After 2 h treatment with 6AN, TMRM fluorescence was similar as in vehicle-treated cells (not shown). (g) Declining viability was also observed in OPC-derived oligodendrocytes that were treated with 6AN for 24 h. In such cultures, Western blot revealed sharply reduced MBP expression suggesting a high sensitivity of maturing oligodendrocytes to PPP inhibition. Equal loading of the blot was confirmed by reprobing of the membrane for GAPDH. (f) Primary mouse Schwann cells were less sensitive to PPP inhibition than OPCs as revealed by the MTT assay at 72 h after initiation of 6AN treatment. These differences are unlikely due to differential sensitivity of mouse versus rat cells as mouse brain OPCs were as sensitive to 6AN as rat spinal cord OPCs (not shown). In (b), (c), (g), (h), data represent averages ± SD of nine sister cultures from three independent experiments; in (f), averages ± SD of at least 54 cells from three independent experiments are shown; ns, p > .05, ***p < .001 (analysis of variance).

Figure 2.

The NADPH antimetabolite 6-amino-NADP (6AN) compromises antioxidant defenses and induces oxidative cytotoxicity in oligodendrocyte precursor cells. Rat OPCs were treated as indicated; in (d–f), treatment times were 24 h (d–e) or 12 h (f). (a–b) Measurements of NADPH/NADP + and GSH/GSSG ratios. As early as 2 h after treatment with 50 μM 6AN, both ratios were dramatically reduced. (c) NBT assay revealed increased superoxide levels at 2 h after adding 6AN. Lack of effect at 6 h is likely due to damage of mitochondria which are the main source of superoxide in most cells (Figure 1(e)–(f)). Hence, compromised supply of antioxidant molecules NADPH and GSH resulted in oxidative stress preceding 6AN-induced cell death. (d–f) The antioxidant drug trolox reduced OPC toxicity of 6AN. (d) Representative phase contrast micrographs reveal that after a 24 h cotreatment the 6AN-induced decline in culture density was attenuated by 100 μM trolox. (e) In 6AN-treated OPCs, MTT survival assays showed protective effects of 100 but not 30 μM trolox. (f) LDH release assay indicated that 6AN-induced permeabilization of the plasma membrane was reduced by 100 μM Trolox. Data represent averages ± SD of three independent experiments (a, b) or nine sister cultures from three independent experiments (c), (e), (f); ns, p > .05; *p < .05; ***p < .001 (U test in (a), (b); analysis of variance in (c), (e), (f)).

Figure 3.

At G6PDH-inhibiting concentrations, dehydroepiandrosterone induces oxidative toxicity in oligodendrocyte precursor cells. OPCs were treated as indicated; treatment time in (a), and (c), was 24 h. (a) MTT cell survival assays revealed declining number of viable OPCs in response to 100 or 300 μM DHEA. (b) As in the case of 6AN, 2 h treatment with 100 μM DHEA increased superoxide levels. Normalized levels of superoxide at 6 h suggest mitochondrial dysfunction. (c) The antioxidant drug trolox protected against DHEA cytotoxicity. Data represent averages ± SD of nine sister cultures from three independent experiments; ns, p > .05, *p < .05, ***p < .001 (analysis of variance in (a), (c); U test in (b)).

Figure 4.

At physiologically relevant concentrations, dehydroepiandrosterone protects NADPH-depleted oligodendrocyte precursor cells. Cells were treated for 24 h as indicated. (a) Representative phase contrast micrographs of OPCs that were treated with low concentration of DHEA (10 nM) in combination with 50 μM 6AN. (b) MTT viability assays revealed that cotreatment with 6AN and low concentrations of DHEA reduced cytotoxicity of 6AN. (c) However, DHEA was ineffective against cytotoxicity of tunicamycin which kills OPCs by inducing ER stress. Data represent averages ± SD of nine sister cultures from three independent experiments, ns, p > .05, **p < .01; ***p < .001 (analysis of variance).

Figure 5.

Requirement of glutathione synthesis for protection of NADPH-depleted oligodendrocyte precursor cells. OPCs were treated as indicated for 6- ((a–b) or 24 h (c–f)). (a) DHEA did not prevent decline of the NADPH/NADP⁺ ratio after PPP inhibition with 6AN. (b) DHEA increased the GSH/GSSG ratio under basal conditions and partially attenuated its decrease in response to 6AN. (c) MTT survival assay revealed that DHEA protection of 6AN-treated OPCs was abolished in the presence of the glutathione synthase inhibitor BSO. (d) Forskolin protected PPP-inhibited OPCs. The protection was sensitive to the PKA inhibitor H89 and the glutathione synthase inhibitor BSO. (e) G1, a selective agonist of GPER, protected 6AN-treated OPCs in a glutathione synthase-dependent manner; conversely, PKA was not involved in G1-mediated protection (data not shown). (f) Cotreatment with a GSH precursor, N-acetyl-L-cysteine (NAC) reduced 6AN toxicity. In (a–b), averages ±  SD of three independent experiments are shown; ns, p > .05; *p < .05 (U test); in (c–f), data represent averages ± SD of nine sister cultures from three independent experiments; ns, p > .05, **p < .01; ***p < .001 (analysis of variance).

Figure 6.

A hypothetical model proposing maintenance of glutathione homeostasis as a common effector for various interventions that support the survival of NADPH-depleted OPCs including physiological levels of DHEA, stimulation of GPER or activation of the cAMP pathway. Increased de-novo glutathione synthesis or its lower use may, at least temporarily, provide sufficient supply of reduced glutathione (GSH) despite disruption of the NDAPH-dependent reduction of oxidized glutathione (GSSG). In this way, the requirement of NADPH (and the PPP) for maintenance of redox homeostasis can be bypassed, and OPCs may survive the metabolic stress. We propose that such a protective mechanism may promote white matter sparing or repair following central nervous system injuries that are associated with oxidative stress-mediated death of OLs/OPCs. At least some of these insults may directly affect NADPH supply by lowering activity of PPP (see the Discussion section for more details).