Glutaredoxin regulates vascular development by reversible glutathionylation of sirtuin 1

Abstract

Embryonic development depends on complex and precisely orchestrated signaling pathways including specific reduction/oxidation cascades. Oxidoreductases of the thioredoxin family are key players conveying redox signals through reversible posttranslational modifications of protein thiols. The importance of this protein family during embryogenesis has recently been exemplified for glutaredoxin 2, a vertebrate-specific glutathione-disulfide oxidoreductase with a critical role for embryonic brain development. Here, we discovered an essential function of glutaredoxin 2 during vascular development. Confocal microscopy and time-lapse studies based on two-photon microscopy revealed that morpholino-based knockdown of glutaredoxin 2 in zebrafish, a model organism to study vertebrate embryogenesis, resulted in a delayed and disordered blood vessel network. We were able to show that formation of a functional vascular system requires glutaredoxin 2-dependent reversible S-glutathionylation of the NAD(+)-dependent protein deacetylase sirtuin 1. Using mass spectrometry, we identified a cysteine residue in the conserved catalytic region of sirtuin 1 as target for glutaredoxin 2-specific deglutathionylation. Thereby, glutaredoxin 2-mediated redox regulation controls enzymatic activity of sirtuin 1, a mechanism we found to be conserved between zebrafish and humans. These results link S-glutathionylation to vertebrate development and successful embryonic angiogenesis.

Keywords: cardiovascular system; proteomics.

Fig. 1.

zfGrx2 is essential for development of the vasculature. (A) Confocal microscopy of fli:EGFP transgenic embryos with or without zfGrx2 at 35 and 48 h postfertilization (hpf). DA, dorsal aorta; DLAV, dorsal longitudinal anastomotic vessel; ISV, intersegmental vessel; PC, parachordal sprouts. (Scale bars, 100 µm.) (B) Statistical analysis of A shows mean ± SEM, significance (***P < 0.01) was calculated by two-tailed Student’s t test, 50 pg WT or mutated zfGrx2 mRNA were injected per single egg for rescue experiments. (C) Statistical analysis of intersegmental vessel formation in fli:EGFP embryos, zfGrx2 knockdown embryos, and zfGrx2 knockdown embryos rescued with 15 ng/embryo zfSIRT1 mRNA (48 hpf) (Fig. S2), mean ± SEM, significance (*P < 0.05) was calculated by two-tailed Student’s t test. (D) Quantitative real-time PCR of zfSirt1 transcripts in WT and zfGrx2 knockdown embryos 48 hpf. Presented are mean ± SEM of triplicates of two independent experiments.

Fig. 2.

Grx2 regulates enzymatic activity of SIRT1. SIRT1 activity (total histone deacetylase activity) was measured in 40 µg protein extracts obtained from WT embryos, zfGrx2 knockdown embryos, and embryos rescued with either zfGrx2 or zfSIRT1 48 hpf (A, mean ± SEM, duplicate measurements of two independent experiments with pooled extracts of ∼200 embryos per experiment per condition) or 50 µg protein extracts of control HeLa cells, hGrx2c overexpressing HeLa cells (Grx2c⁺), and control extract incubated with 50 µM recombinant hGrx2c (mean ± SEM, n = 3) (B). (C) Relative enzymatic activity of 10 µg recombinant zfSIRT1 after S-glutathionylation and following 30 min of incubation with 5× molar excess of prereduced zfGrx2 in relation to zfSIRT1 (mean ± SEM, n = 3). Significance (*P < 0.05, **P < 0.01) was calculated by two-tailed Student’s t test.

Fig. 3.

zfGrx2 regulates reversible S-glutathionylation of zfSIRT1. (A) Ten micromolars zfSIRT1, S-glutathionylated with 0.5 mM fluorescent Eosin-Di-GSSG (1) was incubated for 10 min with prereduced zfGrx2 in single turnover conditions [50 (2) and 100 µM (3)] or in catalytic conditions [0.1 µM zfGrx2/1 mM GSH (4)] and applied to a SDS/PAGE. Bars represent the densitometric analyses for GSH bound to zfSIRT1 of two independent experiments, mean ± SEM. (B) Ten micromolars S-glutathionylated zfSIRT1was incubated without (1) or with prereduced zfGrx2 (2.5× molar excess) for 5 (2), 1 (3), and 0.5 (4) min. Samples were separated by SDS/PAGE, and relative intensity was determined after densitometric analyses for GSH bound to zfSIRT1 of two independent experiments were plotted against incubation time, mean ± SEM. (C) Ten micromolars S-glutathionylated zfSIRT1 (1) was incubated for 10 min with 0.1 µM zfGrx2 (2), zfGrx2C37S (3), or zfGrx2C40S (4). Bars represent the densitometric analyses for GSH bound to zfSIRT1 of two independent experiments, mean ± SEM. GSSG, fluorescence of SIRT1 after incubation with Di-Eosin-GSSG; SIRT1/Grx2, Coomassie staining of the respective proteins.

Fig. 4.

One specific cysteine residue of zfSIRT1 is target for zfGrx2-dependent deglutathionylation. (A) Quantification of zfGrx2-dependent deglutathionylation of the four identified S-glutathionylated zfSIRT1 peptides by mass spectrometry using NEM and its derivative NEMD5 (mean ± SEM, n = 6). (B) Direct quantification of glutathionylated peptide of recombinant zfSIRT1 in reduced and S-glutathionylated protein before and after incubation with 5× molar excess of zfGrx2 for 15 min. (C) Activity measurements of reduced (-SH) and S-glutathionylated (-SG) zfSIRT and zfSIRT1C204S (mean ± SEM, n = 3). (D) Ten micromolars zfSIRT1 and zfSIRT1C204S, both S-glutathionylated with 0.5 mM fluorescent Eosin-Di-GSSG, with and without following incubation with prereduced zfGrx2 (2.5× molar excess) were separated by SDS/PAGE. Bars represent the densitometric analyses of GSH bound to zfSIRT1, mean ± SEM (n = 2). GSSG, fluorescence of SIRT1 after incubation with Di-Eosin-GSSG; SIRT1, Coomassie staining of WT and C204S proteins. Significance (*P < 0.05, ***P < 0.001) was calculated by two-tailed Student’s t test.

Fig. 5.

Grx2 regulates vascular development via SIRT1. (A) Using 50 µg protein extract of zfGrx2 knockdown embryos and WT controls 48 hpf, five transitions for the glutathionylated and NEM coupled (reduced) variant of the zfSIRT1 peptide ILVLTGAGVSVSCGIPDFR were monitored using a triple quadrupole mass spectrometer, mean ± SEM (n = 2). (B) Quantification of Western Blot analyses of glutathionylated zfSIRT1 (zfSIRT1-SG) after biotin labeling of Grx2-specific deglutathionylated thiols in zebrafish embryos 48 hpf, mean ± SEM (n = 4). Bars represent the densitometric analyses of glutathionylated zfSIRT1/total zfSIRT1. (C) Grx2 is essential for vascular development via activation of SIRT1 through reversible S-glutathionylation of a cysteine residue located in the catalytic domain of this protein deacetylase.