Nitric oxide regulates mitochondrial fatty acid metabolism through reversible protein S-nitrosylation

Abstract

Cysteine S-nitrosylation is a posttranslational modification by which nitric oxide regulates protein function and signaling. Studies of individual proteins have elucidated specific functional roles for S-nitrosylation, but knowledge of the extent of endogenous S-nitrosylation, the sites that are nitrosylated, and the regulatory consequences of S-nitrosylation remains limited. We used mass spectrometry-based methodologies to identify 1011 S-nitrosocysteine residues in 647 proteins in various mouse tissues. We uncovered selective S-nitrosylation of enzymes participating in glycolysis, gluconeogenesis, tricarboxylic acid cycle, and oxidative phosphorylation, indicating that this posttranslational modification may regulate metabolism and mitochondrial bioenergetics. S-nitrosylation of the liver enzyme VLCAD [very long chain acyl-coenzyme A (CoA) dehydrogenase] at Cys(238), which was absent in mice lacking endothelial nitric oxide synthase, improved its catalytic efficiency. These data implicate protein S-nitrosylation in the regulation of β-oxidation of fatty acids in mitochondria.

Figure 1

Overview of the cysteine S-nitrosoproteome of the mouse. (A) The number of sites and proteins identified across six organs in wild-type mouse and their dependency to eNOS activity as percentage of the wild type is indicated in parenthesis. Three biological replicates from each organ from wild type and eNOS^−/− mice were analyzed. (B) Sub-cellular localization of the S-nitrosoproteome (white bars) and comparison with the rest of the mouse proteome (grey bars); ^ indicate overrepresentation and # underrepresentation, with both indicating p<0.0001. (C) Percentage of S-nitrosylated mitochondrial proteins. The white bar indicates proteins shared with at least another organ. The hatched bar indicates unique proteins for a particular organ. (D) S-nitrosylated enzymes present in wild-type mouse liver that were absent in eNOS^−/− liver and that are involved in glucose metabolism, TCA cycle and fatty acid metabolism are marked in red. S-nitrosylated proteins that are present in eNOS^−/− liver are depicted in blue. Asterisk (*) indicates proteins that have been reported to be S-nitrosylated. For all the proteins and sites identified see tables S1 to S6.

Figure 2

(A) Rate of ³H-labeled palmitoyl-CoA oxidation in liver homogenates from wild-type and ob/ob mice.* p<0.05 by ANOVA with Bonferroni's post hoc test between wild-type and ob/ob PBS-treated mice (N=4 mice). **p<0.01 by ANOVA with Bonferroni's post hoc test between ob/ob PBS and ob/ob GSNO treated mice (N= 4 mice). (B) Liver triglyceride measurements in wild-type, PBS-treated ob/ob mice, and GSNO-treated ob/ob mice. *p<0.0001 by ANOVA with Bonferroni's post hoc test between wild-type and ob/ob PBS-treated mice (N=4 mice), **p<0.005 by ANOVA with Bonferroni's post hoc test between ob/ob PBS and ob/ob GSNO treated mice (N= 4 mice). (C) Serum triglyceride measurements in wild-type, PBS-treated ob/ob mice, and GSNO-treated ob/ob mice. No statistical difference (N=4 mice). (D) Representative measurements of VLCAD initial velocity measured as a function of palmitoyl-CoA concentration in liver homogenates from wild-type (black line), ob/ob PBS (red line) and ob/ob GSNO (blue line) mice. (E and F) Kinetic analysis of VLCAD enzymatic activity reveals similar V_max but significantly higher K_M in PBS-treated ob/ob mouse liver as compared to wild-type and GSNO-treated ob/ob mouse liver. *p<0.05 by ANOVA with Bonferroni's post hoc test. N=3 biological replicates.

Figure 3

(A) Representative Western blots assessing VLCAD in native gel (top panel) and in SDS-gel (middle panel). Noncontiguous lanes from a single experiment are indicated by black lines. (B) Quantification of abundance of VLCAD in SDS-gels under reducing conditions in total liver homogenates and enriched mitochondria fractions from liver. No statistical difference by ANOVA. N=3 different mice. (C) Representative Western blot for VLCAD in liver homogenates eluted from organomercury resin. The signal intensity was used to determine the abundance of S-nitrosylated VLCAD in the bound (B) fraction and the unmodified VLCAD present in the unbound (U) fraction. The data were repeated in two independent liver homogenates.

Figure 4

Hepa 1-6 cells transiently expressing either FLAG tagged wild-type or C238A VLCAD were exposed to GSNO. (A) Representative Western blot analysis of unbound (VLCAD) and bound fractions (SNO-VLCAD) collected after mercury assisted capture in cell lysates. The unbound fraction indicates the abundance of the unmodified protein. The bound fraction indicates the abundance of S-nitrosylated VLCAD. The abundance of both was determined by using a calibrated antibody binding curve using purified VLCAD. The fraction of S-nitrosylated VLCAD as percentage of VLCAD is indicated. ND, not detected. N=3 biological replicates. Noncontiguous lanes from a single experiment are indicated by black lines. (B) The specific activity of VLCAD was significantly higher in GSNO-treated cells expressing wild-type VLCAD but not in GSNO-treated cells expressing equivalent amount of C238A VLCAD mutant protein. *p< 0.001, **p<0.05, by ANOVA with Bonferroni's post hoc test, N=3 biological replicates. ** “This manuscript has been accepted for publication in Science Signaling. This version has not undergone final editing. Please refer to the complete version of record at http://www.sciencesignaling.org/. The manuscript may not be reproduced or used in any manner that does not fall within the fair use provisions of the Copyright Act without the prior, written permission of AAAS.”