Sulfite oxidase deficiency causes persulfidation loss and hydrogen sulfide release
Chun-Yu Fu, Joshua B. Kohl, Filip Liebsch, Davide D’Andrea, Tamás Ditrói, Seiryo Ogata, Franziska Neuser, Max Mai, Anna T. Mellis, Emilia Kouroussis, Masanobu Morita, Titus Gehling, José Angel Santamaria-Araujo, Sin Yuin Yeo, Heike Endepols, Michaela Křížková, Viktor Kozich, Marcus Krueger, Julia B. Hennermann, Uladzimir Barayeu, Takaaki Akaike, Peter Nagy, Milos Filipovic, Guenter Schwarz
Chun-Yu Fu, Joshua B. Kohl, Filip Liebsch, Davide D’Andrea, Tamás Ditrói, Seiryo Ogata, Franziska Neuser, Max Mai, Anna T. Mellis, Emilia Kouroussis, Masanobu Morita, Titus Gehling, José Angel Santamaria-Araujo, Sin Yuin Yeo, Heike Endepols, Michaela Křížková, Viktor Kozich, Marcus Krueger, Julia B. Hennermann, Uladzimir Barayeu, Takaaki Akaike, Peter Nagy, Milos Filipovic, Guenter Schwarz
View: Text | PDF
Research Article Clinical Research Metabolism

Sulfite oxidase deficiency causes persulfidation loss and hydrogen sulfide release

Abstract

Sulfite oxidase (SOX) deficiency is a rare inborn error of cysteine metabolism resulting in severe neurological damage. In patients, sulfite accumulates to toxic levels, causing a rise in the downstream products S-sulfocysteine, which mediates excitotoxicity, and thiosulfate, a catabolic intermediate/product of hydrogen sulfide (H2S) metabolism. Here, we report a full-body knockout mouse model for SOX deficiency (SOXD) with a severely impaired phenotype. Among the urinary biomarkers, thiosulfate showed a 45-fold accumulation in SOXD mice, representing the major excreted S-metabolite. Consistently, we found increased plasma H2S, which was derived from sulfite-induced release from persulfides, as demonstrated in vitro and in vivo. Mass spectrometry analysis of total protein persulfidome identified a major loss of S-persulfidation in 20% of the proteome, affecting enzymes in amino acids, fatty acid metabolism, and cytosolic iron-sulfur cluster biogenesis. Urinary amino acid profiles indicated metabolic rewiring and mitochondrial dysfunction, thus identifying an altered H2S metabolism and persulfidation in SOXD. Finally, oxidized glutathione and glutathione trisulfide were able to scavenge sulfite in vitro and in vivo, extending the lifespan of SOXD mice and providing a mechanistic concept of sulfite scavenging for the treatment of this severe metabolic disorder of cysteine catabolism.

Authors

Chun-Yu Fu, Joshua B. Kohl, Filip Liebsch, Davide D’Andrea, Tamás Ditrói, Seiryo Ogata, Franziska Neuser, Max Mai, Anna T. Mellis, Emilia Kouroussis, Masanobu Morita, Titus Gehling, José Angel Santamaria-Araujo, Sin Yuin Yeo, Heike Endepols, Michaela Křížková, Viktor Kozich, Marcus Krueger, Julia B. Hennermann, Uladzimir Barayeu, Takaaki Akaike, Peter Nagy, Milos Filipovic, Guenter Schwarz

×

Figure 4

Persulfidome remodeling in SuoxKO/KO mice.

Options: View larger image (or click on image) Download as PowerPoint
Persulfidome remodeling in SuoxKO/KO mice. (A) Setup for the detection o...
(A) Setup for the detection of protein-bound persulfides in the assay described by Zivanovic et al. (18). Crude extract from liver was treated with 4-chloro-7-nitrobenzofurazan (NBF-Cl), which reacts with persulfides, free thiols, sulfenic acids, and amine groups. The NBF-blocked persulfides undergo nucleophilic attack by the DAz-2 moiety, which in turn is cross-linked to a fluorescent Cy5 moiety via click chemistry. (B) Representative Cy5 (persulfonate) and NBF-Cl (control) fluorescence signals measured by a fluorescence scanner (Cy5, red; NBF-Cl, green). (C) Quantification of a specific rectangular area identical for all lanes. All Cy5 signals were normalized to their respective NBF-Cl signals and further normalized to the mean of SuoxWT/WT signals (n = 3). Error bars indicate SD. One-way ANOVA was performed. (D) Heatmap showing the significant changes of protein persulfidation in SuoxKO/KO mouse livers compared with WT animals (Welch’s test, P < 0.05). (E) Volcano plot depicting statistical significance plotted against the log2 fold change of persulfidated proteins in SuoxKO/KO mouse liver relative to WT animals. Significance was established using Welch’s t test (2 sided), with a P value threshold of <0.05. Fold change cutoffs were established at 30%. The numbers of proteins with significantly altered persulfidation states are indicated. (F) KEGG pathway enrichment analysis using DAVID. The graph shows the top 30 significant (Benjamini’s adjusted P value < 0.01) and most enriched terms, with color gradient signifying the adjusted P value and circle size the number of proteins. (G) GO (Biological Process) term enrichment analysis of the 441 proteins found to have decreased persulfidation levels in SuoxKO/KO mice. REVIGO was used to plot the enrichment analysis performed in DAVID. Circle dimensions denote the protein count within specific GO terms, while color gradients depict the degree of significance. Similar biological processes are grouped together with a dashed line. (H) AlphaFold model of human ABCB7 protein with the C-terminal helix pointing into the transport pore. The C-terminal Cys residues are highlighted in a spheres model. (I) Determination of cytosolic Aco2 activity, (J) molybdopterin (MPT)/Moco content via HPLC Form A analysis, and (K) urinary uric acid/xanthine (UA/XA) ratio. ***P < 0.001; **P < 0.01; *P < 0.05.