Thiosulfate sulfurtransferase deficiency promotes oxidative distress and aberrant NRF2 function in the brain

Thiosulfate sulfurtransferase (TST, EC 2.8.1.1) was discovered as an enzyme that detoxifies cyanide by conversion to thiocyanate (rhodanide) using thiosulfate as substrate; this rhodanese activity was subsequently identified to be almost exclusively located in mitochondria. More recently, the emphasis regarding its function has shifted to hydrogen sulfide metabolism, antioxidant defense, and mitochondrial function in the context of protective biological processes against oxidative distress. While TST has been described to play an important role in liver and colon, its function in the brain remains obscure. In the present study, we therefore sought to address its potential involvement in maintaining cerebral redox balance in a murine model of global TST deficiency (Tst−/− mice), primarily focusing on characterizing the biochemical phenotype of TST loss in relation to neuronal activity and sensitivity to oxidative stress under basal conditions. Here, we show that TST deficiency is associated with a perturbation of the reactive species interactome in the brain cortex secondary to altered ROS and RSS (specifically, polysulfide) generation as well as mitochondrial OXPHOS remodeling. These changes were accompanied by aberrant Nrf2-Keap1 expression and thiol-dependent antioxidant function. Upon challenging mice with the redox-active herbicide paraquat (25 mg/kg i.p. for 24 h), Tst−/− mice displayed a lower antioxidant capacity compared to wildtype controls (C57BL/6J mice). These results provide a first glimpse into the molecular and metabolic changes of TST deficiency in the brain and suggest that pathophysiological conditions associated with aberrant TST expression and/or activity renders neurons more susceptible to oxidative stress-related malfunction.

In the context of sulfide metabolism, TST oxidizes the endogenous gasotransmitter H 2 S in a similar fashion as sulfide quinone oxidoreductase (SQOR) and persulfide dioxygenase (ETHE1/PDO) [9].TST also has the capacity to form H 2 S from thiosulfate by utilizing dihydrolipoic acid (DHLA), the reduced form of α-lipoic acid [10,11].Balancing both functions, H 2 S oxidation and production, it seems that the primary function of TST is in the detoxification of H 2 S and less so in the production of H 2 S [12].TST is almost exclusively expressed in mitochondria, the organelle that synthesizes ATP by oxidative phosphorylation (OXPHOS) through stepwise reduction of oxygen in the Electron Transport Chain (ETC) and the generated proton gradient using five enzyme complexes (Complex I-V) [13,14].TST activity appears to modulate reduced nicotinamide adenine dinucleotide (NADH) dehydrogenase (Complex I) and other mitochondrial complexes including succinate dehydrogenase (Complex II) via interaction with their iron-sulfur clusters [6,7].In addition to this interaction between mitochondrial activity and TST, phosphorylation of TST may modulate its activity and thereby affect the stability of sulfur in the iron-sulfur centers of Complexes III and IV, further impacting electron transport and ATP production [15].In mammalian cells, TST is linked to two thiol-dependent antioxidant systems, the thioredoxin system and the GSH system [15].TST exerts anti-oxidative properties via donating sulfur for these two systems and an increase of TST activity can (in) directly activate their reactive oxygen species (ROS) scavenging function [12,16].
Oxidative stress has classically been defined as an imbalance in the generation of pro-oxidant and antioxidant species in favor of the former [17].A series of experimental observations promoted a redefinition of oxidative stress as a condition that is associated with alterations in redox signaling and control, and an updated interpretation of the original concept distinguishes physiological oxidative stress (or 'oxidative eustress') from excessive and deleterious oxidative stress (coined 'oxidative distress') [18].Besides ROS, some other types of short-lived molecules such as reactive nitrogen (RNS) and reactive sulfur species (RSS) contribute to cellular signaling processes and merit consideration specifically in the context of sensing and adaptation to changes in environmental and/or metabolic conditions.These considerations are part of the stress signaling paradigm and have recently been conceptualized in the form of the Reactive Species Interactome (RSI) framework [19,20].
The transcription factor Nrf2 (nuclear factor (erythroid-derived 2)like 2) is a master regulator of cellular homeostasis that controls the expression of genes related to redox homeostasis [21].The primary mechanism of NRF2 regulation is at the protein level.The majority of research has focused on the involvement of the electrophile and redox sensor Kelch-like ECH-associated protein 1 (KEAP1) in regulating NRF2 protein levels in response to metabolic demands [22].The association between KEAP1 and NRF2 is disrupted by electrophilic alteration or oxidation of cysteine thiols in KEAP1, enabling cells/organisms to respond to environmental stress.Following dissociation from KEAP1, NRF2 escapes destruction and targets antioxidant response element (ARE) genes, which leads to an enhancement of antioxidant capacity [23].Moreover, various studies linked the TST-related antioxidant system to nuclear factor erythroid 2-related factor 2 (Nrf2) signaling due to its transcriptional activation of GSH-related enzymes, one of the first lines of defense against oxidative stress [24][25][26][27].Here, H 2 S mediates direct persulfidation of KEAP1 (Kelch-like ECH-associated protein 1) and thereby contributes to sulfide-mediated NRF2 regulation [28].
Several studies have demonstrated that TST provides an essential protective function against disease [15].Studies investigating the mechanistic effects of TST, which were identified in polygenic lean mouse adipocytes, have indicated its potential as a predictor for diabetes [26].In an Icelandic cohort of almost 700 people, an inverse correlation was observed between TST mRNA levels in subcutaneous adipose tissue and body mass index (BMI) [29].Upregulation of adipose TST was detected in lean mice challenged with a high-fat diet (HFD), further supporting its likely protective effects [29].Additionally, mice with transgenic Tst overexpression in mature adipocytes, showed resistance against HFD-induced obesity and exhibited higher protein expression of mitochondrial superoxide dismutase 2 (SOD2) and higher mRNA levels of peroxiredoxin 3 (Prdx3) compared to control mice, while cytosolic superoxide dismutase 1 (SOD1) remained unchanged [29].These findings provide evidence of an interaction between TST and mitochondrial ROS.Consistent with a function of TST for ROS removal, knockdown of Tst in the preadipocyte cell line 3T3-L1 generated higher levels of mitochondrial ROS in the presence of oxidative stress stimuli [29].Furthermore, ROS-related adiponectin release from 3T3-L1 adipocytes was decreased in cells treated with the TST inhibitor, 2-PTS (2-propenyl thiosulfate), while adiponectin was increased following addition of the TST substrate thiosulfate [29] tissue, and has reduced expression in colon mucosa of patients with ulcerative colitis and colon cancer [30][31][32].During the development of dextran sulfate sodium (DSS) induced colitis in mice, the expression of Tst mRNA, protein and activity significantly decrease [33].
Taken together, these findings indicate that TST has an important function in maintaining the redox balance in colon and adipocytes, but there are no studies elucidating the importance of TST in redox signaling within the brain.A clinical report of rhodanese (i.e.TST) deficiency in a rare neurological and mitochondrial human disorder known as Leber's Hereditary Optic Atrophy (LHON) [34] prompted us to investigate the underlying molecular and metabolic changes of TST deficiency in the brain.From physiology to pathology, the brain consumes about 20% of the whole-body supply of oxygen and energy [26].The cerebral cortex, with its intense neuronal activity and high metabolic activity, has a particularly high susceptibility to perturbation by oxidative stress [35][36][37].In Alzheimer's Disease, increased oxidative stress is associated with the progression of the neurodegenerative pathology along with amyloid and hyperphosphorylated tau [38,39].This is due to high mitochondrial generation of ROS, including superoxide anion (O 2 • − ) and hydrogen peroxide (H 2 O 2 ) [19,40].Hence, we hypothesized that TST plays a vital role in enabling an appropriate redox balance in the brain, and that global gene silencing of TST in mice would provide novel insights into the central role of TST in neurological antioxidant signaling and tissue protection.

Experimental animals
All the experiments were performed according to international guidelines set out by the ethical committees of the University of Edinburgh and within the framework of the Animals (Scientific Procedures) Act (1986) of the United Kingdom Home Office.Animals were maintained in standard housing conditions with a 12-h light and 12-h dark cycle (7 a.m.-7 p.m.) and ad libitum access to rodent chow.All studies used female mice housed in cages of 3-6 individual littermates.The female mice used for the study originated from C57BL/6 N Tst − /− mice [29] backcrossed onto a C57BL/6J genetic background for more than 10 generations.Pieces of brain cortex were snap frozen and stored at − 80 • C until use.

Induction of oxidative stress in vivo using paraquat
To test the efficacy of paraquat dichloride (PQ, 36541-SIGMA) as an inducer of oxidative stress in the brain cortex tissue in vivo, PQ (in physiological saline 0.9%) was administered to WT and Tst − /− mice by intraperitoneal injection at a concentration of (25 mg/kg body weight).24 h later, mice were culled by decapitation and organs were harvested for analysis.

Total ATP, GSH/GSSG and MPST activity measurements
Total ATP content was measured with the CellTiter-Glo 2.0 Luminescent assay (G9241, Promega).GSH/GSSG content and ratio were measured with the Glutathione GSH/GSSG Assay Kit (MAK440, Sigma-Aldrich) following the manufacturer's instructions.MPST enzyme activity was detected with MPST ELISA kit (MBS9912889, Mybiosource) according to manufacturer's instructions.All analyses were performed using freshly extracted total protein from brain cortex samples, and using microplate multimode reader Spark (Tecan).For MPST enzyme activity, absorbance reading were obtained by a Synergy H4 Hybrid Reader (Biotek).
The mitochondrial respiration parameters were analyzed using an O2K oxygraphy (Oroboros systems, Innsbruck, Austria).250 μg crude mitochondria from wildtype and Tst − /− mouse cortexes were monitored under continuous stirring at 750 rpm in 1 mL MiR05 (0.5 mM EGTA, mM MgCl2, 60 mM lactobionic acid, 20 mM taurine, 10 mM KH 2 PO 4 , mM HEPES, 110 mM D-Sucrose, BSA, 1 g/l essentially fatty acid free, at pH7.4).Oxygen polarography was performed at 37 • C and the oxygen flux per tissue mass (pmol O 2 /s/mg) were recorded in real-time using the DatLab5 software.The mitochondrial respiration was assessed with the use of different substrates as follows: 5 mM pyruvate, 2 mM malate and 1.5 μM FCCP to uncouple all the ETC complexes and measure complex I respiration; subsequently, 0.5 μM rotenone was added to inhibit complex I, followed by 10 mM succinate to measure complex II respiration.In addition, 2.5 μM antimycin A was added to prevent both complex I and II mediated respiration due to inhibition of complex III, followed by 2 mM ascorbate combined with 0.5 mM TMPD to measure complex IV-mediated respiration.The measurement sequence was completed by injection of 10 mM sodium dithionite.Total SOD activity was measured with the Superoxide Dismutase Activity Assay Kit (AB65354, ABCAM), which assesses the inhibition activity of xanthine oxidase by SOD.Catalase activity was assessed with the Catalase Activity Assay Kit (AB83464, ABCAM).Total Nitric Oxide Synthase activity was measured with the Nitric Oxide Synthase (NOS) Activity Assay Kit (MAK407, Sigma-Aldrich).Total peroxidase activity was assessed with the Peroxidase Activity Assay Kit (MAK092, Sigma-Aldrich).Analyses were performed following manufacturer's instructions, with freshly extracted total proteins from brain cortex samples, and using microplate multimode reader Spark (Tecan).

Glutamate content
Glutamate concentration was measured with the Glutamate Assay Kit (AB138883, ABCAM).
Analysis was performed following manufacturer's instructions, with freshly extracted total proteins from brain cortex samples, and using microplate multimode reader Spark (Tecan).

RNA isolation, cDNA synthesis and quantitative real-time PCR
RNA was extracted from tissue slices according to the instruction of miRNeasy ® Mini Kit (Qiagen, Germany).Total RNA concentration was quantified by Nanodrop® (ND-1000 Spectrophotometer).Equal amount of cDNA synthesis and amplification were performed using Superscript II kit and RANDOM primers (Thermo Fisher Scientific, USA), according to the manufacture's protocol.The cDNA synthesis mix are incubated as follows: 10 min at 25 • C, 50 min at 42 • C, 15 min under 70 • C via a Veriti 96 Well Thermal Cycler (Thermo Fisher Scientific).For qRT-PCR, 500 ng of cDNA samples were mixed with SYBR green, forward primers, and reverse primers (Primers used for quantification are shown in Supplementary Table S1).The qPCR was run on LightCycler® 480 II (Roche) and the protocol was as follows: 2 min at 50

Statistics
The data are presented as mean ± standard deviation (SD).Statistical differences between two conditions were calculated using an unpaired T-test.Statistical differences among more than two conditions were calculated using an one-way ANOVA test.Statistical analysis was performed in GraphPad Prism (version 9.1.0),and repeat numbers are specified in the figure legends.

Tst − /− cerebral cortex exhibits a steady H 2 S level and a dysregulated GSH pathway
To investigate how TST influences the GSH pathway in the brain, we performed studies in the cortical brain region of conventional knock-out mice in which Tst deletion (termed Tst − /− ) has been successfully validated (Fig. 1A).Although the difference in thiosulfate concentrations in the cerebral cortex between wildtype and knockout mice did not reach statistical significance (likely due to small sample size), mean thiosulfate concentration was approx.3-fold higher in Tst − /− mice compared to WT (Fig. 1B).Next, sulfide and polysulfide concentrations were assessed in the cortex.While tissue steady-state levels of both, H 2 S and H 2 S n were found to be maintained at similar levels in Tst − /− and C57BL/6J mice, mean concentrations were lower for H 2 S and higher for polysulfides (Fig. 1B).To investigate whether MPST can compensate for the lack of TST in the brain of the Tst − /− mice, we analyzed Mpst mRNA and protein levels.MPST protein and Mpst mRNA was decreased in Tst − /− mice (Fig. S1A-D).Interestingly, the MPST activity was not changed, although it tended to be higher in the brain of Tst − /− mice compared to C57BL/6J mice (Fig. S1E).To verify whether the GSH pathway was altered in the brain of Tst − /− mice, we measured its GSH and GSSG content.GSH was 36% lower in Tst − /− mice, while its oxidized form, GSSG, increased 5 times, causing the GSH/GSSG ratio to decline 7.2-fold in Tst − /− mice compared to wildtype controls (Fig. 1C).This indicates a redox imbalance secondary to TST enzyme activity loss.Moreover, we found a significant decrease of GPX4 protein (Fig. 1D; Fig. S1F) and the transcriptional level of GR significantly decreased in Tst − /− brain cortex (Fig. 4J).Altogether, our results established that expression and activity of TST exert a major effect on both H 2 S metabolism and the GSH antioxidant pathway (Fig. 1E).

TST deletion increases ATP in the cerebral cortex and elevates ROS via modulation of OXPHOS activity
Since TST exists almost exclusively in mitochondria and is capable of modulating mitochondrial respiratory complexes, we examined the impact of Tst deletion on mitochondrial activity in the cortical brain areas.For that purpose, we performed total ATP measurements generated by OXPHOS via luminescence.Total ATP was increased by 22% in the Tst − /− mice cortex compared with C57BL/6J mice (Fig. 2A).Next, we measured the protein levels of OXPHOS complexes, and found no significant differences between the two mouse genotypes (Fig. 2B; Fig. S2A).In contrast, basal respiration and complex IV activity represented by oxygen consumption rate were markedly increased in the cortex of Tst − /− mice, while no significant differences in Complex I capacity were observed among either mouse genotypes (Fig. 2C).The increase of complex IV activity in the absence of TST indicates an impact of this enzyme on mitochondrial capacity.The function of TST as modifier of GSH to form glutathione persulfide (which is a more efficient ROS scavenger than GSH itself) has been well characterized [12,16].We next determined the extent of O 2 -derived reactive species formation from OXPHOS.The O 2 .−level was 10% higher in the cortex of Tst − /− mice, while tissue levels of H 2 O 2 were 57% higher in the cerebral region of Tst − /− mice compared to C57BL/6J control mice (Fig. 2D).

Tst deficiency reprograms the antioxidant defense against endogenous ROS
To determine whether antioxidant expression/activity changed in the cortex of Tst − /− mice, we checked both protein and total enzyme Polysulfides (H 2 S n ) became novel reactive sulfur species, which is derived from H 2 S with MPST.TST converts thiosulfate produced via non-enzymatic reactions to sulphite.Then oxidation of sulphite to sulfate is dependent on the glutathione system.GPX4 and GR is the cofactor of conversion between GSH and GSSG.(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)activity levels of superoxide dismutase 1 (SOD1) and superoxide dismutase 2 (SOD2).Both, SOD1 and SOD2 are vital endogenous antioxidant enzymes.The expression of SOD1 and SOD2 proteins was unchanged between the two groups, whereas total SOD enzyme activity showed a 34% decrease in the Tst − /− mice (Fig. 3A; Fig. S3A).When H 2 O 2 is excessively produced, catalase participates in H 2 O 2 breakdown.The catalase activity in Tst − /− was 7.6% higher relative to that observed in C57BL/6J mice (Fig. 3B).In addition, the total peroxidase from Tst − /− mice exhibited unaltered activity compared with C57BL/6J mice (Fig. 3C).In the context of the RSI framework, we also measured nitric oxide synthase (NOS) activity and RNS (NO and ONOO − ) production.NOS activity and NO content were similar in the brain cortex of either genotype (Fig. S3B and D), while ONOO − amount in cortexes of Tst − /− mice was 10% lower than in C57BL/6J mice (Fig. S3C).In summary, we found decreased SOD enzyme activity, increased catalase activity and unchanged peroxidase activity, when ROS is elevated due to the global loss of TST in murine brain tissue as schematically depicted in Fig. 3D.

Tst absence results in decrease of Nrf2-Keap1 pathway expression level and increased glutamate content
Activation of the Nrf2-Keap1 system results in an antioxidant defense response which is, at least in part, regulated by H 2 S and KEAP1 protein level [21,22,28].To investigate whether Tst deficiency causes a dysregulation of Nrf2 signaling, qRT-PCR was conducted for genes regulated by antioxidant response elements (AREs) under control of NRF2.Nrf2 and Keap1 mRNA were significantly decreased in Tst − /− cortical brain samples compared with C57BL/6J mice (Fig. 4A and B).NRF2 protein showed significant reduction and KEAP1 protein had an increased expression in Tst − /− mice brain cortexes when compared to those of control mice (Fig. 4C-E).At the same time, different ARE genes including Hmox1 and Txn2 mRNA showed relative reduction in the case of Tst deficiency, while Gstm1, Txn1, Noq1and Cat revealed no significant difference between the two types of mice (Fig. 4G).
GCLC, GCLM and GR are important ARE genes for proteins protecting against oxidative stress.All these genes were found to be significantly reduced in the Tst − /− brain (Fig. 4G).Since both GCLC and GCLM are critical for de novo GSH biosynthesis we also measured glutamate content and found a 1.8-fold increase in Tst − /− compared to wildtype mice, corroborating the effect of Tst deletion on this important antioxidant pathway (Fig. 4F).

Tst − /− mouse brain displayed deteriorated antioxidant capacity when facing paraquat-induced oxidative stress
In order to show the effects of TST loss when facing oxidative stress, we analyzed several oxidative stress markers including ROS, RSS, GSH/ GSSG contents and GPX4 protein abundance.Under basal TST deficiency condition, ROS showed a significant increase in the Tst − /− mouse brains.Upon challenging both genotypes of mice with 25 mg/kg paraquat (PQ) for 24 h, both genotypes of mouse brain cerebral cortices showed significantly increased ROS compared to their basal levels.Cerebral cortex of Tst − /− mice showed 1.5-fold increase of H 2 O 2 compared to those of C57BL/6J mice (Fig. 5A).As for reactive sulfur species, brain cortexes of Tst − /− mice showed similar amounts of H 2 S in the cortical brain area compared to C57BL/6J mice.Upon PQ challenging, there was no difference in both H 2 S and H 2 S n intensity between the two genotypes (Fig. 5B).
The parameters relevant to the GSH content were also assessed.Following paraquat stimulation, both mouse genotypes had significantly higher GSH content compared to their basic levels in brain tissue.However, GSH amount generated in the brain cortex of Tst − /− mice was less than that of C57BL/6J mice.In contrast to the accumulation of GSSG in basal cerebral cortical area of Tst − /− mice, PQ-treated Tst − /− mice also had relatively lower GSSG compared to C57BL/6J mice.Therefore, in the cerebral cortex of PQ-treated mice the ratio of GSH/GSSG showed a significant increase when compared with C57BL/6J mice (Fig. 5C).Furthermore, we tested GPX4 protein expression and found that GPX4 had higher levels in PQ injected animals in response to acute redox damage in both mouse genotypes.In the mouse brain of Tst − /− mice, GPX4 protein level was slightly lower than in C57BL/6J mice (Fig. 5D).In summary, this additional challenge experiment showed an overall deteriorated antioxidant capacity in murine cerebral cortices in conditions of global TST deficiency.

Discussion
TST plays an important role in the oxidation of the endogenous gasotransmitter hydrogen sulfide, complementing the action of sulfide quinone oxidoreductase (SQOR) and persulfide dioxygenase (ETHE1/ PDO) [9,42].Whether or not TST also contributes to antioxidant protection in the brain has not been investigated before.In Tst − /− mice, circulating thiosulfate and sulfide level were markedly elevated, while cerebral cortex displayed similar steady-state levels of sulfide and thiosulfate, as observed in the liver [43].This indicates that the response to Tst knock-down is organ specific in this chronic but viable sulfide elevation model.Starting from organ specificity, the free thiol antioxidant system of brain and liver revealed opposite responses in terms of GSH content.In Tst − /− mice, GSH was decreased in the cerebral cortex, while there was no difference in GSH content between the two genotypes in the liver [43].Since NADPH/H + dependent glutathione reductase (GR) is important for the recycling of GSSG to GSH [44], its down-regulation in the brain of Tst − /− mice explains the alterations in GSSG levels observed.In support of an activation role of TST on ROS scavengers such as GSH [15], Tst − /− mice showed a lower GSH/GSSG ratio and higher ROS levels, indicating a redox signaling imbalance along with lower GPX4 and GR expression.Meanwhile, the Mpst transcription level of brain and liver in Tst − /− mice were reduced.In contrast, the liver functions as a major detoxification organ, and despite lower mRNA for Mpst protein levels were upregulated, perhaps for compensatorily enhanced sulfide removal [43], while MPST protein significantly decreased in the brain cortex and in mitochondria in Tst − /− mice.
TST interacts with iron-sulfur cluster proteins including Complex I (NADH dehydrogenase), Complex II (succinate dehydrogenase) and Complex III (Cytochrome bc1) of the mitochondrial respiratory chain) in oxidative pathway [45].TST directly interacts with Complex I via sulfide transfer. 6In addition, TST transfers sulfane sulfur to Complex II and modifies its iron-sulfur structure via alteration of its phosphorylation status.Those functions led us to explore the metabolic phenotypes of the brain tissues of C57BL/6J and Tst − /− mice.The unexpected finding of mitochondrial activity remodeling was supported by several observations, including unchanged OXPHOS protein expression, increased basal respiration capacity and Complex IV activity in the presence of ascorbate and TMPD.However, the cerebral cortex of Tst − /− mice produced more ATP, and higher amounts of ROS compared to C57BL/6J control mice.ATP, the universal energy currency, is mainly synthesized in mitochondria by OXPHOS.The representative role of Complex IV as regulatory center of OXPHOS limits the rate of respiratory chain reaction [46].Hence, our finding of enhanced Complex IV activity is in line with the observation of enhanced ATP levels in Tst − /− mice.
ROS are generated during the incomplete reduction of O 2 [47,48].In mitochondria, the major sources of O 2 • − are the OXPHOS Complexes I and III [20,49].O Our study of the Nrf2 pathway in Tst − /− mice highlights a potent interaction between Nrf2 signaling and the GSH antioxidant system [53][54][55][56].Importantly, Nrf2 is a major transcriptional factor regulating multiple processes related to GSH synthesis and regeneration [57,58].Beyond GSH content change, the transcriptional and protein levels of Nrf2 appear to be lower and its intracellular inhibitor Keap1 had higher protein expression in the brains of Tst − /− mice.A possible reason for the fact that TST deletion causes disrupted Nrf2-Keap1 signaling, might be found in the abnormal cysteine metabolism within the redox sensor protein KEAP1.TST's main function is sulfide oxidation as cysteine metabolism downstream, and KEAP1 protein contains 25 cysteines within mouse homologue, thus accumulated KEAP1 were detected in the brain of Tst − /− mice, consequently lower amount of Nrf2 protein was released from KEAP1 ubiquitin E3 ligase adapter, which further influenced the ARE gene transcription process [22,59,60].In light with reduced transcription of Keap1, we found an increased KEAP1 protein level, and we suggest a post-transcriptional cellular redox-sensing mechanism as operative.
Similar reduction in Nrf2 activation in Tst − /− mice using transcription factor binding site (TFBS) enrichment analysis were found earlier [43].Additionally, glutamate-cysteine ligase catalytic (GCLC) and modifier (GCLM) subunits together catalyze the rate-limiting step in GSH biosynthesis from glutamate under conditions of oxidative distress [26].Nrf2 − /− mice exhibited a decrease in GSH content and GCL levels [61], which supported our results on the brains of Tst − /− mice.The decrease in Nrf2 mRNA and protein translate into a reduced function of the GSH antioxidant system.Moreover, the decrease in Nrf2-mediated free thiol antioxidants influences various cellular processes, ranging from synthesis, oxidation and reduction of GSH.Among the heterogeneity of brain cell types, the expression of Tst mRNA is known to be highest in astrocytes (https://www.proteinatlas.org/ENSG00000128311-TST/single+cell+type/brain).Astrocytes comprise the majority of glial cells in the mammalian central nervous system (CNS) and serve as major provider of GSH for neurons [62,63].Astrocytes contribute to neuronal activity, and their interaction play distinct, cooperative roles in neuronal redox homeostasis.Astrocytes could initiate and promote an increase in gene expression of Nrf2-mediated antioxidant pathway in response to mild oxidative stress, which results in the synthesis of GSH using glutamate [64].When astrocytes function abnormally, glutamate, which is an excitatory neurotransmitter, will accumulate to an excitotoxic level and contribute to harmful effects on neurons, eventually causing impaired brain function [65].Thus, any deficiency in astrocyte capacity to produce GSH (or reduce GSSG) will render neurons particularly vulnerable to conditions or insults known to be associated with enhanced oxidative stress [66][67][68][69][70].Both oxidative stress and elevated glutamate contribute and are associated with brain pathologies [65].The decreased Nrf2 expression, reduced GSH and increased glutamate content, together with an altered ROS landscape in the CNS, secondary to a loss of TST are therefore key observations of the present study.
Relevant to Nrf2 signaling, another important thiol-dependent antioxidant system is the thioredoxin (TXN) system [71].TXN2 is a mitochondria-specific protein, whereas TXN1 is found in the cytosol, both of them operate using NADPH as cofactor [72].Down-regulated Txn2 mRNA level observed in the cerebral cortex of Tst − /− mice, could be explained by loss of TST's function to interact with mitochondrial NADH dehydrogenase (Complex I) and other iron-cluster proteins [6].Our findings provide a better understanding of the role of TST in linking Nrf2-keap1 signaling and thiol antioxidant pathways.
When challenging mice with paraquat-induced oxidative stress, the ROS production in Tst − /− mouse brain was dramatically increased due to the hampered antioxidant capacity, indicating the importance of the interaction of TST with the antioxidant pathway.The nature of GSH is to provide a protective effect on free radical scavenging and H 2 O 2 detoxification [73] therefore, after stimulation, increased GSH content was detected in both genotypes, although Tst − /− mouse brain had lower GSH levels than control mice.Unlike the basal condition, PQ-treated mice biosynthesized less GSSG when the brain tissue of Tst − /− mice was compared to the WT animals.PQ injection also induced an increase in GPX4 protein expression in response to excessive oxidative distress.GPX4 protein in Tst − /− mice was slightly lower than in C67BL/6J mice.In summary, this additional challenge experiment showed an overall deteriorated antioxidant capacity in murine cerebral cortex under the model of global TST deficiency.
In the context of brain function, both mitochondria and NRF2 functions are particularly important because the brain is highly susceptible to oxidative distress due to its high metabolic rate and relatively low levels of antioxidants compared to other tissues [74].During oxidative stress, structural and functional damage occurs to neurons and disrupts the normal actions of neurotransmitters, leading to altered signaling from neurons.This can lead to impaired cognitive function, affect behavior, and contribute to the development and progression of neurological and neurodegeneration diseases [75].Taken together, our data unmask several tissue-specific mechanisms linked to TST function that are related to mitochondrial and Nrf2-Keap1 signaled antioxidant functions, and the role of TST in maintaining the health and function of brain tissues.

Conclusion
Our study on mouse cerebral cortex revealed that TST deficiency promotes a dysregulation of the reactive species interactome through both generation of ROS and RSS, coupled with mitochondrial OXPHOS remodeling.In addition, TST deficiency elicits the dysregulation of thioldependent antioxidant systems with altered GSH levels and aberrant Nrf2 pathway expression.Moreover, the absence of TST is characterized by a deteriorated antioxidant buffering ability in the murine brain cortex when challenged with a redox cycler as demonstrated by paraquat in vivo.

Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 1 .
Fig. 1.Tst − /− mice brain cortexes display a dysregulated GSH pathway.A-D: Data are expressed as mean ± SD, student T test.(A) Tst mRNA and Tst protein levels of Tst − /− (red) and C57BL/6J (black) mice (n = 4).(B) Thiosulfate content in C57BL/6J mice (Black, n = 3) and Tst − /− mice brain cortexes (Red, n = 4) measured by HPLC-MS/MS.H 2 S and H 2 S n concentrations were measured using fluorescent probes, and show that H 2 S decreased 5% and H 2 S n increased 5% in the Tst − /− (red) compared with C57BL/6J mice (black) (n = 4).(C) GSH and GSSG contents in mice cortexes in the presence (black) or absence (red) of TST (n = 4).(D) Quantification of immunoblots of cerebral cortical area for GPX4 protein level in Tst − /− (red) compared with C57BL/6J mice (black) (n = 3).The original blotting picture is shown in Fig. S1F.(E) Brief schematic showing H 2 S enzymatic pathway.Sulfide is an endogenously produced gaseous signaling molecule via CBS, CSE and MPST.Polysulfides (H 2 S n ) became novel reactive sulfur species, which is derived from H 2 S with MPST.TST converts thiosulfate produced via non-enzymatic reactions to sulphite.Then oxidation of sulphite to sulfate is dependent on the glutathione system.GPX4 and GR is the cofactor of conversion between GSH and GSSG.(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2 .
Fig. 2. Tst − /− mice brain cortexes display metabolism reprogramming.(A-D) Data expressed as mean ± SD, student T test.(A) Total ATP level of C57BL/6J (black) and Tst − /− (red) mice cortical area (n = 4).(B) OXPHOS protein levels of two genotypes detected by OXPHOS cocktail antibody (n = 3).Quantification is normalized to ponceau stain.Immunoblot is displayed in Fig. S2.(C) Corrected oxygen consumption rate of 250 μg crude isolated mitochondria from mice cerebral cortices with various substrates.Respiratory capacity differences of basal respiration, Complex I, Complex II and Complex IV from C57BL/6J (black) and Tst − /− (red) were analyzed.(n = 3 biological samples, independent experiments were repeated at least 3 times).(D) Relative ROS contents as O 2 .-andH 2 O 2 of C57BL/6J (black) and Tst − /− (red) (n = 4).(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4 .
Fig. 4. Tst − /− mice brain cortexes display dysregulated Nrf2-keap1 signaling expression and increased glutamate content.A-B, D-G: Data expressed as mean ± SD, student T test.(A) Expression of Keap1 mRNA in C57BL/6J (black) (n = 4) and Tst − /− (red) (n = 4) mice levels of cortical area.(B) Expression of Nrf2 mRNA in C57BL/6J (black) (n = 4) and Tst − /− (red) (n = 4) mice levels of cortical area.(C) WB analysis of the two proteins.(D) Quantification of immunoblots of cerebral cortical area for NRF2 protein level in mice cortexes in the presence from C57BL/6J (black) (n = 3) and Tst − /− mice (n = 3).(E) Quantification of immunoblots of cerebral cortical area for KEAP1 protein level in mice cortexes in the presence from C57BL/6J (black) (n = 3) and Tst − /− mice (n = 3).(F) Glutamate content measurements in both genotypes (n = 4).(G) Tst − /− mice brain cortexes display dysregulated ARE genes.P value of each gene is indicated between genotypes.All experiments were conducted with repeat number as 4 per genotype.(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Y
. Luo et al.

FundingY
.L. was supported by China Scholarship Council (Grant No.: 202008520033).Z. M. Al-Dahmani was supported by the Islamic Development Bank.A.M.D. was supported by a Rosalind Franklin Fellowship co-funded by the European Union and the University of Groningen.S.M. was supported by a British Heart Foundation 4Y PhD Scholarship (FS/4yPhD/F/20/34,126).L.C was supported by the CNRS.Contribution statement YL, AM and HvG designed and conceptualized the study, wrote and revised the manuscript.YL conducted parts of research, LC designed and performed parts of study, edited the manuscript.YL and LC analyzed the data.ZMM conducted parts of experiments.SM, NZMH and NMM performed experiments on the animal model, provided samples, and edited the manuscript.MG and MF revised the manuscript.AM and HvG supervised this study.
. TST enables H 2 S detoxification in colon 2 •-and H 2 O 2 , RNS NO and ONOO − , and the RSS H 2 S and H 2 S n ylcoumarin, 802,409, Sigma-Aldrich) and H 2 S n (Sulfane Sulfur Probe 4/ SSP4, SB10-10, Tebubio) in freshly extracted total proteins from brain cortex samples.This was done by spectrofluorometry using a microplate multimode reader Spark (Tecan).Y. Luo et al. 2.7.Total SOD activity, catalase activity, total nitric oxide synthase activity, and total peroxidase activity • C, 10 min at 95 • C, 40 cycles of 15 s 95 • C and 60 s 60 • C, and the dissociation curve is run under 15 s 95 • C, 15 s 60 • C, 15 s 95 • C. RT-qPCR data was analyzed on LightCycler 480 software.Expression levels were normalized to GAPDH and displayed as 2-delta Ct.