If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The dual roles of H2S as an endogenously synthesized respiratory substrate and as a toxin raise questions as to how it is cleared when the electron transport chain is inhibited. Sulfide quinone oxidoreductase (SQOR) catalyzes the first step in the mitochondrial H2S oxidation pathway, using CoQ as an electron acceptor, and connects to the electron transport chain at the level of complex III. We have discovered that at high H2S concentrations, which are known to inhibit complex IV, a new redox cycle is established between SQOR and complex II, operating in reverse. Under these conditions, the purine nucleotide cycle and the malate aspartate shuttle furnish fumarate, which supports complex II reversal and leads to succinate accumulation. Complex II knockdown in colonocytes decreases the efficiency of H2S clearance while targeted knockout of complex II in intestinal epithelial cells significantly decreases the levels of thiosulfate, a biomarker of H2S oxidation, to approximately one-third of the values seen in serum and urine samples from control mice. These data establish the physiological relevance of this newly discovered redox circuitry between SQOR and complex II for prioritizing H2S oxidation and reveal the quantitatively significant contribution of intestinal epithelial cells to systemic H2S metabolism.
) in the electron transport chain (ETC), leading to respiratory poisoning (Fig. 1A). The mitochondrial sulfide oxidation pathway begins with the conversion of H2S to glutathione persulfide catalyzed by sulfide quinone oxidoreductase (SQOR), an inner mitochondrial membrane flavoprotein (
). Electrons released from H2S oxidation are transferred to coenzyme Q (CoQ) and enter the ETC at the level of complex III, making H2S an inorganic substrate for oxidative phosphorylation in mammals (
). In this study, we report that a noncanonical redox circuit is established when complex IV is inhibited, via reversal of complex II activity to prioritize H2S oxidation.
Figure 1The mitochondrial NADH pool influences the efficiency of H2S oxidation in HT29 cells.A, scheme showing that multiple CoQ (Q) users compete with SQOR including complexes I and II, dihydroorotate dehydrogenase (DHOD), glycerol 3-phosphate dehydrogenase (G3PDH), proline dehydrogenase (PRODH), and the electron transfer flavoprotein (ETF). B, H2S oxidation is enhanced in cells expressing mitochondrial but not cytoplasmic LbNOX versus the empty vector (EV) control. Rotenone (2 μM) enhanced H2S clearance in control and cytoplasmic but not mitochondrial expressing LbNOX cells. C, disruption of complex I by NDUFS3 knockdown enhanced H2S oxidation. D, mitochondrial expression of TPNOX accelerates H2S oxidation, which is inhibited by NBD-Cl. The data represent the mean ± S.D. of three independent experiments. ns, not significant.
). The dual potential to stimulate electron flux and inhibit the ETC raises questions as to whether modulation of mitochondrial bioenergetics by H2S is pertinent to its cellular signaling mechanism and fans out to other compartments via redox and metabolomic changes (
). SQOR activity has the potential to cause a reductive shift in the CoQ pool, particularly at H2S concentrations that partially or fully inhibit complex IV. H2S also indirectly perturbs the NAD+/NADH and FAD/FADH2 couples that are connected to CoQ/CoQH2via the ETC. We have previously demonstrated that H2S induces a reductive shift in the NAD+/NADH redox couple, creating an electron acceptor insufficiency that leads to uridine and aspartate deficiency and enhanced reductive carboxylation (
). While uridine limitation results from the CoQ dependence of dihydroorotate dehydrogenase in the pyrimidine pathway (Fig. 1A), aspartate deficiency results in part from reduced flux through the TCA cycle and the NADH-linked malate-aspartate shuttle. Furthermore, H2S stimulates the Warburg effect, enhancing glucose consumption and lactate production (
). The observed increase in succinate and decrease in malate at H2S concentrations that inhibit respiration were proposed to result from complex II reversal (
), neither model was evaluated experimentally. A recent study on oligomycin-treated murine microglia reported increased OCR upon exposure to an H2S donor and interpreted this as evidence of reverse electron transfer through complex I (
). The known drivers of mitochondrial reverse electron transfer, which leads to reactive oxygen species (ROS) generation, are a high membrane potential and an overreduced CoQ pool (
). Since respiratory poisons depolarize the mitochondrial inner membrane by limiting electron-coupled proton transfer (Fig. 1A), the premise for H2S-induced reverse electron transfer is unclear. Furthermore, the study contradicted the reported lack of H2S-induced ROS production (
). In this study, we report that rewiring within the ETC circuitry via complex II reversal prioritizes H2S oxidation under conditions of respiratory poisoning with fumarate serving as an electron acceptor. These results have important implications for understanding the mechanism by which intestinal epithelial cells respond to routine exposure to high H2S levels derived from the microbiota and potentially, the role of H2S in signaling a shift in energy metabolism.
Results
SQOR catalyzes sulfide-dependent reduction of O2
We examined whether O2 can serve an alternate electron acceptor for SQOR since complex IV poisoning by H2S should not restrict O2 availability (Fig. S1A). We found that when nanodisc-embedded SQOR (ndSQOR) (
) was reduced in the presence of sulfide and sulfite but in the absence of CoQ, O2 consumption was stimulated (Fig. S1B). From the linear dependence of OCR on O2 concentration, a kon of 3370 ± 290 M−1 s−1 was estimated (Fig. S1C). Oxygen (k ∼ 14 min−1 at 75 μM O2) is, however, a significantly less efficient electron acceptor than CoQ (15 × 103 min−1 at 75 μM CoQ) (
In the presence of a slight excess of sulfide (10 μM) and sulfite (15 μM), SQOR (7.5 μM) catalyzed the consumption of an equimolar concentration of O2 (7.3 ± 0.6 μM) (Fig. S1D). This reaction stoichiometry predicted that the products of O2 reduction by SQOR could be either O2•− and FADH• or H2O2 and FAD. The equivalence between the O2 consumed and the concentration of H2O2 produced (7.6 ± 0.6 μM) is consistent with the two-electron reduction of O2 by SQOR (Fig. S1A). The concentration of H2O2 was significantly diminished (0.2 ± 0.1 μM) when catalase was added to the reaction mixture. The approximately 1:1:1 stoichiometry of sulfide added:O2 consumed:H2O2 produced is consistent with electron transfer from FADH2 to O2via a C4a-hydroperoxy FAD intermediate (Fig. S1E), as proposed in other O2-activating flavoenzymes (
Complex I activity decreases the efficiency of H2S oxidation
Complex I-dependent oxidation of NADH with concomitant reduction of CoQ is a major source of electron flux in the ETC and is expected to influence the efficiency of H2S oxidation. We have previously reported that H2S causes a reductive shift in the NAD+/NADH ratio by inhibiting complex IV (
). H2S oxidation was unaffected by the cytoplasmic, but significantly enhanced by the mitochondrial expression of the water forming NADH oxidase, LbNOX (
) (Fig. 1B). Rotenone, a complex I inhibitor, increased H2S oxidation in control and LbNOX but not mito-LbNOX cells (Fig. 1B). Knockdown of NDUFS3 (Fig. S2), which is required for complex I assembly, increased H2S oxidation (Fig. 1C). Collectively, these data demonstrate that the cellular H2S oxidation capacity can be limited by the mitochondrial NADH pool.
The mitochondrial NADH and NADPH pools are interconnected via the activity of the electrogenic nicotinamide nucleotide transhydrogenase (NNT) located in the inner mitochondrial membrane. Cytoplasmic expression of TPNOX, a genetically encoded water forming NADPH oxidase (
), had no effect on H2S oxidation, while mitochondrial expression enhanced clearance (Fig. 1D). The NNT inhibitor NBD-Cl (4-chloro-7-nitrobenzofurazan chloride) attenuated the mito-TPNOX effect, further demonstrating that the capacity for cellular H2S oxidation is linked to the status of the mitochondrial NAD(P)H redox pool (Fig. 1D).
Succinate accumulates in response to H2S
Metabolomics analysis after exposure to Na2S (100 μM, 1 h) revealed a number of changes in glycolytic, TCA cycle (
), and purine metabolism intermediates in malignant HT29 cells (Fig. 2, A and B). Interestingly, H2S treatment led to ∼5.5-fold higher levels of succinate. To test whether succinate accumulation resulted from reversal of complex II activity (Fig. 2C), we used dimethyl fumarate (DMF), a membrane permeable derivative of fumarate that increases intracellular fumarate concentration (
). DMF accelerated H2S oxidation in four out of five colorectal carcinoma lines but not in RKO cells (Figs. 2D and S3). The molecular basis of the difference in response between RKO and the other cell lines is presently unclear. Two other complex II inhibitors, dimethyl malonate and dimethyl itaconate, also inhibited H2S clearance, while diethyl succinate did not (Fig. S4). Knocking down SDHA (Fig. S5), the complex II subunit that catalyzes the reversible oxidation of succinate to fumarate, reduced H2S clearance (Fig. 2E). DMF shortened the recovery time for return to basal OCR following respiratory inhibition by H2S in HT29 (Fig. 2, F–H), HCT116, LoVo, and DLD cells (Fig. S6) but had no effect when SDHA was knocked down in HT29 cells (Fig. S7). Together, these data are consistent with the model that H2S oxidation is facilitated by reversal of complex II activity.
Figure 2H2S induces succinate accumulation through reversal of complex II activity.A, volcano plot showing changes in select metabolite in response to Na2S (100 μM) treatment of HT29 cells for 1 h and represent a replot of the metabolomics data reported in (
). The gray and colored dots represent metabolites that exhibit statistically significant perturbations in response to H2S treatment. A subset of the metabolites that are pertinent to this study are highlighted in blue, pink, and red as noted. B, heat map showing H2S-induced changes in select metabolites observed in four independent technical repeats. C, scheme showing how complex II reversal can regenerate CoQ for H2S oxidation. D, DMF (100 μM) increases H2S oxidation in EV HT29 cells. E, SDHA knockdown in HT29 cells reduces H2S oxidation. F and G, the duration of respiratory inhibition in HT29 cells by H2S is longer in the absence (F) versus presence (G) of DMF (200 μM). The red arrows indicate when H2S (30 μM) was added. H, comparison of the time required by HT29 cells to return to the basal respiration rate ± DMF. The data in (D and E) represent the mean ± S.D. of 3 to 4 independent experiments.
The effect of complexes I and II on H2S-dependent OCR
To further test the influence of complexes I and II on the cellular response to H2S, OCR was monitored in control versus NDUFS3 and SDHA knockdown cells. NDUFS3 knockdown decreased basal OCR twofold (Fig. 3), consistent with complex I being a major entry point for electrons into the ETC. At a low concentration of H2S (10 μM), OCR activation in NDUFS3 knockdown cells was robust, and the peak increase in OCR was higher than in control and SDHA knockdown cells (Fig. S8). At a higher H2S (20 μM) concentration, differences between the cell lines were clearly visible (Fig. 3, A–C). While the NDUFS3 knockdown showed robust activation of OCR in response to H2S, the control and SDHA knockdown cells showed signs of inhibition. The SDHA knockdown cells also took a longer time to recover basal OCR compared with controls. Following the first and second 20 μM H2S injection, control and SDHA knockdown cells showed signs of partial and severe respiratory inhibition, respectively, in contrast to NDUFS3 knockdown cells. At a higher H2S concentration (30 μM), control and SDHA knockdown cells responded with net inhibition of oxygen consumption in comparison to NDUFS3 knockdown cells, which exhibited a mixed response (Fig. 3, D–F). These results indicate that the CoQ pool limits sulfide clearance and, in the absence of competition from complex I, cells clear sulfide more efficiently. The data also reveal that complex II has the opposite effect, i.e., it is advantageous for sulfide clearance, consistent with our model that complex II reversal supports H2S oxidation by catalyzing CoQH2 oxidation.
Figure 3Complexes I and II influence H2S-linked OCR. Comparison of OCR activation with H2S (20 or 30 μM) in (A and D) scrambled, (B and E) SDHA knockdown, and (C and F) NDUFS3 knockdown HT29 cells. Red arrows indicate when H2S was added. The traces are representative of 3 to 5 independent experiments.
), we tested whether they also contribute to fumarate when the ETC is inhibited by H2S. For this, GOT1 and GOT2 (glutamic-oxaloacetic aminotransferases 1 and 2) expressed in the cytoplasm and mitochondrion, respectively, were knocked down in HT29 cells (Fig. S9). GOT1 but not GOT2 knockdown increased H2S oxidation by ∼38% compared with control cells (Fig. 4C). GOT1 knockdown also promoted H2S clearance as reflected by the shorter recovery time to the basal respiration rate (Fig. S10). Inhibition of adenylosuccinate lyase with AICAR (5-aminoimidazole-4-carboxamide ribonucleotide) decreased H2S clearance by ∼50% (Fig. 4D), consistent with a role for the PNC in this process.
Figure 4The PNC and the malate-aspartate shuttle support fumarate-driven H2S oxidation.A and B, schemes showing that the malate-aspartate shuttle (A) and the PNC (B) are fumarate sources and that AICAR inhibits adenylosuccinate lyase (ASL). MDH1/2, FH, OAA, and CII denote malate dehydrogenase 1/2, oxaloacetate, fumarate hydratase, and complex II, respectively. C, H2S oxidation is stimulated in GOT1 knockdown but unaffected by GOT2 knockdown in HT29 cells. D, AICAR (0.25 mM) inhibits H2S clearance. The data in (C and D) represent the mean ± S.D. of 3 to 4 independent experiments. ns, not significant.
SDHA knockout in murine intestinal epithelial cells decreases H2S oxidation
To assess the physiological relevance of our observation that H2S clearance is supported by complex II working in reverse, we measured the impact of attenuating complex II on organismal H2S metabolism. For this, mice harboring loxP-flanked Sdha were crossed to mice expressing Cre recombinase under control of the villin promoter to specifically target intestinal epithelial cells, to generate Vil1-Cre Sdhafl/fl (SdhaΔIEC) mice as described previously (
). SdhaΔIEC mice showed significantly lower thiosulfate levels compared with control Sdhafl/fl (Fig. 5, B–D) revealing that the loss of complex II in intestinal cells caused local (feces) and systemic (serum and urine) perturbations in H2S oxidation.
Figure 5VillinCreSDHAfl/flmice have reduced thiosulfate levels.A, scheme connecting H2S oxidation to thiosulfate production. B–D, quantitation of thiosulfate levels in control (Sdhafl/fl) and VillinCreSdhafl/fl knockout mice in serum (B), urine (C), and feces (D). The data represent the mean ± S.D. for samples collected from four mice in each group.
In this study, we have uncovered a new mechanism for clearing H2S when its concentrations rise to levels that inhibit complex IV and preclude the use of O2 as the terminal electron acceptor for SQOR-dependent H2S oxidation. Such conditions might be relevant in the gut epithelium (where H2S exposure is high) or in ischemia (where O2 supply is cut off). Reversal of complex II activity under such conditions supports SQOR-dependent H2S oxidation, using fumarate as an alternate electron acceptor and prioritizes H2S clearance.
Metabolomic changes in HT29 cells in response to H2S provided clues to reprogramming driven changes that could potentially impact its clearance. Hypoxanthine and succinate, classic ischemic biomarkers (
), the effect of the oxidative TCA cycle on H2S clearance was not examined. The PNC and the malate aspartate shuttle both impacted H2S clearance (Fig. 4, C and D). The PNC is activated in response to a drop in the adenylate energy charge (
) as well as the observed increase in inosine, which is formed via deamination of adenosine.
Knockdown of GOT1, but not GOT2, increased the efficiency of H2S clearance, suggesting that the cytoplasmic arm of the malate-aspartate shuttle is an important source of fumarate. H2S leads to aspartate deficiency (
), potentially stimulating GOT1-catalyzed transamination of oxaloacetate to aspartate rather than the reverse, which is consistent with lower malate levels in H2S-treated cells (Fig. 2B). In GOT1 knockdown cells, oxaloacetate should be more available for malate dehydrogenase catalyzed reduction to malate, which can be dehydrated to fumarate (Fig. 4A) by fumarate hydratase that is present in the cytoplasm and the mitochondrion (
Our studies support a model for efficient H2S clearance by SQOR when the H2S concentration is low with complexes I and II competing for the CoQ pool and complex III recycling CoQH2 (Fig. 6A). However, when H2S concentrations rise and inhibit complex IV, utilization of fumarate as an electron acceptor by complex II sustains recycling of CoQH2 (Fig. 6B). Complex II catalyzes the reversible oxidation of succinate to fumarate (
). Under in vitro assay conditions, the ratio of succinate oxidation to fumarate reduction catalyzed by the succinate dehydrogenase component of complex II varies substantially with the electron acceptor and ranges from ∼0.1 to 50 for succinate:fumarate consumed (
). Under physiological conditions, flux through the forward versus reverse reaction is governed by the concentration of the respective substrates and by the potentials of the relevant redox couples. In the mitochondrial matrix (pH ∼ 7.7), the standard redox potential for the fumarate/succinate couple (E˚′ = +30 mV) is similar to that for ubiquinone/ubiquinol (+40–60 mV at pH 7.0, decreasing 60 mV per increase in pH unit (
)). The reversibility of complex II in cells is supported by its ability to sustain proficient growth on fumarate as a terminal electron acceptor when expressed under anaerobic conditions in an Escherichia coli strain lacking fumarate reductase (
Anaerobic expression of Escherichia coli succinate dehydrogenase: Functional replacement of fumarate reductase in the respiratory chain during anaerobic growth.
). These data support the plausibility of complex II reversal under conditions when the ETC is blocked, and the CoQ pool is overreduced.
Figure 6Alternate redox cycles for disposing H2S.A and B, CoQH2 formed during H2S oxidation and by complexes I and II enters the ETC at the level of complex III (A). When complex IV is inhibited by H2S, blocking recycling of CoQH2 by complex III, CoQH2 can be oxidized by complex II, concomitant with fumarate reduction and succinate accumulation (B).
Modulation of H2S metabolism by complex I was demonstrated by its inhibition by rotenone and by NDUFS3 knockdown, both enhanced H2S clearance (Fig. 1, B and C), as expected, and is consistent with their increased sulfide-induced OCR compared with control cells (Fig. 3). On the other hand, SDHA knockdown decreased the efficiency of H2S clearance while DMF increased it (Figs. 2 and S3). Under conditions of complete coupling, for every mole of sulfide oxidized by SQOR, ETHE1 and complex IV are predicted to consume 1 and 0.5 mol of O2, respectively. ETHE1 is a mononuclear iron-dependent persulfide dioxygenase, which catalyzes the conversion of glutathione persulfide to sulfite (
Spectroscopic and electronic structure study of ETHE1: Elucidating the factors influencing sulfur oxidation and oxygenation in mononuclear nonheme iron enzymes.
). SDHA knockdown cells exhibited increased sensitivity to H2S-induced inhibition of OCR and took longer to recover, while DMF reduced the time to recovery of the basal OCR (Figs. 2 and 3). Collectively, these results support our model of complex II-dependent recycling of CoQH2 (Fig. 6B). It is important to note, however, that interfering with complex II reduces but does not completely block H2S consumption. Thus, other mechanisms including SQOR-dependent reduction of O2 (Fig. S1) might contribute to H2S removal.
The significant decrease in thiosulfate upon silencing SDHA in murine intestinal epithelial cells (Fig. 5) is notable for three reasons. It supports the physiological relevance of reverse complex II activity for H2S oxidation as loss of the canonical succinate oxidation activity would be expected to stimulate SQOR-dependent H2S oxidation by decreasing competition for the CoQ pool. Second, the observed change in thiosulfate levels in SdhaΔIEC mice reflects the quantitatively significant impact of complex II activity in intestinal epithelial cells on systemic sulfide metabolism. Third, changes in urine and serum thiosulfate in SdhaΔIEC mice reveal the systemic impact of altered H2S metabolism at the host–microbe interface, which warrants further study.
We speculate that H2S-fueled succinate accumulation could have downstream metabolic effects. Succinate is a competitive inhibitor of α-ketoglutarate-dependent dioxygenases and its accumulation could broadly impact histone and DNA methylations (
Inhibition of alpha-KG-dependent histone and DNA demethylases by fumarate and succinate that are accumulated in mutations of FH and SDH tumor suppressors.
), could be enhanced by H2S-driven succinate accumulation. Over 750 protein targets of succinylation have been identified, which are concentrated in mitochondria but also present in other compartments (
). We speculate that succinylation could be enhanced by the opposing effects of H2S on the succinate and NAD+ pools, in an autocorrective loop for activating complex II and prioritizing its removal.
In summary, our study reveals that metabolic reprogramming leads to the establishment of a new redox cycle between SQOR and complex II, permitting sustained H2S clearance. In addition to its relevance at the gut host–microbe interface, this circuitry could be important in the context of ischemia reperfusion injury. H2S is cytoprotective when administered at the time of reperfusion, reducing infarct size, inhibiting myocardial inflammation, and preserving mitochondrial integrity (
). We posit that the cytoprotective effects of H2S could derive from its twin effects on complex IV inhibition and complex II reversal, thereby attenuating succinate-dependent ROS generation during reperfusion. Another cellular context in which H2S-mediated ETC rewiring might be relevant is during the transition from a quiescent to proliferative state. While quiescent cells primarily rely on the high energy yield of oxidative phosphorylation, proliferating cells increase aerobic glycolysis to meet their energy needs and redirect mitochondrial metabolism for macromolecular precursor synthesis (
). The potential for H2S to function as an endogenous modulator of energy metabolism could be significant in this context and needs to be further understood.
Conclusions
Colonocytes are routinely exposed to H2S derived from microbial metabolism and are adapted to remove high concentrations of this toxic gas via a mitochondrial sulfide oxidation pathway that links to the electron transport chain. We have discovered that cells prioritize the removal of H2S when its levels are high enough to inhibit respiration, by utilizing fumarate as an alternate electron acceptor. Specifically, a new redox circuitry is established between SQOR, which reduces CoQ as it oxidizes H2S, and complex II, working in reverse to regenerate CoQ as it reduces fumarate. Mice with targeted deletion of complex II in intestinal epithelial cells exhibit systemic reduction in H2S oxidation, establishing physiological relevance of this redox circuitry and revealing a quantitatively significant contribution of colonocytes to whole-body sulfide homeostasis.
Experimental procedures
Materials
Sodium sulfide nonahydrate (431648), sodium sulfite (S0505), sodium selenite (S5261), CoQ (C7956), dimethyl malonate (63380), dimethyl itaconate (592498), diethyl succinate (8.00680), rotenone (R8775), dimethyl fumarate (242926), 4-chloro-7-nitrobenzofuran (163260), doxycycline (D3447), puromycin (P8833), protease inhibitor cocktail for mammalian tissue extract (P8340), RIPA lysis buffer (R0278), and apo-transferrin (T1147) were from Sigma. RPMI 1640 (11875-093), DMEM (11995-065), FBS (10437-028), trypsin-EDTA (25300-054), penicillin-streptomycin (15140-122), geneticin (10131-035), M199 (11150-059), epidermal growth factor (PHG0311), PBS (10010-023), DPBS (14040-133), and insulin (12585014) were from Gibco. Anti-Flag (20543-1-AP), anti-NDUFS3 (15066-1AP), anti-SDHA (14865-1AP), anti-GOT1 (14886-1AP), and anti-GOT2 (14800-1AP) antibodies were from Proteintech, and the secondary anti-rabbit horseradish peroxidase-linked IgG antibody (NA944V) was from GE Healthcare.
Assays for ndSQOR-catalyzed O2 consumption and H2O2 production
Human SQOR was purified and embedded in nanodiscs as described previously (
). O2 consumption by FADH2 in ndSQOR was monitored using an O2k respirometer (Oroboros Instruments), equipped with two polarographic O2-sensing electrodes housed in separate 2 ml chambers. Each chamber was filled with 100 mM potassium phosphate, pH 7.4, and sulfide (100 μM) and sulfite (200 μM) were added before sealing the chambers and pre-incubating for ∼5 min at 25 °C. The reaction was initiated by injecting ndSQOR (100 nM) and monitored over a period of ∼10 min. Initial O2 concentrations were varied by aerating N2-purged buffer in the chambers before sealing when the desired O2 concentration was reached. H2O2 production was assayed using the Pierce Quantitative Peroxide Assay Kit (Thermo Fisher) according to the manufacturer’s protocol.
Cell culture
HT29 cells were maintained in RPMI 1640 medium. HCT116, LoVo, DLD, and RKO were maintained in DMEM medium. Both RPMI and DMEM media were supplemented with 10% FBS, 100 units/ml penicillin, and 100 μg/ml streptomycin. HCEC cells were cultured as described previously (
). All cells were maintained at 37 °C with ambient O2 and 5% CO2 except HCEC, which were maintained at 2% O2 and 5% CO2.
Ectopic expression of LbNOX and TPNOX
LbNOX and mito-LbNOX and pINDUCER empty vector were obtained from Addgene. The pLVX-TRE3G empty vector, TPNOX, mito-TPNOX, and pLVX TET ON were a generous gift from Dr. Valentin Cracan (Scintillon Institute). The construction of HT29 cell lines stably expressing LbNOX, mito-LbNOX, TPNOX, and mito-TPNOX has been described previously (
). Before the start of an experiment, these cells were incubated for 24 h with 300 ng/ml doxycycline to induce LbNOX expression. The cells were routinely cultured in RPMI 1640 medium supplemented with 10% FBS, 100 units/ml penicillin, 100 μg/ml streptomycin and 300 μg/ml geneticin, and 1 μg/ml puromycin.
Generation of shRNA-mediated knockdown cells
NDUFS3 and SDHA were targeted for knockdown using shRNA purchased from MISSION shRNA Library, Sigma. The clone IDs for NDUFS3 were NM_004551.1-320s21c1 and NM_004551.1-628s21c1. The clone IDs for SDHA were NM_004168.1-619s1c1 and NM_004168.1-1643s1c1. The doxycycline-inducible GOT1 and GOT2 lentiviral constructs were subcloned into the iDox-pLKO vector as described previously (
Biochemical characterization and structure-based mutational analysis provide insight into the binding and mechanism of action of novel aspartate aminotransferase inhibitors.
). Plasmids containing shRNA against specific genes or a scrambled sequence were submitted to the Vector Core (University of Michigan) for lentiviral packaging. For lentiviral infection, 7.5 × 104 HT29 cells were seeded in a six-well plate containing 2 ml per well of RPMI 1640 medium supplemented with 10% FBS, 100 units/ml penicillin, and 100 μg/ml streptomycin. The transduction and selection protocols were the same as described for LbNOX (
), and cells were selected with 1 μg/ml puromycin.
Western blotting
TPNOX expression in HT29 cells was monitored by growing cells in a six-well plate for 24 h in RPMI 1640 medium as described above followed by a 24 h incubation with 300 ng/ml doxycycline. Then, the cells were washed with PBS twice before addition of 250 μl of RIPA lysis buffer containing 10 μl/ml protease inhibitor cocktail for mammalian tissue extracts and collected by scraping. Cells were frozen and thawed three times and centrifuged at 12,000g for 5 min. The protein concentration in the supernatant was measured using Bradford reagent (Bio-Rad). Protein lysates were similarly prepared from cells in which NDUFS3, SDHA, and GOT1/2 were knocked down. Following separation by 10% SDS PAGE, proteins were transferred to a PVDF membrane and incubated overnight at 4 °C with primary anti-Flag antibody at a dilution of 1:1000 for TPNOX. Antibodies against NDUFS3, anti-SDHA, GOT1, and GOT2 (14800-1AP) were used at a dilution of 1:2000. Horseradish-peroxidase-linked anti-rabbit IgG was used at a dilution of 1:10,000. Membranes were developed and visualized using the KwikQuant Digital-ECL substrate and imaging system.
Cellular H2S consumption assay
Cells were grown to ∼90% confluency in 10 cm plates and on the day of experiment, washed with PBS and treated with 0.05% trypsin-EDTA (for ∼10 min at 37 °C). Then, cells were resuspended in 10 ml complete media and centrifuged for 5 min at 4 °C, 1700g. The cell pellet was resuspended in 1 ml modified DPBS (supplemented with 20 mM HEPES, pH 7.4, and 5 mM glucose) in a preweighed Eppendorf tube and centrifuged for 5 min at 4 °C, 1700g. The supernatant was discarded, and the pellet weight was determined. Cells were suspended in modified DPBS to make a 5% cell suspension (w/v) in a 1 ml Eppendorf tube. When the effects of dimethyl malonate (DMM,10 mM) or dimethyl itaconate (DMI, 0.25 mM) were tested, cells were preincubated for 3 h with each reagent before making a 5% cell suspension in which the same concentrations of DMM and DMI were included followed by addition of 100 μM Na2S. Alternatively, when dimethyl fumarate (DMF, 100 μM) and diethyl succinate (DES, 5 mM) were tested, these reagents were added to a 5% cell suspension in modified DPBS for 5 min prior to addition of 100 μM Na2S. The suspension cultures were incubated at 37 °C with shaking (75 rpm). Samples (45 μl) were collected at time 0 and 10 min, mixed with 1 M Tris base (2.5 μl), and stored in dry ice. Control samples containing 10 mM DMM, 0.25 mM DMI, 100 μM DMF, or 5 mM DES and 100 μM Na2S in modified DPBS were incubated in parallel, and the concentration of H2S lost from these samples was subtracted from the values obtained from the cell suspension samples containing the same reagents.
Monobromobimane derivatization of sulfide and HPLC analysis
The samples from the H2S consumption assay described above were thawed and mixed with 2.5 μl of 60 mM monobromobimane (in DMSO) and incubated in the dark at room temperature for 10 min followed by addition of 100 μl of metaphosphoric acid solution (16.8 mg/ml). The samples were vortexed and centrifuged for 5 min at 4 °C and 10,000g. The supernatant was collected in the dark and stored at −20 °C until further use. The samples were analyzed using a Zorbax Eclipse XDB-C18 column (5 μm, 4.6 × 150 mm, Agilent) as described previously (
). Peaks were detected using excitation at 390 nm and fluorescence emission at 490 nm. A calibration curve with known concentrations of sodium sulfide was used to determine the concentration of H2S in samples.
Metabolomics analysis
Metabolomics analysis on HT29 cells treated ±100 μM Na2S for 1 h was performed as described previously (
Oxygen consumption was measured using the O2k respirometer. Cells were grown to ∼90% confluency in 10 cm plates and on the day of experiment, washed with PBS, and then trypsinized with 1.5 ml of 0.05% trypsin-EDTA for ∼10 min at 37 °C. Then, the cells were resuspended in 10 ml of complete medium and centrifuged for 5 min at 1700g, 4 °C. The cell pellet was resuspended in 1 ml of modified DPBS in a preweighed Eppendorf tube, the suspension was centrifuged for 5 min at 1700g, and the weight of the pellet was recorded. The cells were suspended in modified DPBS to make a 5% cell suspension (w/v), which was stored on ice. At the start of the experiment, the cell suspension was diluted to 1% or 1.5% (for NDUFS3 knockdowns which showed lower basal OCR). The cell suspension was placed in the respirometer chamber and the OCR was allowed to stabilize over ∼15 to 20 min at 37 °C with constant stirring at 750 rpm. Na2S (from a freshly prepared 10 mM stock solution in water) was injected into the sample to give the desired final concentration (10–30 μM) per injection.
Mice
B6.Cg-Tg(Vil-cre)1000Gum/J mice were purchased from the Jackson Laboratory. C57BL/6N-Sdhatm2a(KOMP)Wtsi mice were obtained from the Knock Out Mouse Project (KOMP) repository, University of California, Davis and bred to ACTFLPe mice to excise the FRT-flanked region. The resulting Sdhafl/fl mice were bred to Vil1-Cre mice to create Vil1-Cre Sdhafl/fl (SdhaΔ/IEC) mice (
). Then, 12 to 15 week-old mice were used in our experiments. The mice were maintained under specific pathogen-free conditions following procedures approved by the University of Michigan Committee on the Use and Care of Animals, which are based on the University of Michigan Laboratory Animal Medicine guidelines.
Statistical analysis
Statistical analyses were performed using GraphPad Prism 9. Two-tailed tests were used for all t-tests. Errors on measurements are represented as standard deviation.
Data availability
All data are contained within the manuscript and in the supplemental section.
C. A. L. is a consultant for Astellas Pharmaceuticals and is an inventor on patents pertaining to Kras regulated metabolic pathways, redox control pathways in pancreatic cancer, and targeting the GOT1-pathway as a therapeutic approach.
Author contributions
R. K. and A. P. L. conceptualization; R. K., A. P. L., A. G., V. V., H. J. L., C. A. L., K. S., and P. R. data curation; R. K., A. P. L., A. G., V. V., H. J. L., and C. A. L. formal analysis; R. K., A. P. L., and R. B. writing—review and editing.
Funding and additional information
This work was supported in part by the grants from the National Institutes of Health (GM130183 to R. B., NCI R01CA244931 to C. A. L. and HL152605; HL149633; CA203542 to P. R.) and the American Heart Association (826245 to R. K.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Anaerobic expression of Escherichia coli succinate dehydrogenase: Functional replacement of fumarate reductase in the respiratory chain during anaerobic growth.
Spectroscopic and electronic structure study of ETHE1: Elucidating the factors influencing sulfur oxidation and oxygenation in mononuclear nonheme iron enzymes.
Inhibition of alpha-KG-dependent histone and DNA demethylases by fumarate and succinate that are accumulated in mutations of FH and SDH tumor suppressors.
Biochemical characterization and structure-based mutational analysis provide insight into the binding and mechanism of action of novel aspartate aminotransferase inhibitors.
High levels of H2S produced by gut microbiota can block oxygen utilization by inhibiting mitochondrial complex IV. Kumar et al. have shown how cells respond to this inhibition by using the mitochondrial sulfide oxidation pathway and reverse electron transport. The reverse activity of mitochondrial complex II (succinate-quinone oxidoreductase, i.e., fumarate reduction) generates oxidized coenzyme Q, which is then reduced by the mitochondrial sulfide quinone oxidoreductase to oxidize H2S. This newly identified redox circuitry points to the importance of complex II reversal in mitochondria during periods of hypoxia and cellular stress.
01244-8&id=gr1.jpg){kind=link}
01244-8&id=gr2.jpg){kind=link}
01244-8&id=gr3.jpg){kind=link}
01244-8&id=gr4.jpg){kind=link}
01244-8&id=gr5.jpg){kind=link}
01244-8&id=gr6.jpg){kind=link}