Sodium Thiosulfate Attenuates Acute Lung Injury in Mice

ACUTE lung injury (ALI) is characterized by lung inflammation and increased pulmonary vascular permeability.¹  Sepsis is a major cause of ALI,²  and lipopolysaccharide, a cell wall component of Gram-negative bacteria, can reproduce the features of human ALI in mice.³  Studies have revealed that vascular endothelium plays a crucial role in mediating inflammatory response in the lung.^4,5  Therefore, the pulmonary vascular endothelium represents one of the major targets of therapy.⁶ 

Hydrogen sulfide (H₂S) is a reactive gaseous mediator. In mammalian tissues, H₂S is serially oxidized to persulfide, sulfite (SO₃²⁻), thiosulfate (S₂O₃²⁻), and sulfate (SO₄²⁻).⁷  In addition, H₂S may be stored as sulfane sulfur–containing polysulfides in cells.⁸  Although H₂S can exert a host of biological effects on various targets,⁹  it is currently unknown whether the biological effects of H₂S are mediated directly by H₂S itself or its metabolites.⁸  In circulation, reaction with plasma proteins or oxidation maintains free plasma H₂S levels very low.¹⁰  Free H₂S levels only transiently increase and quickly return to its baseline after systemic administration of H₂S donor compounds (e.g., Na₂S [sodium sulfide] or NaHS [sodium hydrosulfide]).

We recently reported that the protective effects of inhaled H₂S on mice subjected to lethal lipopolysaccharide challenge are associated with an increased plasma thiosulfate levels.¹¹  Furthermore, intraperitoneal administration of sodium thiosulfate (STS) improved survival after endoxemia¹¹  and acute liver failure.¹²  These observations suggest that thiosulfate may be one of the “carrier” molecules that mediate antiinflammatory effects of H₂S. STS has been used for decades as an antidote against cyanide poisoning.¹³  STS has also been used for the treatment of calciphylaxis,¹⁴  vascular calcifications,^15,16  and cisplatin-induced cytotoxicity.¹⁷  Therefore, if antiinflammatory effects of STS are confirmed, then it is highly clinically relevant and readily translatable. Although H₂S has been shown to mitigate lung injury^18,19  and vascular endothelial dysfunction in a variety of animal models,^20,21  effects of STS against ALI remain to be determined. Furthermore, mechanisms responsible for the antiinflammatory effects of STS were not investigated in our previous studies.^11,12  It is possible that thiosulfate protects vascular endothelium from inflammatory insults.

H₂S appears to exert antiinflammatory effects at least in part via inhibition of nuclear factor-κB (NFκB)–dependent signaling pathway.^11,22  Upon binding of lipopolysaccharide to the toll-like receptor 4, tumor necrosis factor receptor–associated factor 6 (TRAF6) is recruited to the receptor complex, which facilitates lysine 63 (K63)–linked polyubiquitination of TRAF6.^23,24  Polyubiquitinated TRAF6 induces phosphorylation and activation of transforming growth factor-β–activated kinase 1 (TAK1). TAK1 then activates inhibitor of NFκB kinase (IKK), resulting in NFκB activation.²⁴  Thus, inhibition of TRAF6 ubiquitination has been suggested as a target to modulate NFκB signaling pathway.^25,26 

The objective of the current study is to examine the effects of STS in ALI. We hypothesized that STS prevents ALI via inhibition of NFκB signaling in pulmonary vascular endothelium. We observed that STS inhibited ALI and inflammation after intratracheal lipopolysaccharide challenge and cecal ligation and puncture (CLP). These results highlight the therapeutic potential of STS in the treatment of ALI.

Male C57BL6J mice (Jackson Laboratories, Bar Harbor, ME), 8 to 10 weeks old, 19 to 27 g body weight, were used in this study. Protocols for animal use were approved by the Massachusetts General Hospital Subcommittee on Research Animal Care (Boston, MA). All animal experiments were performed in accordance with the guidelines of the National Institutes of Health (Bethesda, Maryland).

Mice were challenged with 2 mg/kg of lipopolysaccharide (O111: B4; Sigma, St. Louis, MO) in 50-μl saline via intratracheal route as an aerosol using a microsprayer (Penn-Century, Philadelphia, PA) with or without intraperitoneal administration of 2 g/kg STS at 0 and 12 h after intratracheal lipopolysaccharide. Control mice received 50 μl intratracheal saline. Mice breathed spontaneously in ambient air, and bronchoalveolar lavage fluid (BALF) was collected 24 h after saline or lipopolysaccharide challenge. We chose this time point because we found that inflammatory reaction peaks at 24 h after lipopolysaccharide challenge at this dose in pilot studies. Total number of leukocytes in BALF was counted using a hemocytometer. The BALF samples were subsequently centrifuged and cytospin samples were prepared from the cell pellet. Cytoslides were stained with Kwik-Diff stain kit (Thermo Shandon, Pittsburgh, PA). The number of polymorphonuclear neutrophils (PMNs) was determined by the cell counting of neutrophil fraction. Supernatant samples were used to quantify the protein levels using Pierce BCA protein assay kit (Thermo Scientific, Rockford, IL) or stored at −80°C for cytokine and myeloperoxidase activity analysis.

Mice were subjected to CLP as previously described.²⁷  Mice were anesthetized and a midline abdominal incision was made to expose cecum. The cecum was ligated at 1.0 cm from the tip then punctured twice with an 18-gauge needle. A small amount of its contents were extruded before the cecum was returned into the abdominal cavity, and then the abdominal incision was closed in layers. Sham-operated controls underwent laparotomy without CLP. All mice were given fluid resuscitation with prewarmed sterile saline (50 ml/kg) subcutaneously. STS (0.5 g/kg) or saline was intravenously injected via tail vein 10 min after CLP. Lungs were harvested 8 h after CLP, homogenized with 0.5% Triton X-100/phosphate-buffered saline, and centrifuged, and then the supernatants were subjected to cytokine measurement after equalization by the protein concentration.

Enzyme-linked immunosorbent assay kits (R&D Systems, Minneapolis, MN) were used for the measurement of mouse tumor necrosis factor-α (TNFα), interleukin (IL)-6, and myeloperoxidase levels in BALF according to the manufacturer’s instruction.

Mouse lungs were harvested 24 h after lipopolysaccharide challenge, blotted dry, and weighed immediately to obtain the wet lung weight. Lungs were dried in a 70°C incubator for 3 days to obtain dry lung weight, and then wet dry lung ratio was calculated.

Mouse lungs were inflated under a pressure of 23 cm H₂O with 4% paraformaldehyde (Boston Bio Products, Ashland, MA) for histologic evaluation by hematoxylin and eosin staining as previously described.²⁸  Twenty high-power fields (×400 magnification) were taken per mouse, and lung injury was graded using a modified ALI score in each high-power field. Each of three categories, such as (1) thickness of the alveolar walls, (2) infiltration of inflammatory cells, and (3) hemorrhage, was graded in a blinded manner according to the following scale: 0 = minimal damage; 1 = mild damage; 2 = moderate damage; 3 = severe damage; and 4 = maximal damage. The degree of lung damage was assessed by the total of scores ranging from 0 to 12.

Mouse lungs were harvested 24 h after lipopolysaccharide or saline challenge, and total RNA was extracted using the RNAspin mini kit (GE Healthcare, Piscataway, NJ). Complementary DNA was synthesized with Moloney Murine Leukemia Virus Reverse Transcriptase (Promega, Madison, WI), and RNA transcript levels were measured using a Mastercycler Realplex system (Eppendorf North America, Westbury, NY). The primer sequences are listed in table 1. Gene expression was normalized to 18S ribosomal RNA level. The mean value of control mice was set as 1.

To determine the impact of intraperitoneal administration of STS, concentrations of hydrogen sulfide and thiosulfate in lung and plasma were measured by high-performance liquid chromatography as previously described.¹¹  At 2 h after lipopolysaccharide challenge, blood was drawn and lung was dissected. Plasma and lung homogenates were added to 70 μl of 10 mM Tris-HCl buffer (pH 9.5, 0.1 mM diethylenetriamine pentaacetic acid), followed by addition of 50 μl of 200 mM 5-sulfosalicylic acid after 30 min. The mixture was centrifuged and supernatant was analyzed by high-performance liquid chromatography with a fluorescence detector (Waters, Milford, MA).

Human umbilical vein endothelial cells (HUVEC) and human lung microvascular endothelial cells (HMVEC-L) were purchased from Lonza (Walkersville, MD). The cells were cultured in EGM-2 or EGM-2 MV (Lonza) supplemented with 2 or 5% fetal bovine serum and endothelial cell growth factors and used between passages 3 to 5 for all experiments.

Human umbilical vein endothelial cells or HMVEC-L were seeded in 96-well plates at 1 × 10⁴ cells per well and incubated with lipopolysaccharide (10 μg/ml) or recombinant human TNFα (10 ng/ml) (R&D Systems) with or without STS. Cell culture supernatants were collected after incubation for 20 h with lipopolysaccharide or TNFα. Human IL-6 and IL-8 levels were measured by enzyme-linked immunosorbent assay (R&D Systems) according to the manufacturer’s instruction.

Chloromethyl-2’, 7’-dichlorodihydrofluorescein diacetate (CM-H₂DCFDA; Invitrogen, Eugene, OR) was used to measure intracellular production of general reactive oxygen species (ROS) such as H₂O₂ (hydrogen peroxide), hydroxyradical, and peroxides. HUVEC were seeded at 1 × 10⁴ cells per well in 96-well plate and cultured to confluent. Cells were loaded with 10 μM of CM-H₂DCFDA for 30 min. Then cells were washed twice, lipopolysaccharide (10 μg/ml) or TNFα (10 ng/ml) was added with or without STS for 30 min, followed by fluorescence measurement at excitation and emission wavelengths of 480 and 530 nm. The Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Invitrogen) was also used to measure H₂O₂ and peroxide production. After treatment with lipopolysaccharide or TNFα, 100 μM of Amplex Red reagent solution (Invitrogen) was added to each well. The cells were treated for 30 min, and the fluorescence intensity was measured at excitation and emission wavelengths of 560 and 590 nm.

Protein extracts were prepared by lysing cells or homogenizing lung tissue in radio-immunoprecipitation assay buffer (Boston BioProducts) supplemented with protease inhibitor cocktail (Roche, Indianapolis, IN) and phosphatase inhibitor cocktail (Sigma). The samples were subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membrane (Millipore, Billerica, MA), and immunoblotted. Antibodies used were as follows: TRAF6 (H274), TAK1 (M579) from Santa Cruz (Dallas, TX), K63-linkage–specific ubiquitin (HWA4C4) from Millipore, phospho-TAK1 (4531), phospho-IKKα/β (2697), IKKβ (2678), phospho-NFκB p65 (3033), NFκB p65 (8242), inhibitor of NF-κBα (IκBα) (4812), phospho-IκBα (2859), glyceraldehyde-3-phosphate dehydrogenase (5174), and β-tubulin (2146) from Cell Signaling (Danvers, MA). In densitometric analysis, the mean value of control group was set as 1.

Cell lysates were incubated with TAK1 or TRAF6 antibodies with Protein G Mag Sepharose (GE Healthcare) at 4°C for overnight. The complexes were washed four times with Tris buffered saline with Tween20 and eluted in sodium dodecyl sulfate–sample buffer (Boston Bioproducts) at 90°C for 2 min. For TRAF6 immunoprecipitation, cells were pretreated with 10 μM MG132 (Sigma) for 4 h before treatment with lipopolysaccharide and STS. Cells were lysed in radio-immunoprecipitation assay buffer supplemented with protease inhibitor cocktail (Roche) and 10 mM N-ethylmaleimide.

Two novel fluorescent probes, HSip-1DA²⁹  and SSP4, an improved version of SSP2,³⁰  were used for the detection of H₂S and sulfane sulfur, respectively. SSP4 was prepared using the same method reported previously.^30 1H NMR (300 MHz, DMSO-d₆) δ 5.49 (s, 2H), 6.97 (d, J = 9.0 Hz, 2H), 7.12 (d, J = 9.0 Hz, 2H), 7.32 (t, J = 6.0 Hz, 2H), 7.44 to 7.55 (m, 5H), 7.66 (d, J = 9.0 Hz, 2H), 7.76 to 7.88 (m, 2H), 8.09 (d, J = 9.0 Hz, 1H), 8.19 (d, J = 6.0 Hz, 2H). ¹³C NMR (75 MHz, CD₃Cl) δ 81.9, 110.9, 116.9, 118.2, 124.3, 124.7, 125.2, 125.5, 129.3, 131.4, 132.5, 133.7, 140.1, 151.8, 152.2, 153.2, 164.9, 169.4; MS (ESI⁺) m/z 627.6 (M+Na⁺). HUVEC were seeded at 1 × 10⁴ cells per well in 96-well plate and cultured to confluent. Cells were loaded with 30 μM of HSip-1DA, a cell membrane-permeable derivative of HSip-1, or 50 μM of SSP4 for 30 min. Then cells were washed twice and treated with or without lipopolysaccharide and STS for 20 h, followed by fluorescence measurement at excitation and emission wavelengths of 490 and 515 nm. Intracellular H₂S and sulfane sulfur levels were determined by relative fluorescence intensities normalized to the levels of untreated control at 1 h after treatment.

All data are presented as means ± SD. We did not conduct an a priori statistical power calculation. We estimated our sample size based on our previous studies in which the effects of STS were examined in sepsis.¹¹  No randomization methods were used to assign animals or cells to each treatment. Treatment conditions were alternated between samples in an effort to maintain experimental conditions as constant as possible among groups. Two-tailed hypothesis testing method was used throughout this study. Data were analyzed by one-way or two-way ANOVA followed by Bonferroni post hoc comparisons tests as required using GraphPad Prism 5 (GraphPad Software, La Jolla, CA). Thiosulfate and sulfide levels of plasma and lung were analyzed by Mann–Whitney U test because the values were not normally distributed. P values of less than 0.05 were considered significant.

The number of cells in BALF was evaluated to examine the effects of STS on lipopolysaccharide-induced lung inflammation and permeability. The recovery rate of BALF was more than 90% in all groups. In lipopolysaccharide-challenged mice, a marked increase in the number of total cells and PMNs in BALF was observed (fig. 1, A and B). However, STS significantly decreased the influx of cells into the alveolar space after lipopolysaccharide challenge. STS treatment alone did not affect the number of cells in BALF. Myeloperoxidase levels, a marker of PMN infiltration, were also measured by enzyme-linked immunosorbent assay. In accordance with the result of PMNs influx into the lung, myeloperoxidase levels were significantly increased in lipopolysaccharide-challenged mice, whereas it was attenuated by STS administration (fig. 1C). We also evaluated the effect of STS on lung vascular leak by measuring BALF protein and lung wet dry weight ratio. Lipopolysaccharide challenge induced a significant increase in BALF protein concentrations and lung wet dry ratio (fig. 1, D and E). STS attenuated the pulmonary vascular leakage and lung edema in lipopolysaccharide-challenged mice. To evaluate inflammatory mediators recruiting PMNs to the lung, we measured cytokine levels in BALF by enzyme-linked immunosorbent assay. STS significantly decreased the lipopolysaccharide-induced IL-6 and TNFα increase in the BALF (IL-6, TNFα [pg/ml]; 49 ± 12, 57 ± 6, control [n = 8]; 313 ± 164, 336 ± 156, lipopolysaccharide [n = 10]; 79 ± 27, 76 ± 88, lipopolysaccharide + STS [n = 10]; fig. 1, F and G). Furthermore, intravenous STS attenuated IL-6 induction in the lung 8 h after CLP (pg ml⁻¹ g⁻¹ protein; 95 ± 33, control [n = 8]; 424 ± 256, CLP [n = 12]; 241 ± 128, CLP + STS [n = 12]; fig. 1H). These results suggest that STS exerts potent antiinflammatory effects and prevents the increase in lung permeability after lipopolysaccharide challenge or polymicrobial sepsis.

Histologic assessment revealed that lipopolysaccharide stimulated a marked influx of PMNs to the alveolar space. Administration of STS attenuated the PMN infiltration into the lung (fig. 2, A and B). Semiquantitative analysis of lung sections by lung injury score demonstrated that lipopolysaccharide-induced lung injury was attenuated by STS treatment (fig. 2C).

We measured the messenger RNA expression levels of proinflammatory mediators, including cytokines, chemokines, and adhesion molecules, in the whole lung of mice after lipopolysaccharide challenge (fig. 3). STS attenuated lipopolysaccharide-induced up-regulation of cytokines (TNFα, IL-6, and IL-1β), intracellular adhesion molecule-1, chemokines, and nitric oxide synthase 2. STS also attenuated the lipopolysaccharide-induced up-regulation of matrix metalloproteinase 9, a family of proteinases that remodel extracellular matrix components, but not matrix metalloproteinase 2.

Activation of IKK by lipopolysaccharide induces phosphorylation and degradation of IκB, leading to the nuclear translocation of NFκB and transcriptional activation.²³  To elucidate the molecular mechanisms involved in the attenuated inflammatory responses by STS, we examined the effects of STS on IκB/NFκB p65 pathway. STS inhibited lipopolysaccharide-induced IκB phosphorylation and p65 nuclear translocation in mice lung 24 h after lipopolysaccharide challenge (fig. 4). These results suggest that STS inhibits lipopolysaccharide-induced activation of NFκB signaling by inhibiting IκB phosphorylation.

Administration of STS markedly increased thiosulfate levels in plasma and lung with or without lipopolysaccharide challenge (fig. 5). Plasma thiosulfate concentrations reached 1.8 ± 0.3 mM and 1.8 ± 0.9 mM in mice challenged with saline or lipopolysaccharide, respectively, at 2 h after STS administration. Plasma sulfide levels were also augmented to 56 ± 16 μM and 46 ± 22 μM by STS administration in mice challenged with saline or lipopolysaccharide, respectively. Administration of STS markedly increased the levels of thiosulfate in the lung with or without lipopolysaccharide challenge. Similarly, sulfide levels doubled in the lungs of saline-challenged mice and tended to increase in lipopolysaccharide-challenged mice lungs after STS administration.

To examine the effects of STS on endothelium, we next evaluated the effects of STS on lipopolysaccharide or TNFα-induced cytokine production in HUVEC and HMVEC-L. STS (20 mM) per se had no significant cytotoxicity on HUVEC (data not shown). STS inhibited the lipopolysaccharide or TNFα-induced IL-6 production in HUVEC (fig. 6A) and HMVEC-L (fig. 6B) in a dose-dependent manner. Similarly, STS attenuated lipopolysaccharide or TNFα-induced IL-8 production in HUVEC (fig. 6C).

We assessed whether or not STS affects lipopolysaccharide or TNFα-induced production of ROS, which can lead further inflammatory response in endothelium.²³  Lipopolysaccharide (10 μg/ml) or TNFα (10 ng/ml) significantly increased the fluorescence intensity of CM-H₂DCFDA, an intracellular probe of ROS in HUVEC (fig. 7A). STS inhibited the intracellular ROS production induced by lipopolysaccharide or TNFα. The H₂O₂/peroxidase levels in HUVEC measured by fluorescent intensity of Amplex red (Invitrogen) were significantly increased with lipopolysaccharide or TNFα, which were markedly inhibited by STS (fig. 7B).

To determine whether or not STS attenuates lipopolysaccharide-induced activation of IKK/NFκB pathway in endothelial cells, we analyzed IKKα/β, IκB, and p65 in HUVEC. STS inhibited lipopolysaccharide-induced phosphorylation of IKKα/β, IκB, and p65 in a dose-dependent manner (fig. 8, A–C).

We examined the effect of STS on the activation of TAK1 and polyubiquitination of TRAF6 as upstream modulators of NFκB.²⁴  Cell lysates were immunoprecipitated with TAK1 antibody and phosphorylated TAK1 was detected by immunoblot. STS inhibited lipopolysaccharide-induced TAK1 phosphorylation (fig. 9A). Next, TRAF6 was immunoprecipitated and probed with anti-K63–specific ubiquitin antibody. STS inhibited lipopolysaccharide-induced K63-linked polyubiquitination of TRAF6 (fig. 9B). These results suggest that STS attenuates lipopolysaccharide-induced IKK/NFκB activation through inhibiting the polyubiquitination of TRAF6 and activation of TAK1.

To identify the sulfide metabolites increased by STS in the cells challenged with lipopolysaccharide, intracellular levels of free H₂S/HS⁻ and sulfane sulfur were evaluated using fluorescent probes HSip-1DA and SSP4, respectively. We performed kinetic measurement of fluorescence up to 20 h after treatment with lipopolysaccharide and STS. HSip-1 and SSP4 fluorescent intensity gradually increased over 20 h in control cells that are loaded with HSip-1DA or SSP4 but without lipopolysaccharide and STS. Lipopolysaccharide alone or lipopolysaccharide with STS at concentrations less than 5 mM did not affect fluorescence intensity of HSip-1 and SSP4 compared with control at all times. Intracellular sulfide levels were augmented by 20 mM of STS between 2 and 20 h after treatment in cells incubated with lipopolysaccharide (fig. 10A). STS at 10 mM augmented intracellular sulfide levels at 6 and 20 h after treatment in lipopolysaccharide-treated cells. Intracellular sulfane sulfur levels were augmented by 20 mM of STS between 6 and 20 h after treatment (fig. 10B). These results suggest that STS increases intracellular levels of sulfide and sulfane sulfur in lipopolysaccharide-treated cells, with the former increasing faster than the latter.

In the current study, we demonstrated that intraperitoneal administration of STS attenuated ALI in mice. STS inhibited the increase of lung permeability, influx of PMN, and expression of proinflammatory mediators in mice lung subjected to intratracheal lipopolysaccharide challenge. STS also attenuated lung tissue inflammation after polymicrobial sepsis. Administration of STS markedly increased the levels of sulfide and thiosulfate in lung and plasma of mice. We also observed that STS markedly attenuated lipopolysaccharide or TNFα-induced cytokine/ROS production in cultured endothelial cells and prevented the increase in endothelial permeability in vascular endothelial monolayer (data not shown). The beneficial effects of STS were associated with down-regulation of IKK/NFκB signaling pathways. Our results also revealed that STS markedly inhibited the lipopolysaccharide-induced activation of TAK1 and TRAF6 polyubiquitination, suggesting a novel regulatory mechanism responsible for the inhibitory effects of sulfide on NFκB signaling. Last, the current results suggest that STS exerts its beneficial effects at least in part by increasing intracellular sulfide and sulfane sulfur levels in vascular endothelium. Taken together, these observations suggest a therapeutic potential of STS against ALI.

Role of sulfide and sulfide metabolites in inflammatory organ injury remains incompletely defined. Although acute administration of high doses of H₂S donor compounds appears to be invariably toxic, lower and steady levels of H₂S may be cytoprotective against systemic inflammation.^11,22  Along these lines, we have recently reported that breathing low concentration of H₂S prevents lethal endotoxemia and lipopolysaccharide-induced lung and liver injury in mice at least in part by increasing thiosulfate.¹¹  We also observed that administration of STS dose dependently prevents death from endotoxin shock in the previous study.¹¹  These studies prompted us to further examine the lung-protective effects of STS in the current study.

Thiosulfate is a potent antioxidant and STS has been used for the treatment of cyanide poisoning and calciphylaxis with a remarkable safety track record.^13,14  In the current study, we observed that STS markedly inhibited ROS production induced by lipopolysaccharide in endothelial cells. Although we used relatively high doses of STS, previous clinical studies have shown extremely low cytotoxicity and effectiveness of STS at similar doses in patients. For example, intravenous administration of STS at dose of 4 or 12 g/m² showed no evidence of neuro- or nephrotoxicity in humans.^31,32  Furthermore, Neuwelt et al.¹⁷  used as much as 16 and 20 g/m² doses of intravenous STS against carboplatin-induced ototoxicity in human. In their study, after intravenous administration of 16 or 20 g/m² of STS, serum thiosulfate levels reached 308 mg/dl (12.3 mM) and 330.8 mg/dl (13.2 mM), respectively, immediately after bolus infusion, with no signs of toxicity.¹⁷  Given the weight and height of the patient are 50 kg and 160 cm (1.5 m² surface area), the doses of 16 to 20 g/m² STS are assumed to be 0.48 to 0.6 g/kg, which correspond to the effective intravenous STS dose in mouse subjected to CLP in the current study. Therefore, the doses of STS used in the current study and the resultant plasma concentrations of thiosulfate fall within the doses of STS and plasma levels of thiosulfate that have been observed in patients. These observations suggest clinical relevance of our findings.

It is well established that lipopolysaccharide activates toll-like receptor 4–dependent signaling cascade. Binding of lipopolysaccharide to toll-like receptor 4 in endothelial cells up-regulates production of proinflammatory cytokines, chemokines, and adhesion molecules, predominantly via the transcription factor NFκB pathway.²³  These actions lead to cell adhesion or increased vascular permeability that causes neutrophil migration and edema in the lung.²³  STS markedly inhibited lipopolysaccharide-induced phosphorylation of IκBα and nuclear translocation of NFκB p65 in mice lung. STS inhibited the rapid phosphorylation of IKK, IκBα, and p65 in HUVEC that occurred within 1 h after exposure to lipopolysaccharide. Although these observations are consistent with the hypothesis that sulfide exerts antiinflammatory effects via inhibition of NFκB,^11,22  impact of sulfide or thiosulfate on NFκB-dependent signaling is incompletely defined. For example, although the majority of studies reported that H₂S inhibits NFκB,^22,33  some studies suggested that NFκB can be activated by sulfide.³⁴  To further characterize the molecular mechanisms responsible for the inhibitory effects of STS on toll-like receptor 4-NFκB signaling, we examined the impact of STS on lipopolysaccharide-induced TRAF6 ubiquitination and TAK1 activation that are upstream of NFκB.

The protein ubiquitination is carried out through a three stepwise enzymatic reactions, involving E1 (ubiquitin-activating enzyme), E2 (ubiquitin-conjugating enzyme), and E3 (ubiquitin ligase).²⁴  TRAF6 functions as an E3, which catalyzes K63 polyubiquitination. K63-linked autoubiquitination of TRAF6 is required for the activation of TAK1 and subsequent NFκB activation.²⁴  A crucial role of ubiquitinated TRAF6 on NFkB signaling has been demonstrated by inhibiting TRAF6 polyubiquitination.^25,26  The inhibitory effects of STS on lipopolysaccharide-induced TRAF6 ubiquitination revealed in this study shed light on the novel mechanisms responsible for the antiinflammatory effects of STS and sulfide. In addition, STS inhibited TNFα-induced cytokine production in HUVEC. Of the six TRAF family members, TRAF6 is the only TRAF that mediates both of the TNF receptor and toll-like receptor signaling.³⁵  Although this unique property of TRAF6 may explain the inhibitory effects of STS on TNFα-induced cytokine production, the effects of STS on TNFα-induced signaling remain to be further elucidated in future studies.

It has been proposed that some of the effects of H₂S are mediated via the properties of sulfide metabolites containing reactive sulfane sulfur (S⁰), a labile, highly reactive sulfur atom.^8,10  Although thiosulfate is one of the sulfide metabolites that contain sulfane sulfur,⁸  thiosulfate itself appears to have limited reactivity. We therefore hypothesized that STS is converted to other sulfide metabolites that exert beneficial effects after lung injury. To determine the sulfide metabolites that are responsible for the beneficial effects of STS, we measured the levels of sulfide and reactive sulfane sulfur in HUVEC using novel fluorescent probes HSip-1²⁹  and SSP4,³⁰  respectively. Gradual increase of the HSip-1 and SSP4 fluorescence intensity in control cells may reflect endogenous production of sulfide or sulfane sulfur or leakage of the fluorescence probes to extracellular spaces where the probes can react with sulfide metabolites in culture media. We observed that STS increased intracellular H₂S levels within 2 h after the start of incubation of HUVEC with lipopolysaccharide and STS. In contrast, SSP4-reactive sulfane sulfur levels did not increase in HUVEC treated with lipopolysaccharide and STS until 6 h after the treatment. Because lipopolysaccharide triggered NFκB activation within 1 h and STS markedly inhibited NFκB activation, our data suggest that the antiinflammatory effects of STS are primarily mediated by intracellular sulfide that is converted from STS. It has been reported that thiosulfate can be converted to H₂S via 3-mercaptopyruvate sulfurtransferase which is expressed in vascular endothelium.^36,37  However, it has been suggested that H₂S converted from thiosulfate is stored as sulfane sulfur.³⁷  It is likely that levels of sulfide and sulfane sulfur are dynamically regulated in cells.

In summary, our results revealed that STS exhibits robust antiinflammatory effects on the lung. Taken together with our recent studies,^11,12  results of this study suggest a role of STS in inflammatory ALI in addition to its established role as a therapeutic agent for cyanide toxicity, calciphylaxis, and chemotoxicity. Our results also revealed for the first time that the beneficial effects of STS are associated with the inhibition of TRAF6 ubiquitination, suggesting a novel regulatory mechanism of NFkB signaling by sulfide metabolites. Considering the clinical availability and established safety track record of STS and the critical role of NFkB signaling in cellular survival, further studies examining the beneficial effects of STS in other diverse forms of organ injury are warranted.

This study was supported by a National Institutes of Health (Bethesda, Maryland) grant R01HL101930 (to Dr. Ichinose).

The authors declare no competing interests.