The mitochondria-targeted sulfide delivery molecule attenuates drugs-induced gastropathy. Involvement of heme oxygenase pathway

Abstract

Hydrogen sulfide (H₂S) signaling and H₂S-prodrugs maintain redox balance in gastrointestinal (GI) tract. Predominant effect of any H₂S-donor is mitochondrial. Non-targeted H₂S-moieties were shown to decrease the non-steroidal anti-inflammatory drugs (NSAIDs)-induced gastrotoxicity but in high doses. However, direct, controlled delivery of H₂S to gastric mucosal mitochondria as a molecular target improving NSAIDs-pharmacology remains overlooked. Thus, we treated Wistar rats, i.g. with vehicle, mitochondria-targeted H₂S-releasing AP39 (0.004-0.5 mg/kg), AP219 (0.02 mg/kg) as structural control without H₂S-releasing ability, or AP39 + SnPP (10 mg/kg) as a heme oxygenase (HMOX) inhibitor. Next, animals were administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) as NSAIDs representative or comparatively with 75% ethanol to induce translational hemorrhagic or necrotic gastric lesions, that were assessed micro-/macroscopically. Activity of mitochondrial complex IV/V, and DNA oxidation were assessed biochemically. Gastric mucosal/serum content of IL-1β, IL-10, TNF-α, TGF-β1/2, ARG1, GST-α, or phosphorylation of mTOR, NF-κB, ERK, Akt, JNK, STAT3/5 were evaluated by microbeads-fluorescent xMAP®-assay; gastric mucosal mRNA level of HMOX-1/2, COX-1/2, SOD-1/2 by real-time PCR. AP39 (but not AP219) dose-dependently (0.02 and 0.1 mg/kg) diminished NSAID- (and ethanol)-induced gastric lesions and DNA oxidation, restoring mitochondrial complexes activity, ARG1, GST-α protein levels and increasing HMOX-1 and SOD-2 expression. AP39 decreased proteins levels or phosphorylation of gastric mucosal inflammation/oxidation-sensitive markers and restored mTOR phosphorylation. Pharmacological inhibition of HMOX-1 attenuated AP39-gastroprotection. We showed that mitochondria-targeted H₂S released from very low i.g. doses of AP39 improved gastric mucosal capacity to cope with NSAIDs-induced mitochondrial dysfunction and redox imbalance, mechanistically requiring the activity of HMOX-1.

Keywords: Gastric mucosa; Heme oxygenase; Hydrogen sulfide; Mitochondria; Non-steroidal anti-inflammatory drugs.

Fig. 1

Gastric mucosal damage induced by i.g. administration of acetylsalicylic acid (ASA, 125 mg/kg) or 75% ethanol (EtOH) in rats treated or not with AP39 (0.004–0.5 mg/kg i.g.) or its structural negative control (NC-AP39, 0.02 mg/kg i.g.). Intact represents healthy gastric mucosa without any lesions (the same for both panels). Results are mean ± SD of five rats per each experimental group. A: One group of rats was treated with tin protoporphyrin IX (SnPP, 10 mg/kg i.g.) followed by i.g. administration of AP39 (0.02 mg/kg). Asterisk (*) indicates significant change as compared with intact (p < 0.05, ANOVA with Dunnett’s post hoc test). Cross (+) indicates significant change as compared with vehicle (p < 0.05, ANOVA with Dunnett’s post hoc test or unpaired t-test). Double asterisk (**) indicates significant change as compared with AP39 (0.02 mg/kg i.g.) (p < 0.05, unpaired t-test). B, C: Representative macroscopic images and histological slides of intact gastric mucosa and gastric mucosal damage induced by ASA (yellow arrows) or 75% ethanol (red arrows) in rats pretreated i.g. with vehicle or AP39 (0.02 mg/kg). C: Upper panel: hematoxylin/eosin (H&E) stained slides; lower panel: alcian blue/periodic acid-Schiff (AB/PAS) stained slides. D: Serum concentration of interleukin (IL)-1β, IL-10, tumor necrosis factor (TNF)-α in rats administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) or 75% ethanol (EtOH) and treated 30 min earlier with vehicle or AP39 (0.02 mg/kg). Results are mean ± SD of four-five values per each experimental group. Cross (+) indicates significant changes compared with the respective values in vehicle-treated rats (p < 0.05, unpaired t-test). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Concentration of 8-hydroxy-deoxyguanosine (8-OHdG) (A) and the activity of complex V (B) and complex IV (C) in gastric mucosa of rats administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) and treated 30 min earlier with vehicle, AP39 (0.02 mg/kg i.g.) or NC-AP39 (0.02 mg/kg i.g.). Intact represents healthy gastric mucosa without any lesions. Results are mean ± SD of four-five samples per each experimental group. Significant changes as compared with the respective values in intact rats are indicated by asterisk (*) (p < 0.05, ANOVA with Dunnett’s post hoc test). Cross (+) indicates significant changes as compared with vehicle (p < 0.05, ANOVA with Dunnett’s post hoc test or unpaired t-test). Significant changes in NC-AP39 group as compared with the respective values in AP39-treated rats are indicated by double asterisk (**) (p < 0.05, unpaired t-test).

Fig. 3

Gastric mucosal mRNA expression of heme oxygenase (HMOX)-1 (A) and HMOX2 (B), superoxide dismutase (SOD-1) (C), SOD-2 (D), cyclooxygenase 1 (COX-1) (E) and COX-2 (F), in rats administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) and treated 30 min earlier with vehicle or AP39 (0.02 mg/kg i.g.). Intact represents healthy gastric mucosa without any lesions. Results are mean ± SD of five rats per each experimental group. Significant changes as compared with the respective values in intact rats are indicated by asterisk (*) (p < 0.05, ANOVA with Dunnett’s post hoc test). Cross (+) indicates significant changes as compared with vehicle (p < 0.05, ANOVA with Dunnett’s post hoc test).

Fig. 4

Protein level ofarginase 1 (ARG1) (A) and glutathione transferase (GST)-α (B) in gastric mucosa of rats administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) and treated 30 min earlier with vehicle or AP39 (0.02 mg/kg). Intact represents healthy gastric mucosa without any lesions (B data for intact is derived from the same technical experimental series, published already elsewhere [30]). Results are mean ± SD of five rats per each experimental group. Significant changes compared with intact are indicated by asterisk (*) (p < 0.05, ANOVA with Dunnett’s post hoc test or unpaired t-test). Cross (+) indicates significant changes compared with vehicle (p < 0.05, ANOVA with Dunnett’s post hoc test).

Fig. 5

Gastric mucosal phospho/total protein ratio of nuclear factor-kappa B (NF-κB) (A), extracellular signal-regulated kinase (ERK) (B), c-Jun N-terminal kinases (JNK) (C), serine–threonine kinase Akt (D), signal transducer and activator of transcription (STAT) 3 (E), STAT5 (F) and mammalian target of rapamycin (mTOR) (E) in rats treated i.g. with vehicle or AP39 (0.02 mg/kg) followed by i.g. administration of acetylsalicylic acid (ASA, 125 mg/kg). Intact represents healthy gastric mucosa without any lesions. Results are mean ± SD of four-five samples per each group. Asterisk (*) indicates significant change compared to intact (p < 0.05, ANOVA with Dunnett’s post hoc test or unpaired t-test). Cross (+) indicates significant change compared with vehicle (p < 0.05, ANOVA with Dunnett’s post hoc test or unpaired t-test).

Fig. 6

Serum concentration of transforming growth factor (TGF)-β1 (A) and TGF-β2 (B) in rats administered i.g. with acetylsalicylic acid (ASA, 125 mg/kg) and treated 30 min earlier with vehicle or AP39 (0.02 mg/kg). Results are mean ± SD of three values per each experimental group. Cross (+) indicates significant changes compared with the respective values in vehicle-treated rats (p < 0.05, unpaired t-test).

Fig. 7

Overview on the possible sequence of events for NSAIDs-induced gastrotoxicity (red lines) and for the mechanistic approach of mitochondria-targeted AP39 gastroprotection vs. acetylsalicylic acid (ASA) (blue lines). Abbreviations: NSAIDs: non-steroidal anti-inflammatory drugs; H₂S: hydrogen sulfide; mTOR: mammalian target of rapamycin; ROS: reactive oxygen species; SOD: sodium dismutase; COX: cyclooxygenase; HMOX: heme oxygenase; ARG: arginase; GST: glutathione S-transferase; ERK: extracellular signal-regulated kinase; STAT: signal transducer and activator of transcription; NF-κB: nuclear factor kappa-light-chain-enhancer of activated B cells; IL: interleukin; TNF: tumor necrosis factor. Comment: H₂S-releasing AP39 increased mitochondrial resistance in gastric mucosa and therefore led to the more effective counteraction to ASA-evoked i) redox imbalance (reflected by altered DNA oxidation and altered complex IV/V activities) and ii) decreased activity of mTOR (that physiologically regulates gastric epithelial progenitor homeostasis). As a downstream effect, AP39 enhanced anti-oxidative cellular response by overexpression of cyto- and mito-protective HMOX-1 and SOD-2. This interplay further stimulated anti-oxidative defensive response supported by mTOR1 homeostasis maintenance. Therefore, ASA-triggered DNA oxidation (by ROS) was blocked, epithelial damage was decreased what prevented the activation/inhibition of subsequent cascade, including the phosphorylation of particular elements of intracellular signaling and systemic inflammatory response. Alterations of SOD-2 were evaluated on mRNA level. Due to the implementation of the selective inhibitor (in line with the gastric mucosal expression analysis), we assumed that AP39-gastroprotection vs. ASA involves the activity of HMOX-1 pathway. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)