Soluble epoxide hydrolase deficiency or inhibition enhances murine hypoxic pulmonary vasoconstriction after lipopolysaccharide challenge

Abstract

Hypoxic pulmonary vasoconstriction (HPV) is the response of the pulmonary vasculature to low levels of alveolar oxygen. HPV improves systemic arterial oxygenation by matching pulmonary perfusion to ventilation during alveolar hypoxia and is impaired in lung diseases such as the acute respiratory distress syndrome (ARDS) and in experimental models of endotoxemia. Epoxyeicosatrienoic acids (EETs) are pulmonary vasoconstrictors, which are metabolized to less vasoactive dihydroxyeicosatrienoic acids (DHETs) by soluble epoxide hydrolase (sEH). We hypothesized that pharmacological inhibition or a congenital deficiency of sEH in mice would reduce the metabolism of EETs and enhance HPV in mice after challenge with lipopolysaccharide (LPS). HPV was assessed 22 h after intravenous injection of LPS by measuring the percentage increase in the pulmonary vascular resistance of the left lung induced by left mainstem bronchial occlusion (LMBO). After LPS challenge, HPV was impaired in sEH^+/+, but not in sEH^−/− mice or in sEH^+/+ mice treated acutely with a sEH inhibitor. Deficiency or pharmacological inhibition of sEH protected mice from the LPS-induced decrease in systemic arterial oxygen concentration (Pa_O2) during LMBO. In the lungs of sEH^−/− mice, the LPS-induced increase in DHETs and cytokines was attenuated. Deficiency or pharmacological inhibition of sEH protects mice from LPS-induced impairment of HPV and improves the Pa_O2 after LMBO. After LPS challenge, lung EET degradation and cytokine expression were reduced in sEH^−/− mice. Inhibition of sEH might prove to be an effective treatment for ventilation-perfusion mismatch in lung diseases such as ARDS.

systemic blood vessels dilate in response to a decrease in arterial oxygen concentration (10). In the pulmonary circulation, blood vessels constrict when exposed to low oxygen levels, a vasoconstrictor response described as “hypoxic pulmonary vasoconstriction” (HPV). HPV, as the physiological response to alveolar hypoxia, directs pulmonary blood flow away from hypoxic lung regions and toward well-ventilated regions and therefore matches perfusion with alveolar ventilation. The diversion of blood to regions of the lung with adequate alveolar ventilation maintains systemic arterial oxygenation, highlighting the physiological importance of HPV (25).

Phospholipase A₂ (PLA₂) mediates the release of arachidonic acid (AA) from membrane phospholipids (Fig. 1) (19). Cyclooxygenase, lipoxygenase, and the cytochrome P-450 (CYP) epoxygenases are enzymes that further metabolize AA (3). CYP epoxygenases metabolize AA to 14,15-, 11,12-, 8,9-, and 5,6- epoxyeicosatrienoic acids (EETs) (23). These EETs are pulmonary vasoconstrictors and are converted to less vasoactive dihydroxyeicosatrienoic acids (DHETs) by the enzyme soluble epoxide hydrolase (sEH) (34). Mice with a congenital deficiency of cytosolic PLA₂ and CYP epoxygenase CYP2J, both enzymes upstream of EET metabolism, lack HPV (11, 36). In mice with a deficiency of sEH, the enzyme that metabolizes EETs, HPV has been reported to be enhanced (14).

Fig. 1.

Overview of arachidonic acid (AA) metabolism. The enzyme cytosolic phospholipase A₂ (cPLA₂) releases AA from phospholipids. AA is further metabolized by cytochrome P-450 epoxygenases (CYP-epoxygenases) to epoxyeicosatrienoic acids (EETs). EETs are metabolized to their corresponding diols, the dihydroxyeicosatrienoic acids (DHETs), by the enzyme soluble epoxide hydrolase (sEH). Pharmacological inhibition of CYP-epoxygenases [e.g., with N-methylsulfonyl-6-(2-propargy-loxyphenyl) hexanamide (MS-PPOH)] reduces the level of EETs. Inhibition of sEH (e.g., with IK-950) increases EET levels. COX, cyclooxygenase. LOX, lipoxygenase.

HPV is impaired in human acute respiratory distress syndrome (ARDS) (27) and in endotoxemic mice (12, 28, 29). The aim of the present study was to elucidate the effects of deficiency and pharmacological inhibition of sEH on LPS-induced impairment of HPV. We hypothesized that either a congenital deficiency or pharmacological inhibition of sEH would decrease the conversion of EETs to their biologically less active DHETs (8–9, 14, 18, 24) and would therefore enhance HPV in endotoxemic mice (26). We report that sEH deficiency or inhibition enhances murine HPV and protects mice from the decrease in systemic arterial oxygen concentration (Pa_O2) after challenge with LPS.

MATERIALS AND METHODS

Animals.

All of the animal experiments described in this study conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health and were approved by the Subcommittee on Research Animal Care of the Massachusetts General Hospital. We studied C57Bl/6 male wild-type (sEH^+/+) mice, which were obtained from the Jackson Laboratory (Bar Harbor, ME), and sEH-deficient (sEH^−/−) mice. Homozygous C57Bl/6 sEH^−/− mice were generously provided by Dr. Bruce D. Hammock (Department of Entomology at the University of California, Davis, CA) and were bred in the hospital's animal resource facility (Center for Comparative Medicine at Massachusetts General Hospital). Male mice were studied between 2 and 5 mo of age and weighed 20 to 30 g. Mice in each experimental group were matched for body weight and age.

Surgical preparation and invasive hemodynamic measurements in anesthetized and mechanically ventilated mice.

To assess HPV, left pulmonary vascular resistance index (LPVRI) was estimated before and during alveolar hypoxia induced by left mainstem bronchial occlusion (LMBO) (2, 11, 36). Mice were anesthetized with an intraperitoneal (ip) injection of ketamine (120 mg/kg) and fentanyl (0.09 mg/kg) and placed on a heating pad to maintain core temperature at 37°C. Following a tracheostomy, rocuronium (1 mg/kg) was injected ip to induce muscle relaxation and mice were subjected to a median thoracotomy. Volume-controlled ventilation was provided at a respiratory rate of 110 breaths/min, a tidal volume of 8 ml/kg, a positive end-expiratory pressure of 1–2 cmH₂O, and an inspired fraction of oxygen (Fi_O2) of 1.0 (MiniVent 845; Harvard Apparatus, Holliston, MA). For hemodynamic measurements, fluid-filled polyethylene 10 catheters were inserted into the right carotid artery and the main pulmonary artery. In some mice, a 1.0-mm ultrasonic flow probe (1RB, Transonic Systems, Ithaca, NY) was placed around the lower thoracic aorta and connected to a flowmeter (T106, Transonic Systems) to measure lower thoracic aortic flow (Q_LTAF). A 0.5-mm ultrasonic flow-probe (0.5VB, Transonic Systems) was placed around the left pulmonary artery and connected to a flowmeter (T402, Transonic Systems) to obtain left pulmonary arterial blood flow (Q_LPA). Heart rate (HR), mean arterial blood pressure (MAP), pulmonary arterial pressure (PAP), Q_LPA, and Q_LTAF were continuously measured and recorded (Chart 5 software, ADInstruments, Colorado Springs, CO) before and during LMBO. To estimate total systemic vascular resistance (TSVRI) or left pulmonary vascular resistance index (LPVRI), the inferior vena cava was partially occluded to transiently reduce Q_LTAF or Q_LPA to ∼50%, as previously described (12). TSVRI or LPVRI was calculated from the slope of the MAP/Q_LTAF or PAP/Q_LPA relationship. In each mouse, the increase in LPVRI induced by LMBO was obtained by calculating the percentage change in the mean value of the PAP/Q_LPA slopes. HPV is expressed as the percentage increase in LPVRI during LMBO (21). Arterial blood was sampled from the right carotid artery after obtaining hemodynamic measurements during LMBO, and blood-gas analysis was performed using an ABL800 FLEX analyzer (Radiometer America, Westlake, OH). The following exclusion criteria were used: preparation time of more than 60 min, heart rate less than 400 beats/min, and inadvertent displacement of the catheters or the flow probes.

Treatment of mice with LPS, IK-950, and MS-PPOH.

The epoxygenase inhibitor N-methylsulfonyl-6-(2-propargy-loxyphenyl) hexanamide (MS-PPOH) was purchased from Cayman Chemical (Ann Arbor, MI). MS-PPOH (30 mg/kg), dissolved in 45% 2-hydroxypropyl-cyclodextrin in saline, or the solvent alone, was injected via tail vein 90 min before HPV measurements. The sEH inhibitor IK-950 was kindly provided by Jeffrey D. Winkler (University of Pennsylvania, Philadelphia, PA) and was synthesized as previously described (16–17). The sEH inhibitor IK-950 (10 mg/kg), dissolved in saline, was injected via the jugular vein 30 min before HPV measurements. The doses of MS-PPOH and IK-950 and the timing of infusion relative to HPV measurements were based on previous studies (30, 36).

LPS from Escherichia coli 055:B5 was purchased from List Biological Laboratories (Campbell, CA). Awake mice received an injection of either LPS (20 mg/kg) dissolved in saline or saline alone (0.01 ml/g body wt) via a tail vein injection 22 h before HPV measurement. The dose and timing of LPS injection used in these studies was based on a previous study (28).

Measurement of DHET levels and cytokine mRNA levels in lung tissue.

Twenty-two hours after treatment with either LPS or saline, lungs of sEH^−/− and sEH^+/+ mice were perfused with saline (4°C) via the pulmonary artery. The lungs were removed and left lungs were used for measurement of DHET levels; right lungs were used to prepare mRNA for measuring cytokine mRNA levels.

For determination of DHET levels, frozen lung tissue was homogenized by hand using a glass tissue homogenizer in 400 μl 66% methanol. Internal standards were added, and samples were stored at −80°C prior to use. Eicosanoids were extracted using solid-phase Strata-X Polymeric Reversed Phase 60-mg cartridges (Phenomenex, Torrance, CA). Each sample was centrifuged for 15 min at 4°C. The supernatant was then collected, diluted with 5 volumes of water, and acidified to pH 5.0 with HCl. Cartridges were primed with 6 ml of methanol followed by 6 ml of water. Samples were loaded, washed with 2 ml 20% methanol, and eluted with 1.5 ml of ethyl acetate. The collected ethyl acetate fraction was dried under nitrogen and resuspended in 50 μl of methanol and stored at −80°C until analysis. The total protein level in lung pellets was quantified by Bradford assay. Identification and quantification of eicosanoids were performed using a Shimadzu Triple Quadrupole Mass Spectrometer LCMS-8050 equipped with a Nexera UHPLC using multiple reaction monitoring mode. MS conditions were 1) Ionization mode: negative heated electrospray; 2) Applied voltage: −4.5∼−3 kV; 3) Nebulizer gas: 3.0 l/min N₂; 4) Drying gas: 5.0 l/min N₂; 5) Heating gas: 12.0 l/min air; 6) Interface temperature: 400°C; 7) Desolvation line temperature: 100°C; 8) Heat block temperature: 500°C; 9) Internal standards: d6 20-HETE, d8 5-HETE, d4 PGE2, d11 14(15) EET, and d11 11,12 DHET. UHPLC conditions were 1) Analytical column: Zorbax Eclipse Plus C18 RRHD (50 mm length × 2.1 mm ID, 1.8 μm); 2) Mobile phase A: 95% water 5% acetonitrile 0.05% acetic acid; 3) Mobile phase B (B.): acetonitrile 0.05%; 4) Time program 40% B. 0 min)→75% B. 3 min)→85% B. 7.5 min); 5) Flow rate: 0.4 ml/min; 6) Injection volume: 5 μl; 7) Column oven temperature: 40°C. Synthetic standards were used to obtain standard curves (0.5–500 pg) for each compound. The standard curves were used to calculate the final concentrations of the eicosanoids. All solvents were HPLC grade or higher.

To measure levels of IL-6, TNF-α, and IL-1β mRNAs, total RNA from mouse lungs was extracted by the phenol/guanidinium method, as previously described (5). Reverse transcription was performed using MultiScribe MuLV Reverse Transcriptase (Thermo Fisher Scientific, Waltham, MA). A Mastercycler ep realplex (Eppendorf, Hamburg, Germany) was used for real-time amplification and quantification of mRNA transcripts. All mRNA levels were determined by the relative CT method normalized to 18S ribosomal RNA. TaqMan gene expression assays were used to quantify mRNA levels encoding IL-6, TNF-α, and IL-1β, as well as the level of 18S RNA.

Statistical analysis.

Statistical analyses were performed using Prism 6 software (GraphPad Software, La Jolla, CA). Variables were tested for normality by the Shapiro-Wilk test. For hemodynamic experiments, a two-way ANOVA with repeated measures was used to compare differences between groups. When the interaction P value between time and condition was significant, a one-way ANOVA with post hoc Bonferroni-adjusted comparison testing for normally distributed data was performed. A Kruskal-Wallis test with post hoc Dunn's comparison testing was used for data that were not normally distributed. Measurements within the same experimental group were compared by either the paired t-test or the Wilcoxon signed-rank test, as appropriate. Measurements between two experimental groups were compared either with the unpaired t-test or the Mann-Whitney test, as appropriate. Comparisons between groups for the DHET lung levels were performed on log-transformed data. P values <0.05 were considered statistically significant. All data are expressed as means ± SE.

RESULTS

Effects of sEH deficiency and inhibition on HPV and hemodynamic parameters.

To assess the role of sEH in the regulation of pulmonary vascular tone, we measured LPVRI at baseline and during LMBO in sEH^−/− and sEH^+/+ mice. At baseline breathing pure oxygen, LPVRI was similar in sEH^−/− and sEH^+/+ mice (Table 1). LMBO induced a similar increase in LPVRI after LMBO in sEH^−/− and sEH^+/+ mice (%increase in LPVRI in sEH^−/− vs. sEH^+/+ mice: 101 ± 21 vs. 81 ± 11%, P = not significant; Fig. 2A). There were no significant differences in PAP or Q_LPA between sEH^−/− and sEH^+/+ mice either before or after LMBO (Table 1). To examine the effect of sEH deficiency on systemic vascular tone, we measured systemic hemodynamic parameters [blood flow through the descending thoracic aorta (Q_LTAF) and MAP] in sEH^−/− and sEH^+/+ mice. Q_LTAF did not differ between sEH^−/− and sEH^+/+ mice either at baseline (205 ± 30 vs. 188 ± 11 μl·min⁻¹·g⁻¹, P = not significant, Table 1) or during LMBO (215 ± 32 vs. 200 ± 16 μl·min⁻¹·g⁻¹, P = not significant, Table 1). At baseline and during LMBO, MAP was lower in sEH^−/− mice compared with sEH^+/+ mice (Table 1). Because TSVRI was calculated from the slope of the MAP/Q_LTAF relationship, TSVRI was lower in sEH^−/− mice compared with sEH^+/+ mice (Table 1). These results show that the absence of sEH results in decreased systemic vascular resistance in anesthetized mice but does not alter either the baseline pulmonary vascular resistance (PVR) or the increase in PVR after LMBO.

Table 1.

Hemodynamic measurements at baseline and 5 min after LMBO in sEH^−/− or sEH^+/+ mice with or without treatment with LPS or MS-PPOH

			HR, bpm		MAP, mmHg		PAP, mmHg		QLPA, μl·min-1·g-1		LPVRI, mmHg·min·g·ml-1		QLTAF, μl·min-1·g-1		TSVRI, mmHg·min·g·ml-1
Genotype	Treatment	n	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO
sEH+/+	Saline	11	621 ± 12	609 ± 13	80 ± 4	82 ± 3	17 ± 1	19 ± 0a	132 ± 9	89 ± 8a	59 ± 5	105 ± 9a	188 ± 11	200 ± 16	150 ± 17	128 ± 12
sEH−/−	Saline	10	517 ± 22b	528 ± 26b	52 ± 6b	52 ± 5b	16 ± 1	17 ± 1a	106 ± 9	72 ± 6a	58 ± 5	109 ± 7a	205 ± 30	215 ± 32	88 ± 16b	81 ± 12b
sEH+/+	LPS (20 mg/kg)	11	527 ± 15b	539 ± 16b	84 ± 3	83 ± 5	17 ± 1	21 ± 1a	166 ± 11	141 ± 12abd	58 ± 3	83 ± 6a	260 ± 15	281 ± 15	86 ± 8b	75 ± 7b
sEH−/−	LPS (20 mg/kg)	9	498 ± 13b	513 ± 13b	76 ± 4	69 ± 5	16 ± 0	18 ± 1a	128 ± 10	83 ± 10ace	70 ± 5	123 ± 12ace	212 ± 21	210 ± 27	122 ± 18	115 ± 21
sEH+/+	β-Cyclodextrin in saline	7	641 ± 12	627 ± 15a	87 ± 7	78 ± 7	19 ± 2	20 ± 1a	151 ± 19	96 ± 10a	57 ± 5	101 ± 12a	172 ± 13	167 ± 19	174 ± 25	141 ± 18
sEH+/+	MS-PPOH (30 mg/kg)	7	630 ± 14	617 ± 19	89 ± 4	81 ± 6a	19 ± 2	21 ± 1	158 ± 11	121 ± 10a	55 ± 5	79 ± 6a	180 ± 12	192 ± 11	175 ± 31	138 ± 31
sEH−/−	β-Cyclodextrin in saline	5	573 ± 28b	555 ± 26	79 ± 6	74 ± 7	20 ± 3	20 ± 2	136 ± 21	91 ± 22a	57 ± 14	107 ± 31a	251 ± 25	274 ± 20	98 ± 12	83 ± 15
sEH−/−	MS-PPOH (30 mg/kg)	5	621 ± 16	582 ± 34	99 ± 13	93 ± 9	19 ± 1	20 ± 1a	169 ± 14	130 ± 16a	55 ± 3	79 ± 7a	233 ± 20	251 ± 23	116 ± 12	99 ± 4

Data represent means ± SE.

^aP < 0.05 vs. baseline value of the same parameter in the same treatment group.

^bP < 0.05 vs. saline- or β-cyclodextrin-treated sEH^+/+ mice at the same time point.

^cP < 0.05 vs. sEH^+/+ mice in the same treatment group.

^dP < 0.05 vs. saline-treated sEH^+/+ mice for the delta between LMBO and baseline of measured parameter.

^eP < 0.05 vs. sEH^+/+ mice in the same treatment group for the delta between LMBO and baseline of the measured parameter.

Fig. 2.

Hypoxic pulmonary vasoconstriction (HPV) in mice lacking active sEH. Left pulmonary vascular resistance index (LPVRI) was calculated from the slope of the PAP/Q_LPA relationship at baseline and during left mainstem bronchial occlusion (LMBO). HPV is expressed as the percentage increase in LPVRI induced by LMBO. A: the percentage increase in LPVRI after LMBO was similar in sEH^−/− (n = 10) and sEH^+/+ mice (n = 11, P = not significant). B: treatment with the sEH inhibitor IK-950 (10 mg/kg) 30 min before HPV measurements enhanced HPV in sEH^+/+ mice compared with control sEH^+/+ mice (n = 7, *P < 0.05). C: mice were treated with the CYP-epoxygenase inhibitor MS-PPOH (30 mg/kg) 90 min before the HPV measurements. MS-PPOH impaired HPV in sEH^−/− (n = 5) and sEH^+/+ (n = 7) mice compared with control mice (#P < 0.05 vs. control sEH^−/− mice; *P < 0.05 vs. control sEH^+/+ mice). Data represent means ± SE and were compared by unpaired t-test (A and B) and one-way ANOVA followed by post hoc Bonferroni-adjusted comparison testing (C).

To assess the effect of pharmacological inhibition of sEH on HPV, we measured changes in LPVRI in response to LMBO in sEH^+/+ mice that were treated with or without the sEH inhibitor IK-950. At baseline, LPVRI was similar in IK-950- or saline-treated (control) sEH^+/+ mice (Table 2). LMBO reduced Q_LPA to a greater extent in IK-950-treated compared with control sEH^+/+ mice (Q_LPA during LMBO: 39 ± 3 vs. 54 ± 2 μl·min⁻¹·g⁻¹, P < 0.05; Table 2). At baseline and during LMBO, PAP did not differ between IK-950-treated and control sEH^+/+ mice (Table 2). Due to the decrease in Q_LPA with no difference in PAP, LMBO increased LPVRI more in IK-950-treated sEH^+/+ mice than in control sEH^+/+ mice (%increase in LPVRI: 169 ± 27 vs. 81 ± 9%, P < 0.05; Fig. 2B). MAP did not differ between sEH^+/+ mice treated with IK-950 or control sEH^+/+ mice either before or after LMBO (Table 2). Taken together, the results show that acute pharmacological inhibition of sEH with IK-950 augments HPV.

Table 2.

Hemodynamic measurements at baseline and 5 min after LMBO in sEH^+/+ mice with or without treatment with IK-950 or LPS

		HR, bpm		MAP, mmHg		PAP, mmHg		QLPA, μl·min-1·g-1		LPVRI, mmHg·min·g·ml-1
Treatment	n	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO	Baseline	LMBO
Saline + Saline	7	527 ± 12	543 ± 18	100 ± 5	98 ± 6	20 ± 1	22 ± 1*	100 ± 3	54 ± 2*	71 ± 4	128 ± 6*
Saline + IK-950 (10 mg/kg)	7	519 ± 12	527 ± 21	91 ± 5	91 ± 6	22 ± 0.3	24 ± 1*	97 ± 2	39 ± 3*§	64 ± 4	172 ± 18*§
LPS (20 mg/kg) + Saline	10	516 ± 10	526 ± 13	84 ± 5	83 ± 5	19 ± 1	18 ± 1	105 ± 6	82 ± 5*†§	79 ± 9	102 ± 8*
LPS (20 mg/kg) + IK-950 (10 mg/kg)	10	518 ± 12	535 ± 14	89 ± 4	87 ± 3	17 ± 1†	20 ± 1*‡	109 ± 6	69 ± 6*‡	70 ± 8	137 ± 15*‡

Data represent means ± SE.

^*P < 0.05 vs. baseline value of the same parameter in the same treatment group.

^†P < 0.05 vs. control sEH^+/+ mice.

^‡P < 0.05 vs. LPS- and non-IK-950-treated sEH^+/+ mice for the delta between LMBO and baseline of the measured parameter.

^§P < 0.05 vs. control sEH^+/+ mice for the delta between LMBO and baseline of the measured parameter.

Effects of pharmacological inhibition of cytochrome P-450 epoxygenase activity on HPV in sEH^−/− and sEH^+/+ mice.

CYP epoxygenases are enzymes that increase the level of EETs by metabolizing AA (Fig. 1) (4). Pharmacological inhibition of CYP epoxygenases with MS-PPOH impairs HPV in sEH^+/+ mice by reducing the level of pulmonary vasoconstricting EETs (30). To test whether HPV in sEH^−/− mice is also modulated by the production of pulmonary vasoconstricting EETs generated by CYP epoxygenases, sEH^−/− and sEH^+/+ mice were treated with MS-PPOH or the solvent alone 90 min before HPV measurements. Inhibition of CYP epoxygenase with MS-PPOH impaired HPV in sEH^−/− mice (%increase in LPVRI after LMBO in MS-PPOH- vs. control sEH^−/− mice: 44 ± 9 vs. 83 ± 9%, P < 0.05; Fig. 2C) and sEH^+/+ mice (%increase in LPVRI after LMBO in MS-PPOH- vs. control sEH^+/+ mice: 46 ± 6 vs. 78 ± 12%, P < 0.05; Fig. 2C). These results indicate that HPV after LMBO is dependent on the enzymatic activity of the EET-generating CYP epoxygenases in both sEH^−/− and sEH^+/+ mice.

Effects of congenital deficiency or acute inhibition of sEH on HPV after LPS challenge.

HPV was impaired in sEH^+/+ mice 22 h after challenge with intravenous (iv) LPS (Fig. 3, A and B). To assess the effect of congenital sEH deficiency on LPS-induced impairment of HPV, we measured changes in LPVRI in response to LMBO 22 h after challenge with iv LPS. After LPS-exposure but before LMBO, LPVRI was similar in sEH^−/− and sEH^+/+ mice (Table 1). Although Q_LPA was reduced during LMBO compared with baseline in LPS-challenged sEH^−/− and sEH^+/+ mice, after LPS exposure there was a greater reduction in Q_LPA during LMBO in sEH^−/− mice compared with sEH^+/+ mice (83 ± 10 vs. 141 ± 12 μl·min⁻¹·g⁻¹, P < 0.05; Table 1). Twenty-two hours after challenge with LPS, the LMBO-induced increase in LPVRI was greater in sEH^−/− mice compared with sEH^+/+ mice (%increase in LPVRI: 77 ± 12 vs. 43 ± 3%, P < 0.05; Fig. 3A). At 22 h after LPS, there was no significant difference in either baseline PAP or in the increase in PAP following LMBO in either genotype (Table 1). Compared with control sEH^+/+ mice, exposure to LPS reduced TSVRI in sEH^+/+ mice. In contrast, when sEH^−/− mice were challenged with LPS, TSVRI did not change compared with control sEH^+/+ mice (Table 1). These results show that, 22 h after LPS challenge, a congenital absence of sEH preserves HPV and protects mice from the LPS-induced decrease in systemic vascular resistance.

Fig. 3.

Measurements of hypoxic pulmonary vasoconstriction (HPV) in mice challenged with lipopolysaccharide (LPS). Left pulmonary vascular resistance index (LPVRI) was calculated from the slope of the PAP/Q_LPA relationship at baseline and during left mainstem bronchial occlusion (LMBO). HPV is expressed as the percentage increase in LPVRI induced by LMBO. A and B: 22 h after challenge with LPS, HPV was impaired in sEH^+/+ mice (n = 7 and 11, *P < 0.05 vs. control sEH^+/+ mice). A: after challenge with LPS, sEH^−/− mice were protected from the loss of HPV (n = 9, #P < 0.05 vs. LPS-challenged sEH^+/+ mice). B: in LPS-challenged sEH^+/+ mice, acute treatment with IK-950 (10 mg/kg) restored HPV (n = 10, #P < 0.05 vs. LPS-challenged sEH^+/+ mice). Data represent means ± SE and were compared by one-way ANOVA followed by post hoc Bonferroni-adjusted comparison testing (A and B).

To determine whether pharmacological inhibition of sEH would enhance HPV after exposure to LPS, we measured changes in LPVRI in response to LMBO in sEH^+/+ mice that were exposed to LPS and received either IK-950 or the solvent alone. After LPS exposure but before LMBO, LPVRI was similar in IK-950- and non-IK-950-treated sEH^+/+ mice (Table 2). Treatment with IK-950 restored HPV in LPS-challenged sEH^+/+ mice (%increase in LPVRI: 101 ± 13 vs. 37 ± 11%, P < 0.05; Fig. 3B). In LPS-challenged sEH^+/+ mice, IK-950 did not alter HR and MAP (Table 2). These results demonstrate that acute pharmacological inhibition of sEH with IK-950 restores HPV in LPS-challenged sEH^+/+ mice.

Effects of congenital sEH deficiency or acute inhibition of sEH on systemic arterial oxygenation in LPS-challenged mice after LMBO.

Twenty-two hours after iv challenge with LPS, HPV was preserved in sEH^−/− mice and in IK-950-treated sEH^+/+ mice. To determine the impact of HPV on systemic arterial oxygenation, we measured Pa_O2 (at Fi_O2 1.0) after LPS challenge and during LMBO in sEH^−/− mice and in sEH^+/+ mice treated with IK-950. The Pa_O2 was higher in LPS-challenged sEH^−/− mice compared with LPS-challenged sEH^+/+ mice (Pa_O2: 289 ± 51 vs. 129 ± 7 mmHg, P < 0.05; Fig. 4A and Table 3). In sEH^+/+ mice exposed to LPS for 22 h and treated with IK-950 30 min before LMBO, the Pa_O2 also was significantly higher than in LPS-challenged sEH^+/+ mice that did not receive the sEH inhibitor (Pa_O2: 225 ± 33 vs. 121 ± 9 mmHg, P < 0.05; Fig. 4B). Therefore, after LPS challenge, congenital deficiency of sEH or inhibition of sEH by IK-950 in sEH^+/+ mice improves systemic arterial oxygenation during LMBO.

Fig. 4.

Systemic arterial oxygen concentration (Pa_O2) during left mainstem bronchial occlusion (LMBO) after lipopolysaccharide (LPS) challenge in mice lacking active sEH. A: after treatment with LPS, sEH^−/− mice had a greater Pa_O2 during LMBO compared with sEH^+/+ mice (n = 8–9, *P < 0.05 vs. control sEH^+/+ mice; #P < 0.05 vs. LPS-challenged sEH^+/+ mice). B: after challenge with LPS (20 mg/kg), systemic oxygenation during LMBO was augmented when sEH^+/+ mice were treated with IK-950 (10 mg/kg) (n = 7–10, #P < 0.05 vs. LPS-challenged and non-IK-950-treated sEH^+/+ mice). Data represent means ± SE and were compared by Kruskal-Wallis test followed by post hoc Dunn's comparison testing (A) or one-way ANOVA followed by post hoc Bonferroni-adjusted comparison testing (B).

Table 3.

Arterial blood gas analyses during left mainstem bronchial occlusion

Genotype	Treatment	n	pHa	PaO2, mmHg	PaCO2, mmHg	Hemoglobin, mg/dl	Base Excess, mmol/l
sEH+/+	Saline	9	7.30 ± 0.01	235 ± 34	37 ± 2	14.8 ± 0.3	−9.3 ± 1.0
sEH−/−	Saline	8	7.29 ± 0.01	263 ± 45	33 ± 3	14.4 ± 0.4	−8.8 ± 0.8
sEH+/+	LPS (20 mg/kg)	9	7.16 ± 0.03*	129 ± 7*	44 ± 4	12.2 ± 0.3*	−13.0 ± 1.3
sEH−/−	LPS (20 mg/kg)	8	7.13 ± 0.03*	289 ± 51†	36 ± 1	12.1 ± 0.3*	−16.0 ± 0.9*
sEH+/+	β-Cyclodextrin in saline	7	7.29 ± 0.02	196 ± 30	34 ± 2	14.8 ± 0.2	−12.3 ± 0.4
sEH+/+	MS-PPOH (30 mg/kg)	7	7.31 ± 0.02	181 ± 25	35 ± 2	15.4 ± 0.3	−11.7 ± 0.4
sEH−/−	β-Cyclodextrin in saline	5	7.36 ± 0.01	240 ± 40	32 ± 2	14.6 ± 0.2	−9.7 ± 0.6*
sEH−/−	MS-PPOH (30 mg/kg)	5	7.35 ± 0.02	169 ± 26	33 ± 2	14.3 ± 0.5	−9.1 ± 0.8†

Data represent means ± SE. Pa_O2, systemic arterial oxygen concentration during LMBO (Fi_O2 = 1.0).

^*P < 0.05 vs. saline- or β-cyclodextrin-treated sEH^+/+ mice.

^†P < 0.05 vs. sEH^+/+ mice in the same treatment group.

Effects of LPS challenge on DHET levels in lung tissue of sEH^−/− and sEH^+/+ mice.

The enzyme sEH mediates conversion of pulmonary vasoconstricting EETs to less vasoactive DHETs (7). Because EETs are unstable and have a very short half-life (4, 31), it is technically difficult to measure EET levels in blood and tissue. In this study, we therefore measured the levels of the more stable DHETs as an indirect measure of EET levels. At 22 h after iv LPS challenge, an increase in conversion of pulmonary vasoconstricting EETs to the less vasoactive DHETs in sEH^+/+ mice might contribute to the observed impairment of HPV after LMBO. Conversely, a decrease in conversion of EETs to DHETs in sEH^−/− mice might contribute to the preservation of HPV in sEH^−/− mice after LPS challenge. We therefore measured the levels of stable DHETs in lung tissue obtained from sEH^−/− and sEH^+/+ mice, including LPS-challenged and control mice. Levels of 14,15-, 11,12-, 8,9-, and 5,6-DHET were higher in LPS-challenged sEH^+/+ mice than in control sEH^+/+ mice (Fig. 5, A–D). In contrast, compared with control sEH^+/+ mice, exposure to LPS in sEH^−/− mice did not change levels of lung 14,15- and 11,12-DHET (Fig. 5, A and B), and only increased lung levels of 8,9- and 5,6-DHET (Fig. 5, C and D). Taken together, the LPS-induced impairment of HPV in sEH^+/+ mice is accompanied by an increase in lung DHET levels, suggesting an increase in EET degradation in these mice. The LPS-induced increase in 14,15- and 11,12-DHET levels was attenuated in sEH^−/− mice, suggesting a lower rate of 14,15- and 11,12-EET degradation in sEH^−/− mice.

Fig. 5.

Levels of DHETs in murine lung tissue. Levels of lung 14,15-DHET (A), 11,12-DHET (B), 8,9-DHET (C), 5,6-DHET (D), and total (14,15-, 11,12-, 8,9-, 5,6-) DHET (E) increased markedly 22 h after sEH^+/+ mice were challenged with lipopolysaccharide (LPS, 20 mg/kg) (n = 6–7, *P < 0.05 vs. control sEH^+/+ mice). Compared with control sEH^+/+ mice, levels of lung 14,15-DHET (A) and 11,12-DHET (B) did not change in sEH^−/− mice 22 h after LPS challenge. Levels of 8,9-DHET (C) and 5,6-DHET (D) increased in both sEH^+/+ and sEH^−/− mice 22 h after challenge with LPS (n = 6–7, *P < 0.05 and #P < 0.05 vs. control sEH^+/+ mice). E: 22 h after challenge with LPS, total DHET levels (14,15-, 11,12-, 8,9-, and 5,6-DHET) were higher in sEH^+/+ mice compared with control sEH^+/+ mice (n = 6–7, *P < 0.05 vs. control sEH^+/+ mice) but were markedly decreased in sEH^−/− mice compared with LPS-challenged sEH^+/+ mice (n = 6–7, §P < 0.05 vs. LPS-challenged sEH^+/+ mice). Data represent means ± SE and were compared by one-way ANOVA (A–E) followed by post hoc Bonferroni-adjusted comparison testing on log-transformed data.

Level of cytokine mRNA in lung tissue of sEH^−/− and sEH^+/+ mice after exposure to LPS.

To determine the possible contribution of sEH to the inflammatory response associated with LPS exposure, we measured cytokine mRNA levels in lung tissue obtained from LPS-challenged or control sEH^−/− and sEH^+/+ mice. In the lung tissue of sEH^+/+ mice, LPS-exposed mice had higher IL-6, TNF-α, and IL-1β mRNA levels than control mice (Fig. 6, A–C). Compared with LPS-challenged sEH^+/+ mice, IL-6, TNF-α, and IL-1β mRNA levels were lower in lung tissue of LPS-challenged sEH^−/− mice (Fig. 6, A–C). These results show that the lack of sEH blunts the LPS-induced increase in pulmonary cytokine mRNA levels.

Fig. 6.

Relative levels of IL-6, TNF-α, and IL-1β mRNA in lung tissue from sEH^−/− and sEH^+/+ mice 22 h after challenge with lipopolysaccharide (LPS). Levels of lung IL-6 mRNA (A), TNF-α mRNA (B), and IL-1β mRNA (C) increased markedly in sEH^+/+ mice after challenge with LPS (n = 9–11, *P < 0.05 vs. control sEH^+/+ mice). The lung tissue of LPS-challenged sEH^−/− mice showed markedly decreased levels of lung IL-6 mRNA (A), TNF-α mRNA (B), and IL-1β mRNA (C) 22 h after LPS challenge compared with LPS-challenged sEH^+/+ mice (n = 6–11, #P < 0.05 vs. LPS-challenged sEH^+/+ mice). Data represent means ± SE and were compared by one-way ANOVA (A–C) followed by post hoc Bonferroni-adjusted comparison testing.

DISCUSSION

In the present study, we investigated the role of sEH on HPV in endotoxemic mice. In mice that were exposed to LPS, congenital absence or acute pharmacological inhibition of sEH augmented HPV and increased systemic arterial oxygenation. The increase in lung DHET levels after LPS challenge seen in sEH^+/+ mice suggests that LPS promotes EET degradation, which may contribute to LPS-induced impairment of HPV. Deficiency of sEH prevented the LPS-induced increase in lung DHET levels, suggesting a lower degradation rate of EETs in the lungs of sEH^−/− mice. After challenge with LPS, sEH^−/− mice had reduced pulmonary IL-6, TNF-α, and IL-1β mRNA levels compared with LPS-challenged sEH^+/+ mice. Interestingly, in this study we observed that in mice that were not challenged with LPS, HPV was enhanced in IK-950-treated sEH^+/+ mice but not in sEH^−/− mice. In both sEH^−/− and sEH^+/+ mice, HPV is dependent on the enzymatic activity of EET-generating CYP epoxygenases.

Previous investigators demonstrated that deficiency of sEH increases HPV in mice (14–15). In the present study, we found that in non-LPS-challenged sEH^−/− mice, pulmonary vascular tone (Table 1) and HPV (Fig. 2A) were unaffected. Previous investigators also found that pharmacological inhibition of sEH enhances HPV in mice (14, 30). Although deficiency of sEH did not affect HPV in the present study, acute inhibition of sEH using IK-950 increased HPV in non-LPS-challenged sEH^+/+ mice (Fig. 2B). Acute inhibition of sEH may increase the level of EETs and thereby induce pulmonary vasoconstriction and enhance HPV (34). A possible reason for the difference between our results and previous studies might be that the chronic absence of sEH in non-LPS-challenged mice influences the activity of other, potentially compensatory enzymes involved in the AA pathway, such as cytosolic PLA₂ or CYP epoxygenases. Mice with cytosolic PLA₂ or CYP2J epoxygenase deficiency have impaired HPV, suggesting that these enzymes are critically involved in the maintenance of HPV (11, 36). The activity of cytosolic PLA₂ or CYP epoxygenases may be altered in sEH-deficient mice due to feedback mechanisms. Because the previous studies, which demonstrated enhanced HPV in sEH-deficient mice, employed isolated, ex vivo lung perfusion to measure HPV, it is possible that in vivo feedback mechanisms that compensate for chronic sEH deficiency were not detected. In contrast, in the present study, we measured HPV in anesthetized and mechanically ventilated mice during alveolar hypoxia in the left lung, using dynamic measurements of pulmonary blood flow and pressure to quantify PVR.

CYP epoxygenases are enzymes that produce pulmonary vasoconstricting EETs by metabolizing AA (Fig. 1) (4), and treatment with the selective CYP epoxygenase inhibitor MS-PPOH impairs HPV in sEH^+/+ mice by reducing levels of EETs (30, 36). In the present study we examined whether sEH^−/− mice are also susceptible to impairment of HPV when CYP-epoxygenase mediated EET-generation is inhibited. Treatment with the CYP epoxygenase inhibitor MS-PPOH impaired HPV in both sEH^−/− and sEH^+/+ mice (Fig. 2C). These findings demonstrate that HPV is dependent on the ability of CYP epoxygenases to generate EETs in sEH^−/− and sEH^+/+ mice.

We and others (12, 28, 29) found that challenge with LPS markedly impaired HPV in sEH^+/+ mice. The main finding of the present study is that mice with a sEH deficiency were protected from the LPS-induced impairment of HPV after LPS challenge and that HPV was restored in sEH^+/+ mice that were treated with an sEH inhibitor (Fig. 3, A and B). The enhancement of HPV in LPS-challenged mice that lacked active sEH was associated with an increased Pa_O2 during LMBO (Fig. 4, A and B). The observed increase in HPV and Pa_O2 highlights the important ability of HPV to reduce flow through nonventilated lung regions (22, 34).

There are at least two possible mechanisms by which the absence of active sEH might enhance HPV after challenge with LPS. The impairment of HPV in LPS-challenged sEH^+/+ mice was associated with higher levels of DHETs, suggesting increased EET degradation (Fig. 5, A and E), which might reflect a lower level of pulmonary vasoconstricting EETs. In previous studies, 11,12-EET was shown to be a potent pulmonary vasoconstrictor (13, 14). In the present study, preserved HPV in LPS-challenged sEH^−/− mice was associated with reduced 14,15- and 11,12-DHET levels in lung tissue, suggesting reduced 14,15- and 11,12-EET degradation (Fig. 5, A and B). Reduced degradation of 14,15- and 11,12-EETs in LPS-challenged sEH^−/− mice might imply that 14,15-EET and 11,12-EET are both important pulmonary vasoconstrictors and that both 14,15- and 11,12-EET may be responsible for the preservation of HPV in these mice.

A second possible mechanism by which the absence of sEH activity may enhance HPV involves inhibition of inflammation. Previous investigators showed that sEH inhibition reduced LPS-induced TNF-α release and NF-κB p65 nuclear translocation in human endothelial cells in vitro (1). Inhibition of sEH also reduced hypotension 6 h after LPS challenge in mice (20). In the present study, sEH^+/+ mice were treated with IK-950 only 30 min before measurement of HPV and 22 h after LPS challenge. Therefore, the effects of sEH inhibition by IK-950 on HPV are most likely caused by a direct effect on pulmonary vascular tone rather than an anti-inflammatory effect.

Previous investigators found that endotoxemia, which is commonly present in human ARDS, impairs HPV (12, 28, 29, 32, 33, 35). In the present study, the lack of sEH activity protects mice from LPS-induced impairment of HPV and the decrease in systemic arterial oxygenation. Deng and colleagues (6) demonstrated that, after exposure to LPS, sEH^−/− mice had a significant reduction in endotoxin-induced activation of nuclear factor NF-κB signaling, cellular adhesion molecules, chemokine and cytokine expression, and neutrophil infiltration in lung tissue. The reduced pulmonary cytokine expression in sEH^−/− mice after exposure to LPS in the present study is likely due to the anti-inflammatory effects of EETs, which inhibit NF-κB activation (7). The decrease in pulmonary cytokine mRNA level in sEH^−/− mice after LPS challenge suggests that there may be added, long-term beneficial effects associated with inhibition of sEH.

In the present murine model of endotoxemia, deficiency or inhibition of sEH did not affect PAP at baseline or after LPS challenge with or without LMBO, suggesting that sEH inhibition does not increase overall PVR. Nonetheless, as in the case with any vasoconstrictor, it is possible that sEH inhibition will increase PVR and thereby increase right ventricular afterload in some patients with acute lung injury, such as those who have coexisting increased pulmonary vascular reactivity. Effects of sEH inhibition on PVR and right ventricular function need to be carefully assessed before sEH inhibitors are considered for patients with lung injury.

In conclusion, in this study we demonstrate that after LPS challenge HPV was preserved in sEH^−/− mice and restored in sEH^+/+ mice treated with the sEH inhibitor IK-950. Preservation of HPV was accompanied by increased levels of systemic arterial oxygenation in both groups. A challenge with LPS induced an increase in the degradation of lung 14,15- and 11,12-EETs and in pulmonary cytokine mRNA levels in sEH^+/+ mice. These changes were not detected in mice that lacked sEH. Because inhibition of sEH enhances HPV and is also able to inhibit inflammation, inhibitors of this enzyme might improve lung function and decrease pulmonary inflammation in patients with ARDS.

GRANTS

This work was supported by funds of the German Research Foundation (DFG WE 5471/2-1) (M. Wepler), Grant K08HL111210 from the National Heart, Lung, and Blood Institute (R. Malhotra), NIH Grant 5R01-EY022746 (E. S. Buys), NIH Grant R01DK082971 and the LeDucq Foundation (D. B. Bloch), and funds of the Department of Anesthesia, Critical Care and Pain Medicine at Massachusetts General Hospital (W. M. Zapol).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

M.W., A.B., and W.M.Z. conceived and designed research; M.W., A.B., and M.D.B. performed experiments; M.W., A.B., M.D.B., D.P., R.M., E.S.B., P.R., F.I., D.B.B., and W.M.Z. analyzed data; M.W., A.B., M.D.B., D.P., R.M., E.S.B., P.R., F.I., D.B.B., and W.M.Z. interpreted results of experiments; M.W. prepared figures; M.W. and D.B.B. drafted manuscript; M.W., A.B., M.D.B., D.P., E.S.B., P.R., F.I., D.B.B., and W.M.Z. edited and revised manuscript; M.W., A.B., M.D.B., D.P., R.M., E.S.B., P.R., F.I., D.B.B., and W.M.Z. approved final version of manuscript.