Differential effects of sphingosine 1-phosphate receptors on airway and vascular barrier function in the murine lung

Abstract

The therapeutic options for ameliorating the profound vascular permeability, alveolar flooding, and organ dysfunction that accompanies acute inflammatory lung injury (ALI) remain limited. Extending our previous finding that the intravenous administration of the sphingolipid angiogenic factor, sphingosine 1-phosphate (S1P), attenuates inflammatory lung injury and vascular permeability via ligation of S1PR(1), we determine that a direct intratracheal or intravenous administration of S1P, or a selective S1P receptor (S1PR(1)) agonist (SEW-2871), produces highly concentration-dependent barrier-regulatory responses in the murine lung. The intratracheal or intravenous administration of S1P or SEW-2871 at < 0.3 mg/kg was protective against LPS-induced murine lung inflammation and permeability. However, intratracheal delivery of S1P at 0.5 mg/kg (for 2 h) resulted in significant alveolar-capillary barrier disruption (with a 42% increase in bronchoalveolar lavage protein), and produced rapid lethality when delivered at 2 mg/kg. Despite the greater selectivity for S1PR(1), intratracheally delivered SEW-2871 at 0.5 mg/kg also resulted in significant alveolar-capillary barrier disruption, but was not lethal at 2 mg/kg. Consistent with the S1PR(1) regulation of alveolar/vascular barrier function, wild-type mice pretreated with the S1PR(1) inverse agonist, SB-649146, or S1PR(1)(+/-) mice exhibited reduced S1P/SEW-2871-mediated barrier protection after challenge with LPS. In contrast, S1PR(2)(-/-) knockout mice as well as mice with reduced S1PR(3) expression (via silencing S1PR3-containing nanocarriers) were protected against LPS-induced barrier disruption compared with control mice. These studies underscore the potential therapeutic effects of highly selective S1PR(1) receptor agonists in reducing inflammatory lung injury, and highlight the critical role of the S1P delivery route, S1PR(1) agonist concentration, and S1PR(1) expression in target tissues.

Figures

Figure 1.

Effects of sphingosine 1–phosphate (S1P) and selective S1P receptor (S1PR1) agonist (SEW-2871) on LPS-induced increases in lung permeability. (A) Effects of S1P (0.001 and 0.1 mg/kg) in C57BL/6 mice delivered intratracheally, 2 hours after LPS injection (2.5 mg/kg) and harvested 16 hours later. Bronchoalveolar lavage (BAL) protein accumulation was significantly decreased in S1P-treated mice compared with the LPS-only group (*P ≤ 0.05). Similarly, intravenously delivered S1P (0.03 mg/kg) produces significant barrier enhancement 18 hours after LPS injection, reflected by decreased BAL protein concentrations (*P ≤ 0.05 versus vehicle, **P ≤ 0.05 versus LPS) compared with LPS-treated animals (n = 5). Veh, vehicle. (B) Results after injection of SEW-2871 (SEW) both intratracheal and intravenous, 0.1 and 0.3 mg/kg, respectively, after LPS injection (2.5 mg/kg) and harvested after 16 and 18 hours, respectively. The accumulation of BAL protein was significantly decreased, reflecting lung barrier function enhancement (*P ≤ 0.05 versus vehicle, **P ≤ 0.05 versus LPS).

Figure 2.

Histologic evaluation of barrier enhancement properties of SEW-2871 in LPS-induced acute lung injury. Sections were stained with hematoxylin–eosin for histologic evaluations of vehicle (A) and SEW-2871 (B), respectively, without evidence of inflammation. (C) In contrast, histologic assessments of LPS-mediated barrier dysfunction revealed a moderate increase in infiltrating polymorphonuclear leukocytes (PMNs) in lung tissue and prominent increases in alveolar edema. (D) These LPS-mediated histologic alterations were attenuated by treatment with SEW-2871. (E) Relative levels of leukocyte infiltration, quantified as aggregated histologic scores, reflect SEW-2871–mediated decreases in leukocyte influx compared with LPS alone.

Figure 3.

Effects of S1PR1 inhibition on S1P/SEW-2871–mediated barrier enhancement. In these experiments, 8–10-week-old C57Bl/6 mice were treated with 0.1 mg/mouse SB-649146 (intravenous) 30 minutes after challenge with LPS, LPS/S1P, or LPS/SEW-2871. Mice were killed 18 hours later, and BAL fluid was collected for evaluation of protein levels. (A) The administration of 0.1 mg/mouse of SB-649146 completely reversed S1P-mediated barrier enhancement properties, as reflected by raising BAL total protein concentrations significantly (*P = 0.034, S1P/LPS/SB-649146 versus S1P/LPS; **P ≤ 0.05, S1P/LPS versus LPS). (B) Administering 0.1 mg/mouse of SB-649146 attenuates the barrier-protective effects of SEW-2871 significantly in LPS-challenged mice, as reflected by significant increase in BAL total protein level (*P = 0.044, SEW-2871/LPS/SB-649146 versus SEW/LPS; **P ≤ 0.05, SEW-2871/LPS versus LPS) (n = 4).

Figure 4.

Responses of S1PR1+/− heterozygous mice to murine LPS-induced acute lung injury. (A) S1PR1+/− or wild-type (WT) mice were challenged with LPS (intratracheally) and killed 18 hours later. Disruption of vascular barrier integrity was reflected by significant increase in BAL PMNs of S1PR1+/− mice compared with WT mice (#,†P ≤ 0.05, S1PR1+/−/LPS versus LPS; #P ≤ 0.05, S1PR1/LPS versus S1PR1+/−/vehicle). (B) Effects of S1P on lung tissue myeloperoxidase (MPO) activity in S1P1+/− heterozygous mice challenged with LPS. S1PR1+/− or wild-type mice were treated with an intravenous injection of S1P 30 minutes after LPS. As reflected by lung-tissue MPO activity, barrier-enhancing properties of S1P were significantly attenuated in LPS-challenged S1PR1+/− mice compared with WT mice (*P ≤ 0.05, WT-LPS versus WT-LPS/S1P; **P ≤ 0.05, WT-S1P/LPS versus S1PR1+/+-S1P/LPS).

Figure 5.

Responses of S1PR2−/− knockout (KO) mice to murine LPS-induced acute lung injury. BAL was collected 18 hours after intratracheal injection of LPS, and total protein level was measured in S1PR2−/− KO and WT S1P+/+ mice. (A) BAL protein concentration in S1PR2−/− KO mice was significantly lower (*P = 0.02 and **P ≤ 0.05 LPS versus vehicle) than in WT mice, supporting our hypothesis of the barrier-disruptive role of S1P2 receptors. (B) No changes occurred in white blood cell counts in S1PR2−/− KO mice compared with WT mice.

Figure 6.

LPS-induced acute lung injury responses in mice with S1PR3 silencing via angiotensin-converting enzyme (ACE) antibody–conjugated nanocarriers with short, interfering RNA (siRNA)–S1P3 cargo. (A) Western blot analysis of lung tissue after ACE antibody–conjugated nanocarrier delivery of siRNA targeting S1PR3. Nanocarriers were injected into jugular veins to deliver 10 mg/kg S1PR3 siRNA specifically into lung vessels, and after 5 days, tissues were assessed by examining lung homogenates by immunoblotting in control (NP control) and in siS1PR3-depleted murine lungs. (B) Lung injury induced by LPS is attenuated by intravenous injection of S1PR3 siRNA, using ACE antibody–conjugated nanoparticle delivery as reflected by BAL albumin content (*P < 0.01) compared with LPS controls. **P < 0.05 in siS1PR3/vehicle versus LPS injured mice. (C) Graphic representation of tissue albumin content indicates ACE antibody–conjugated nanocarrier delivery significantly reduces tissue albumin content in LPS-induced lung injury (*P < 0.01) compared with the LPS group. **P < 0.05, significant increases in tissue albumin after LPS, compared with siS1PR3 control mice.