Hyperoxia-induced S1P1 signaling reduced angiogenesis by suppression of TIE-2 leading to experimental bronchopulmonary dysplasia

Abstract

Introduction: We have earlier shown that hyperoxia (HO)-induced sphingosine kinase 1 (SPHK1)/sphingosine-1-phosphate (S1P) signaling contribute to bronchopulmonary dysplasia (BPD). S1P acts through G protein-coupled receptors, S1P₁ through S1P₅. Further, we noted that heterozygous deletion of S1pr1 ameliorated the HO-induced BPD in the murine model. The mechanism by which S1P₁ signaling contributes to HO-induced BPD was explored.

Methods: S1pr1^+/+ and S1pr1^+/- mice pups were exposed to either room air (RA) or HO (75% oxygen) for 7 days from PN 1-7. Lung injury and alveolar simplification was evaluated. Lung protein expression was determined by Western blotting and immunohistochemistry (IHC). In vitro experiments were performed using human lung microvascular endothelial cells (HLMVECs) with S1P₁ inhibitor, NIBR0213 to interrogate the S1P₁ signaling pathway.

Results: HO increased the expression of S1pr1 gene as well as S1P₁ protein in both neonatal lungs and HLMVECs. The S1pr1^+/- neonatal mice showed significant protection against HO-induced BPD which was accompanied by reduced inflammation markers in the bronchoalveolar lavage fluid. HO-induced reduction in ANG-1, TIE-2, and VEGF was rescued in S1pr1^+/- mouse, accompanied by an improvement in the number of arterioles in the lung. HLMVECs exposed to HO increased the expression of KLF-2 accompanied by reduced expression of TIE-2, which was reversed with S1P₁ inhibition.

Conclusion: HO induces S1P₁ followed by reduced expression of angiogenic factors. Reduction of S1P₁ signaling restores ANG-1/ TIE-2 signaling leading to improved angiogenesis and alveolarization thus protecting against HO-induced neonatal lung injury.

Keywords: Angiogenesis.; Neonatal lung disease; Oxidative stress; Sphingosine 1 phosphate receptor.

Fig. 1

Hyperoxia (HO) increases the expression of S1P₁. WT neonatal mice were exposed to HO (75% O₂) or room air (RA) from postnatal day 1 (PN1) for 7 days. Quantitative real time-PCR showed increased expression of S1pr1 following HO (A). Western blot analysis of whole lung tissue lysates showed increased expression of S1P₁ following HO in WT compared to RA controls (B, C). Statistical analysis was done with ANOVA test. *** indicates p < 0.0001, n = 5–8/group

Fig. 2

S1pr1 heterozygous (S1pr1^+/−) newborn mice showed protection against hyperoxia (HO) induced lung injury whereas S1pr3^−/− did not. Representative H&E photomicrographs of lung sections obtained from WT or S1pr1^+/− neonatal mice exposed to HO (75% O₂) from postnatal day 1 (PN1) for 7 days showed improved alveolarization as compared to S1pr1^+/+ controls (A). Objective assessment of alveolarization demonstrated smaller alveoli with shorter mean linear intercept (MLI) in the S1pr1^+/− group as compared to controls (B) and less inflammatory exudate in bronchoalveolar lavage fluid (C). Representative H&E photomicrographs of lung sections obtained from WT (S1pr3^+/+) or S1pr3^−/− neonatal mice exposed to HO (75% O2) from PN1 for 7 days showed no protection against HO-induced BPD as compared to WT controls (D). Objective assessment of alveolarization demonstrated larger alveoli with longer mean linear intercept (MLI) in both the HO-exposed group as compared to RA controls (E) and increased inflammatory exudate in bronchoalveolar lavage fluid (F). Statistical analyses were done with ANOVA. ***p < 0.001, n = 5–8/group

Fig. 3

Hyperoxia (HO) reduced the expression of TIE-2 and ANG-1. Western blot analysis of mouse lung tissue subjected to HO showed decreased expression of Tie2 at RNA (A) and TIE-2 at protein levels (B) in WT mice. Immunohistochemistry (IHC) of lung tissue showed HO induced decrease in expression of TIE-2 in S1pr1^+/+ mice which recovered in S1pr1^+/− mice (C, D). IHC showed a decreased expression of ANG-1 in the HO exposed S1pr1^+/+ mice which was rescued in the S1pr1^+/− mice exposed to HO (E, F). The statistical analysis was carried out using ANOVA where ** indicate p < 0.01, ***p < 0.001, ****p < 0.0001, n = 5–8/group

Fig. 4

Hyperoxia (HO) exposure reduced the expression of VEGF and blood vessel count. HO exposure resulted in reduced VEGF in lung tissue of S1pr1^+/+ mice as shown by immunohistochemistry (A, B). This reduction was ameliorated in S1pr1^+/− mice which showed higher expression of VEGF. HO was accompanied by a significant reduction in the number of order-1 arterioles in the WT mice which was improved in the S1pr^+/− mice (C). The statistical analysis was carried out using ANOVA where *** indicate p < 0.001, n = 5–8/group

Fig. 5

Hyperoxia (HO) induced the expression of Klf2 and Klf4 accompanied by a translocation of NF-κB to the nucleus. In silico analyses showed a strong probability for NF-κB and KLF-2 binding to the promoter site of S1P₁ with a score of 0.98 and 0.92, respectively (A–C). Real time PCR analysis of lung tissues obtained from WT neonatal mice exposed to HO (75% O₂) showed increased mRNA expression of Klf2 (D), and Klf4 (E). Western blot analysis showed increased expression of KLF-4 (F, G). Immunofluorescence showed an increased translocation of NF-κB to the nucleus (H, I). Statistical analysis was done using ANOVA test. *** for p < 0.001 and **** for p < 0.0001, HO vs RA and HO vs treatment. n = 6–8/group

Fig. 6

Hyperoxia (HO) increased the expression of S1P₁ in human lung microvascular endothelial cells (HLMVECs) and S1P₁ inhibition blocked HO-induced reactive oxygen species (ROS) generation. Western blot analysis of HLMVECs exposed to 95% HO or NO for 24 h showed increased expression of S1P₁ (A, B). Luciferase reporter assay on HLMVECs exposed to 95% O₂ or RA for 24 h showed increased S1PR1 gene transcription following HO (C). VEGF was used as a known promoter of S1PR1. HLMVECs treated with the vehicle or S1P₁ inhibitor and then exposed to HO for 3 h showed increased production of ROS as determined by the DCFDA fluorescence method. The HO induced increase in the production of ROS was inhibited by the S1P₁ inhibitor (D, E). The statistical analysis was carried out using ANOVA where * indicate p < 0.05, **p < 0.01, ***p < 0.001, n = 5

Fig. 7

Hyperoxia (HO)-induced alteration in TIE-2 and KLF-2 were restored by S1P₁ inhibition. Hyperoxia (HO)-reduced TIE-2 expression in human lung microvascular endothelial cells (HLMVECs) which was restored with the S1P₁ inhibitor treatment (A, B). HLMVEC exposed to HO increased the expression of KLF-2 which was reduced by inhibitor treatment (C, D). Statistical analysis was done using ANOVA test. * for p < 0.05, *** for p < 0.001 and, HO vs RA and HO vs treatment

Fig. 8

Proposed mechanism. Hyperoxia (HO) triggers an increase in KLF-2 and KLF-4 and activates NF-κB. These transcription factors putatively bind to S1P₁ promoter resulting in increased transcription of S1P₁. The increased S1P₁ results in reduced ANG-1, TIE-2, and VEGF which are factors important for blood vessel maintenance and angiogenesis, there by inhibiting angiogenesis. S1P₁ inhibition reversed the inhibition of angiogenesis by increased S1P₁ ameliorating BPD. Thus, S1P₁ could serve as a potential druggable target for treating BPD