Sphingosine Kinase 1 Plays an Important Role in Atorvastatin-Mediated Anti-Inflammatory Effect against Acute Lung Injury

Atorvastatin is a 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) inhibitor and inhibits cholesterol synthesis. Recently, atorvastatin also showed anti-inflammatory effect in acute lung injury, ameliorating pulmonary gas-blood exchanging function. Sphingosine kinase 1 plays a central role in endothelial (EC) cytoskeleton rearrangement and EC barrier integrity regulation. In this study, the role of sphingosine kinase 1 in atorvastatin anti-inflammatory effect against acute lung injury was investigated. Both wild-type (WT) and SphK1−/− mice were challenged with high tidal volume ventilation (40 ml/kg body weight, 65 breathing/min, 4 hours). The acute lung injury was evaluated and the mechanisms were explored. In WT mice, atorvastatin treatment significantly decreased acute lung injury responding to high tidal volume ventilation (HT), including protein, cellular infiltration, and cytokine releasing; comparing to WT mice, SphK1−/− mice showed significantly worsen pulmonary injuries on HT model. Moreover, the atorvastatin-mediated anti-inflammatory effect was diminished in SphK1−/− mice. To further confirm the role of SphK1 in VILI, we then compared the inflammatory response of endothelial cells that were isolated from WT and SphK1−/− mice to cyclic stretching. Similarly, atorvastatin significantly decreased cytokine generation from WT EC responding to cyclic stretching. Atorvastatin also significantly preserved endothelial junction integrity in WT EC against thrombin challenge. However, the inhibitory effect of atorvastatin on cytokine generation induced by cyclic stretching was abolished on SphK1−/− mice EC. The endothelial junction integrity effects of atorvastatin also diminished on SphK1−/− mouse EC. Signal analysis indicated that atorvastatin inhibited JNK activation induced by cyclic stretch. SphK1 knockout also blocked atorvastatin-mediated VE-cadherin junction enhancement. In summary, by inhibition of MAPK activity and maintenance of EC junction homeostasis, SphK1 plays a critical role in atorvastatin-mediated anti-inflammatory effects in both cellular and in vivo model. This study also offers an insight into mechanical stress-mediated acute lung injury and potential therapy in the future.


Introduction
Atorvastatin is a member of statin family inhibiting of 3hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) and modulates cholesterol synthesis [1]. In addition to inhibiting cholesterol synthesis, statins have multiple effects on clinical purposes, such as meliorating the patient's quality of life [2,3]. Among these properties of statin, the anti-inflammatory effects and enhancement of pulmonary function have been widely investigated [4,5].
Simvastatin and rosuvastatin have anti-inflammatory effects against LPS and virus infection such as influenza A [6,7]. Similarly, atorvastatin also attenuated the lung injury and the inflammatory response induced by various stimuli [8,9].
Ventilation-induced lung injury (VILI) often happens in infant ICU, the patients with coma and in patients with ARDS. The acute lung injury may include the mechanic injury and alkalic metabolic disturbance [10][11][12][13]. The pathological basis of VILI includes volutrauma, barotrauma, atelectrauma, biotrauma, and shear strain [14]. The longterm effects include pulmonary dysplasia, mental health issues, and organ dysfunction [15,16]. The VILI-mediated cellular and molecular pathological changes include endothelial-epithelial barrier dysfunction and increasing endothelial inflammatory responses [17,18]. Interestingly, atorvastatin demonstrates therapeutic properties against VILI on rabbit isolated lung model, notably the improvement of alveolar capillary permeability and hemodynamics and the attenuation of lung injury induced by high tidal volume mechanical ventilation [4].
In present studies, we explored the mechanisms and importance of atorvastatin in high tidal volume ventilation-mediated acute lung injury using a sphingosine kinase 1 knockout mouse model. By regulating the SphK1-MAPK signal pathway, atorvastatin meliorated high tidal volume ventilation-mediated acute lung injury. In addition, atorvastatin also regulates endothelial junction integrity by rearrangement of cytoskeleton system. These findings may offer a new view and potential therapy for the high tidal volume ventilation-mediated acute lung injury.

Materials and Methods
2.1. Materials. The RT-PCR and qPCR kit were purchased from Applied Biosystems (Beverly, MA). The mouse ELISA kits were purchased from BioLegend (San Diego, CA) and Millipore Sigma (Burlington, MA). The CD31-Alexa and CD45-FITC antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). All other antibodies and reagents were purchased from Cell Signaling Technology (Beverly, MA). . The isolated mouse primary endothelial  cells were cultured in EGM-2 supplemented with 2% FBS,  hydrocortisone, hFGF, VEGF, ascorbic acid, hEGF, GA-1000, heparin, and R3-IGF-1 (Lonza, Walkersville, MD). The cells were cultured in 75 cm 2 flask at 37°C in 5% CO 2 and 95% air. Once the cells got confluence, the cells were used for experiment or frozen for future usage. Before the cells were used, the purity was checked by flow cytometry and maintained above 90%.

Cell Culture
2.3. SphK1 -/-Mouse Breeding and Models. The C57 black 6 (C57BL6) and SphK1 homozygous knockout (SphK1 -/-) mice were purchased from Shanghai Animal Model Center, Shanghai, China. SphK1 -/mice were backcrossed with C57BL6 mice for five generations to eliminate the background interference to the observed phenotypes. The siblings of the backcross of the SphK1 -/and C57BL6 mice were used in this study. All experiments and animal care procedures were reviewed and approved by the Institutional Animal Care and Use committee at West China Hospital of Sichuan University (protocol number: 2019292A). The homozygous SphK1 -/mice and the background control mice (WT, 8-12 weeks old) were treated with atorvastatin or vehicle (10 mg/ kg body weight overnight, IP) followed by high tidal volume ventilation (40 ml/kg bodyweight, 65 breaths/min, 4 hours). After ventilation, the mice were sacrificed and bronchoalveolar lavage (BAL) was harvested in select animals as previously described [32]. BAL fluid was assessed for cell counts and protein content. The lungs then were used for histological analysis using lung injury score [33].

Mouse Lung Endothelial Cell Isolation and Purification.
The SphK1 -/mouse and WT mouse blood in vessel were flushed with 20 ml PBS containing 10 U/ml heparin (Hospira, Lake Forest, IL) via the right ventricle. The mouse lungs were minced and digested in an enzyme cocktail of DMEM containing 1% bovine serum albumin, 2 mg/ml collagenase, 100 μg/ml DNase, and 2.5 mg/ml Dispase at 37°C for 1 hour. The result products were meshed though a 70 μm nylon cell strainer. Once resuspended in 5 ml PBS (containing 1% BSA) and pelleted by centrifuge, the cells were incubated with Fc blocker antibody for 15 min on ice. Then, the cells incubated with anti-CD31 and anti-CD45 for 15 min on ice. After washing with 5 ml PBS (containing 1%BSA), the pulmonary vascular endothelial cells (CD31 +-CD45 -) were sorted by flow cytometry (BD FACSAria III, Becton, Dickinson and Company, NJ). The CD31 + CD45cells from WT mice and SphK1 -/mice were incubated in gelatin-coated plates, and the medium was changed (EGM2 complete medium) next morning [34]. Once setting up the stable cell line, the mRNA of EC was extracted and qPCR was performed to quantitatively determine SphK1 mRNA level in EC. And western blots also were performed to test the protein level in EC. The data is represented as a mean normalized expression by GAPDH [35]. The primers used in real-time PCR are as follows: mouse SphK1: forward 5 ′ -ACAGTGGGCACCTTCTTTC, reverse 5′-CTTCTGCACCAGTGTAGAGGC; mouse GAPDH: forward 5′-CAACTACATGGTCTACATGTTC, reverse 5′-CACCAGTAGACTCCACGAC [36].

Cytokine Measurement in Cellular Medium and in BAL.
The endothelial cells isolated from SphK1 -/and WT mice were grown to 80% confluence in Flexcell cyclic stretch plate (Flexcell Tension System, Burlington, NC). After being treated with 5% atorvastatin overnight, the cells were cyclic stretched on Flexcell for 4 hours (18% elongation, 0.5 Hz). After finishing the cyclic stretch, the media were harvested and centrifuged briefly to remove the cellular debris. The supernatants were collected to measure the indicated cytokines. The cytokines in BAL were measured by the same kits that were used to measure the cytokines in cellular medium. The ELISA kits were performed according to the manufacturer's instruction.

Western
Blotting. Samples were harvested with cellular lysis buffer containing proteinase inhibitors and phosphatase inhibitors as per standard protocols. After sonication and centrifugation, the supernatant was collected. Laemmli sample buffer was added into the supernatant and then was boiled and subsequently analyzed by SDS-PAGE. After transferring to a nitrocellulose membrane (Bio-Rad, Inc., Hercules, CA), western blotting was performed using appropriate primary antibodies and horseradish peroxidaseconjugated secondary antibodies prior to visualization via chemiluminescence (Amersham Biosciences, Piscataway, NJ). Blot density was determined by Alpha Imager software (Alpha Innotech, San Leandro, CA).

Immunofluorescence
Microscopy. Confluent EC grown on coverslips were exposed to 1 U/ml thrombin for 1 hour, fixed with 3.7% formaldehyde, and permeabilized with 0.25% Triton X-100. After blocking with 2% bovine serum albumin for 1 hour, cells were exposed to primary antibodies of interest for 2 hours. Fluorescent-tagged secondary antibodies were applied for 60 min in the dark. Cells were imaged using a Nikon video imaging system [37].
2.9. Statistical Analysis. Student's t-test was used to compare the means of data from two experimental groups, while significant differences (P < 0:05) among multiple group comparisons were performed by one-way ANOVA followed by Tukey's test. Results were expressed as means ± SE.
Histology assay showed that HT induced a mass cell infiltration in alveolar interstitium in WT mice and atorvastatin inhibited this HT-mediated cell infiltration. In contrast, HT induced a higher cell infiltration in alveolar interstitium in SphK1 -/mice and atorvastatin only mildly inhibited this HT-mediated cell infiltration (Figures 1(c) and 1(d)).

Identification of Endothelial Cell Isolation from WT and
SphK1 -/-Mice. To further confirm the role of sphingosine kinase 1 in atorvastatin-mediated anti-inflammatory effect, we then isolated the endothelial cells from WT and SphK1 -/mouse lung. As shown in Figure 3(a), the endothelial cells were isolated (CD31+/CD45-) from WT mouse lung (SphK1 +/+ ) and from SphK1 -/mouse lung (SphK1 -/-) by flow cytometry (Figures 3(a) and 3(b)). The isolated endothelial cells showed classic morphological properties of endothelial cells, such a triangle cellular body and two major lamellipodia. Interestingly, the isolated SphK1 -/endothelial cells showed bigger and flatter cellular body with more lamellipodia (Figure 3(c)). After culture for several passages, the sphingosine kinase 1 mRNA and protein levels in WT and in SphK1 -/endothelial cells were confirmed by qPCR (Figure 3(d)) and western blots (Figure 3(e)).

Sphingosine Kinase 1 Knockout Diminished the Atorvastatin-Mediated Anti-Inflammatory Effect in Mouse
Endothelial Cells Responding to Cyclic Stretch. The isolated endothelial cells on Flexcell plates were cyclic stretched for 4 hours (18% elongation, 0.5 Hz), and the cytokines released from endothelial cells were measured by ELISA kits. As shown in Figure 4, the 4-hour cyclic stretch induced significantly higher cytokine generations in WT mouse endothelial cells than the static controls.    Figure 6: JNK plays a critical role in sphingosine kinase 1-mediated inflammatory response to cyclic stretch. The isolated ECs were seeded on Flexcell plates overnight in the presence of the indicated inhibitors. After cyclic stretching (CS) (18% elongation, 0.5 Hz, 4 hours) as described in Figure 4, the medium was harvested for ELISA. (a) IL6 and KC assay showed that SphK1 knockout significantly enhanced CS-mediated IL6 and KC generation. The SphK1 knockout-mediated IL6 generation enhancement was not significantly abolished by PD98059 but partially by SB203580 and completely by JNK inhibitor II. (b) The SphK1 knockout-mediated KC generation enhancement was not significantly abolished by PD98059 and SB203580, but completely by JNK inhibitor II ( * P < 0:05, * * P < 0:01).
Similar with WT endothelial cells, 18% elongation 0.5 Hz cyclic stretch also induced P38 and JNK phosphorylation increase in SphK1 -/mouse endothelial cells. Pretreatment of atorvastatin significantly inhibited JNK but not ERK and P38 phosphorylation ( Figure 5).

Sphingosine Kinase 1 Knockout-Mediated Cytokine Generation Enhancement in Endothelial Cells May Be
Regulated by JNK and P38. To further confirm the regulation of sphingosine kinase 1 knockout on MAPK, the isolated mouse endothelial cells were cyclic stretched in the presence of ERK, JNK, and P38 inhibitors. After 4-hour cyclic stretch, the cytokines in medium were measured. As shown in Figure 6, sphingosine kinase 1 knockout induced significant IL6 and KC generation in SphK1 -/endothelial cells comparing to WT endothelial cells. The IL6 generation induction in SphK1 -/endothelial cells was inhibited by JNK inhibitor and partially inhibited by SB203580 (P38 inhibitor) but not by PD58059 (ERK inhibitor) (Figure 6(a)). The CSmediated KC generation in SphK1 -/endothelial cells showed similar trends (Figure 6(b)).

Sphingosine Kinase 1 Plays a Critical Role in
Atorvastatin-Mediated Endothelial Junction Integrity Enhancement. As pretreatment of atorvastatin decreased both protein and cellular filtration on the in vivo model, we then investigated the importance of atorvastatin on the endothelial barrier integrity against pathogenic challenge like thrombin. We first tested whether atorvastatin changed SphK1 expression in mouse endothelial cells. As shown in Figure 7(a), atorvastatin did not change SphK1 in WT and SphK1 -/endothelial cells (upper panel). Atorvastatin also did not change another isotype of sphingosine kinase, SphK2 (middle panel).
We next investigated whether SphK1 is involved in the endothelial barrier integrity enhancement. Immunostaining showed that VE-cadherin mostly located in the cytosolic plasma and a few small gaps existed between WT endothelial To further confirm the role of SphK1 in atorvastatinmediated endothelial junction integrity enhancement effect, the permeabilities of monolayer WT and SphK1 -/endothelial cells in transwell plates were compared at both control and thrombin-challenging conditions (Figure 7(c)). Thrombin challenge caused FITC-dextran permeability significantly increased at WT endothelial cells compared to control. Additionally, the thrombin-mediated FITCdextran permeability was attenuated by pretreatment by atorvastatin (Figure 7(c)). Compared to WT endothelial cells, thrombin-challenged SphK1 -/endothelial cells had significantly higher FITC-dextran permeability. Moreover, this permeability increase was not significantly reversed by atorvastatin (Figure 7(c)).

Discussion
Based on the murine high tidal volume ventilation model and cellular models, the importance of sphingosine kinase in atorvastatin-mediated anti-inflammatory effects was investigated. Atorvastatin significantly attenuated VILImediated protein and cell infiltration as well as cytokine release on in vivo and cellular model of WT mice. Knockout SphK1 diminished the atorvastatin-mediated inflammatory inhibitory effects. Signal pathway assay showed that SphK1-mediated anti-inflammatory effect was partially through the JNK pathway. In addition, endothelial junction studies indicated that SphK1 plays a central role in endothelial junction reorganization and integrity maintenance, which also partially contributes to the anti-inflammatory effect of atorvastatin ( Figure 8).
Mechanical stress-mediated inflammatory response has its own unique properties, such as sever junctional damage and in homogenous lung injury due to uneven pleural pressure [38,39]. Long-term side effects include COPD [40]. The molecular mechanism of mechanical ventilation includes barotrauma, volutrauma, atelectrauma, and biotrauma [41]. Among all the mechanism basis of VILI, endothelial barrier is the most sensitive and plays a central role in the process of inflammatory response to VILI. The involvements of endothelium in VILI include (1) increased lung vascular permeability due to loss of endothelial cell barrier integrity, (2) mechanical stress-induced inflammatory cytokine release and cell infiltration, and (3) genetic and epigenetic regulation of critical target genes such as junction protein and signal proteins that are involved in VILI response [17]. The anti-inflammatory effects of statins have been widely investigated on various models [6,[42][43][44]. These antiinflammatory effects include decreasing proinflammatory cytokine generation, enhancing endothelial cell junction integrity, decreasing ROS and chemokine generation, decreasing lipid oxidation, and inflammation [31,45]. The beneficial effects of statins on mechanical stress-mediated acute lung injury have been relatively poorly understood [43]. In current studies, we identified the atorvastatinmediated anti-inflammatory mechanisms on cellular model and VILI model. The inhibitory effect of atorvastatin was mediated by inhibition of MAPK activity, which is similar to other statins [32,46]. Moreover, sphingosine kinase 1 also contributes to atorvastatin-mediated endothelial junction integrity enhancement [47,48].
SphK1 is involved in many critical cellular activities [49,50]. Sphingosine-1-phosphate, a product of SphK1 and sphingosine, has shown enhancing endothelial junction. By binding with the EDG receptors, S-1-P activates Rac GTPase, causes cytoskeleton reorganization, and improves endothelial junction tightness [51]. Interestingly, pharmaceutical inhibitor studies demonstrate that the MAPK signal F -a c t in Figure 8: Schematic description of the potential role of SphK1 in atorvastatin-mediated anti-inflammatory effect. Under control condition, sphingosine exists as an inactive form and there is balance between the stress fiber system and the cortactin ring system. When atorvastatin is added, stress fiber is inactivated, the cortical ring is enhanced, and the VE-cadherin junction is tightened. Meanwhile, atorvastatin inhibits the MAPK signal pathway and attenuates inflammatory responses. Sphingosine kinase 1 is a mediator in these processes.

11
Mediators of Inflammation pathway is not involved in this process [52]. Knockout SphK1 resulted in the increase of LPS-or PAR-induced edema in mouse lung and cellular barrier permeability in cellular model [27]. Recent studies indicated that SphK1 stabilizes JNK and inhibits the interaction of JNK and JNKinteracting protein 3 (JIP3), further abrogating the activation of NADPH oxidase and NF-κB activation [28]. In addition, SphK1 also plays a protective role in LPS-mediated inflammatory response in the central nervous system as well as other inflammatory responses [53,54]. The results in the current study show that SphK1 plays a critical role in atorvastatin-mediated protective effects against high tidal volume ventilation-induced lung injury (VILI). One of the major mechanisms of atorvastatin may mediate MAPK activity inhibition during the inflammatory response. Knockout of the sphingosine kinase 1 abolished the inhibition effect of atorvastatin on MAPK activity indicating that sphingosine kinase 1 plays a vital role in atorvastatin-mediated MAPK inactivation. These are also supported by VILI-mediated cytokine releases from SphK1 -/cellular and mice.
Another important contribution of sphingosine kinase 1 in atorvastatin-mediated anti-inflammatory effects is the enhancement of endothelial junction integrity. Endothelial junction includes many components and is regulated by multiple dynamic systems. So far, endothelial junction includes atypical tight junction, which consists of ZO-1, occluding, claudins, JAM, etc. Another major junction is adherens junction, which consists of VE-cadherin, α-catenin, β-catenin, etc. Moreover, endothelial junction integrity also depends on cytoskeleton protein such as F-actin and MLCK as they associate with and tightly regulate junctional proteins. When stimuli like LPS or thrombin bind with their receptor on endothelial cells, the relative signal pathways regulate cytoskeleton rearrangement. High tidal volume ventilation may activate cytoskeleton system by modulating mechanical receptors. The consequence of these challenges to the lung will present as edema and inflammatory cell infiltration. SphK1 turns sphingosine into an active form, sphingosine-1-phosphate (S-1-P). S-1-P in turn regulates the VEcadherin junction and cortical actin ring [55]. Consistently, endothelial junction imaging and permeability assays indicate that sphingosine kinase 1 is a critical mediator of atorvastatinmediated endothelial junction integrity enhancement.
In summary, the regulatory mechanism of atorvastatin on anti-inflammatory response to mechanical stress has been investigated. By inactivation of MAPK in endothelial cells and enhancement of endothelial junction integrity, atorvastatin attenuates VILI-mediated inflammatory response. This anti-inflammatory response of atorvastatin was abolished by sphingosine kinase 1 knockout. These findings in the current study open a new field for us to explore the new medicine for acute lung injury, such as high tidal volume ventilation-mediated lung injury.

Data Availability
The data (Excel) used to support the findings of this study are available from the corresponding author upon request.