Effect of sphingosine kinase 1 inhibition on blood pressure

Abstract

Accumulating evidence suggests that sphingosine kinase 1 (SphK1) plays a key role in carcinogenesis by regulating cyclooxygenase-2 (COX-2) expression. Recent clinical studies have revealed that COX-2 inhibitors cause adverse cardiovascular side effects, likely due to inhibition of prostacyclin (PGI₂). In this work, we investigated the roles of SphK1 inhibition on blood pressure (BP). The results show that lack of SphK1 expression did not exacerbate angiotensin II (Ang II)-induced acute hypertension, whereas celecoxib, a COX-2 inhibitor, augmented and sustained higher BP in mice. Interestingly, SphK1-knockout mice inhibited prostaglandin E₂ (PGE₂) but not PGI₂ production in response to Ang II, whereas celecoxib blocked both PGE₂ and PGI₂ production. Mechanistically, SphK1 down-regulation by siRNA in human umbilical vein endothelial cells decreased cytokine-induced PGE₂ production primarily through inhibition of microsomal PGE synthase-1 (mPGES-1), not COX-2. SphK1 down-regulation also decreased MKK6 expression, which phosphorylates and activates P38 MAPK, which, in turn, regulates early growth response-1 (Egr-1), a transcription factor of mPGES-1. Together, these data indicate that SphK1 regulates PGE₂ production by mPGES-1 expression via the p38 MAPK pathway, independent of COX-2 signaling, in endothelial cells, suggesting that SphK1 inhibition may be a promising strategy for cancer chemoprevention with lack of the adverse cardiovascular side effects associated with coxibs.—Furuya, H., Wada, M., Shimizu, Y., Yamada, P. M., Hannun, Y. A., Obeid, L. M., Kawamori, T. Effect of sphingosine kinase 1 inhibition on blood pressure.

Cyclooxygenase-2 (COX-2) is a pivotal proinflammatory enzyme, which regulates prostaglandin E₂ (PGE₂) production. Numerous studies reveal that the COX-2/PGE₂ pathway plays an important role in colon carcinogenesis. Selective COX-2 inhibitors significantly reduced colon carcinogenesis in rodents (1, 2) and inhibited intestinal polyp formation in patients with familial adenomatous polyposis (3). In addition, PGE₂ administration enhanced colon carcinogenesis (4) and overcame indomethacin-reduced intestinal polyp formation in Apc^Min mice (5). Thus, COX-2 can be a promising target for colon cancer prevention. However, recent clinical trials have revealed that COX-2 inhibitors increase cardiovascular risks, such as hypertension and atherosclerosis (6–9). The mechanism of this effect is suggested, in part, by evidence that inhibition of COX-2 can block the production of prostacyclin (PGI₂) without affecting the synthesis of thromboxane A₂ (TXA₂). TXA₂ and PGI₂ are functionally antagonistic prostanoids (7, 10); i.e., TXA₂ causes platelet aggregation and vasoconstriction, while PGI₂ inhibits platelet aggregation and causes vasodilation. Thus, selective COX-2 inhibition causes a shift in balance between TXA₂ and PGI₂ toward TXA₂ dominance and may enhance cardiovascular risks.

Sphingolipids, especially ceramide, sphingosine, and sphingosine 1-phosphate (S1P), play key roles as regulatory molecules in cancer development (11). S1P promotes cell proliferation and survival, and regulates angiogenesis, whereas sphingosine and ceramide inhibit cell proliferation and stimulate apoptosis. Sphingosine kinase (SphK) is an enzyme that phosphorylates sphingosine to produce S1P and is a critical regulator of sphingolipid-mediated functions. Two isotypes of SphK have been characterized, SphK1 and SphK2, numbered in order of their discovery (12). Several studies have shown that the SphK1/S1P pathway mediates COX-2 expression and PGE₂ production in several cell types (13–16). Previously, we found that SphK1 deficiency significantly reduced azoxymethane (AOM)-induced aberrant crypt foci (ACFs; preneoplastic lesions) and colon cancer development in an inflammation-related colon carcinogenesis mouse model by regulating the COX-2/PGE₂ pathway (17). Recently, we also found that SphK1-knockout (KO) mice significantly reduce carcinogen-induced tongue carcinogenesis, indicating that SphK1 is involved in head and neck squamous cell carcinoma development (18). Thus, we believe that the SphK1/S1P pathway plays a pivotal role in carcinogenesis. On the other hand, it is speculated that inhibition of the SphK1/S1P pathway may increase cardiovascular risks by disrupting the COX-2 signaling and the subsequent decrease in PGI₂ production. In addition, S1P also has COX-2-independent cardiovascular functions. S1P generates signals in the vasculature that can result in either vasoconstriction or vasodilation (19, 20). Moreover, it is also possible that the SphK1/S1P pathway regulates microsomal PGE synthase 1 (mPGES-1; the main enzyme of PGE₂ synthesis) and affects the balance between TXA₂ and PGI₂. Several studies have reported that S1P may activate extracellular signal-regulated kinase 1/2 (ERK1/2) and p38 mitogen-activated protein kinase (MAPK) pathway and early growth response-1 (Egr-1) expression in various cells (21–25). Egr-1 is regulated by both ERK1/2 and p38 MAPK pathway and has been reported to regulate mPGES-1 expression by binding specifically to GC-rich elements in the promoter region (26). On the basis of these results, it is speculated that SphK1/S1P pathway plays an important role in cardiovascular functions, but it is still unclear whether inhibition of SphK1/S1P pathway increases cardiovascular risks in vivo.

In the present study, we provide in vivo evidence for the effect of lack of SphK1 expression in angiotensin II (Ang II)-induced acute hypertension compared with celecoxib, a selective COX-2 inhibitor. We also show that down-regulation of SphK1 by small interfering RNA (siRNA) reduces only PGE₂ production via a decrease in mPGES-1 induced by cytokines, while celecoxib reduces both PGE₂ and PGI₂ production through inhibition of COX-2 in human umbilical vein endothelial cells (HUVECs). In addition, we demonstrate that down-regulation of SphK1 also decreases the expression of mitogen-activated protein kinase kinase 6 (MKK6) and Egr-1 in HUVECs. We then discuss the implications of these results for the role of the SphK1/S1P pathway and its potential in cancer chemoprevention and chemotherapy.

MATERIALS AND METHODS

Materials

Celecoxib, a selective COX-2 inhibitor, was purchased from LC Laboratories (Woburn, MA, USA). Human recombinant tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) were purchased from PeproTech (Rocky Hill, NJ, USA). Ang II was purchased from Sigma-Aldrich (St. Louis, MO, USA).

Animals

Mice were housed and handled in the laboratory animal resources facilities at the Medical University of South Carolina (MUSC) and the University of Hawaii (UH). Mice were maintained under controlled conditions of humidity (50±10%), light (12-h light-dark cycle) and temperature (23±2°C). All mouse experiments were approved by the institutional animal care and use committees at MUSC and UH. SphK1 homozygous KO mice and SphK2 homozygous KO mice of the 129SV-C57BL/6 background, kind gifts from Dr. Richard L. Proia [U.S. National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)/National Institutes of Health (NIH), Bethesda, MD, USA], were backcrossed to C57BL/6 wild-type (WT) mice (purchased from Charles River Laboratories, Wilmington, MA, USA) ≥10 times (27). Genotypes of SphK1- and SphK2-KO mice were determined by PCR analysis of genomic DNA isolated from tail biopsies (27).

Blood pressure (BP) measurement

Systolic BP (SBP) was measured in anesthetized male mice (8-12 wk old) using a computerized CODA high-throughput noninvasive BP acquisition system (Kent Scientific Corp., Torrington, CT, USA). WT C57BL/6, SphK1-KO, and SphK2-KO mice were administrated celecoxib [100 mg/kg body weight (BW)] or vehicle (polyethylene glycol 400) by oral gavage 1 h before SBP measurement. These mice were anesthetized using ketamine (200 mg/kg BW, i.p.), and then were injected with Ang II (640 μg/kg BW, i.p.). SBP of these mice was measured before Ang II injection for 5 times, and the average of the 5 SBP data are expressed as basal levels. After Ang II injection, SBP was monitored for 35 min (∼2×/min). Data were collected and analyzed using CODA data acquisition software (Kent Scientific).

Determination of prostanoid levels in kidney and aorta, and S1P levels in blood

Kidney, aorta, and blood were collected at 5 and 25 min after Ang II injection from WT C57BL/6 mice with and without celecoxib treatment, SphK1-KO mice, and SphK2-KO mice, and tissues were snap-frozen in liquid nitrogen. Kidney and aorta (50 mg) were homogenized at ice temperature with a microtube pestle and vortexed thoroughly for 2 min in 0.25 ml of 0.1 M Tris–HCl buffer containing 5.6 μM indomethacin and 15% methanol. Samples were centrifuged at 400 g at 4°C for 10 min. Supernatants were collected, and then levels of PGE₂ and 6-keto-prostaglandin F_1α (PGF_1α) were quantified by enzyme immunoassay (EIA; Cayman Chemical Co., Ann Arbor, MI, USA), according to the manufacturer's instructions. The results were expressed as picograms per milligram of protein, calculated according to the 6-keto PGF_1α and PGE₂ standard curve. All assays were performed in triplicate. S1P in blood samples was analyzed by sphingolipid profiling using tandem mass spectrometry with positive-mode electrospray ionization in the MUSC Lipidomics Core Facility, as described previously (15). Results are expressed as picomoles of sphingolipids per 100 μl of blood.

Cells

HUVECs were purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA) and maintained in an endothelial medium system (EGM-2; Cambrex, East Rutherford, NJ, USA) at 37°C in a 5% CO₂ incubator.

RNA interference

The gene silencing of human SphK1 was performed essentially as described before using sequence-specific siRNA reagents (Qiagen, Valencia, CA, USA; ref. 28). The human SphK1 (29) silencing target starts at the 70th nucleotide from the start codon. The siRNA for human SphK1 (GGGCAAGGCCUUGCAGCUCdTdT and GAGCUGCAAGGCCUUGCCCdTdT) and scrambled (SCR) negative control were from Qiagen. The specificity of SphK1 siRNA was verified by sequence comparison with the human genome database using the NIH Basic Local Alignment Search Tool (BLAST; http://blast.ncbi.nlm.nih.gov/). To down-regulate SphK1 expression, HUVECs were seeded in 6-well plates at a concentration of 2.5 × 10⁵ cells/well. After 1 d of incubation, they were transfected with 1 μM of SphK1 or SCR siRNA using OligofectAmine (Invitrogen, Carlsbad, CA, USA) as recommended by the manufacturer. After 48 h of transfection, some of the cells were harvested for the analysis of SphK1 expression, and the others were treated with vehicle, TNF-α (5 nM), or IL-1β (2.5 ng/ml).

Measurement of 6-keto PGF_1α and PGE₂ in cell culture

The culture supernatants of HUVECs were collected after they were incubated with vehicle, TNF-α (5 nM), or IL-1β (2.5 ng/ml), and celecoxib at 1 μM in growth medium for 8 h and assayed for 6-keto PGF_1α and PGE₂ production using EIA, as described above. Results are expressed as picograms per milliliter, calculated according to the 6-keto PGF_1α and PGE₂ standard curve. All assays were performed in triplicate.

Isolation of lung endothelial cells (ECs) from WT and SphK1-KO mice

Lung ECs were isolated from WT and SphK1-KO mice using Dynabeads (Invitrogen), as previously reported (30, 31). Briefly, lung was removed aseptically and digested in type I collagenase (2 mg/ml; Worthington, Lakewood, NJ, USA) at 37°C for 45 min with occasional agitation. Dynabeads precoated with CD31 and CD102 antibodies were used for EC purification.

Real-time RT-PCR

RNA was extracted from cell lysates using RNeasy mini kit (Qiagen), according to the manufacturer's instructions. cDNA was synthesized from these RNAs using M-MLV reverse transcriptase (Promega, Madison, WI, USA) and oligo d(T)_12–16. Real-time PCR was performed on an iCycler iQ multicolor real-time PCR detection system (Bio-Rad Laboratories, Hercules, CA, USA). The standard real-time PCR reaction volume was 25 μl, including 12.5 μl iQ SYBR Green Supermix (Bio-Rad), 1 μl forward primer, 1 μl reverse primer, and 1 μl cDNA. The following primers were used for real-time PCR amplification: human SphK1, forward 5′-AGGCTGAAATCTCCTTCACGC-3′ and reverse 5′-GTCTCCAGACATGACCACCAG-3′; COX-2, forward 5′-ATATGTTCTCCTGCCTACTGGAA-3′ and reverse 5′-GCCCTTCACGTTATTGCAGATG-3′; mPGES-1, forward 5′-CTGCTGGTCATCAAGATGTACG-3′ and reverse 5′-GGTTAGGACCCAGAAAGGAGT-3′; prostacyclin synthase (PGIS), forward 5′-CTGTTGGGCGATGCTACAGAA-3′ and reverse 5′-CCTCAATTCCGTAAAGAGTCAGG-3′; MKK6, forward 5′-TGCAGAGTTTGTTGACTTTACCT-3′ and reverse 5′-TGCCACATCTGTTCCTTTGGA-3′; Egr-1, forward 5′-ACCTGACCGCAGAGTCTTTTC-3′ and reverse 5′-GCCAGTATAGGTGATGGGGG-3′; and β-actin (32), forward 5′-TCCTCCCTGGAGAAGAGCTA-3′ and reverse 5′-CCAGACAGCACTGTGTTGGC-3′. Cycles (n=40) consisted of a 10-s melt at 9°C, followed by 1 min annealing/extension at 60°C. The final step was 1 min incubation at 72°C. All reactions were performed in triplicate. Threshold cycle (C_T) analysis for all samples was set at 0.15 relative fluorescence units. The data were analyzed using Q-Gene software (33) and expressed as mean normalized expression (MNE). MNE is directly proportional to the amount of RNA of the target gene (SphK1, COX-2, mPGES-1, PGIS, MKK6, or Egr-1) relative to the amount of RNA of the reference gene β-actin.

Statistical analysis

SBP was analyzed by 1-way ANOVA and Tukey multiple comparison tests among groups at 0, 5, 10, 15, 20, 25, 30, and 35 min. Production of PGE₂ and 6-keto PGF_1α was analyzed by 1-way ANOVA and Tukey multiple-comparison tests among siSCR-, siSphK1-, and celecoxib/siSCR-treated HUVECs. mRNA expression of COX-2, mPGES-1, and PGIS in HUVECs was analyzed by 1-way ANOVA and Tukey multiple-comparison tests among siSCR-, siSphK1-, and celecoxib/siSCR-treated HUVECs. mRNA expression of mPGES-1 and PGIS in lung ECs was analyzed with the unpaired Student's t test between WT and SphK1-KO mice. Differences were considered statistically significant at values of P < 0.05.

RESULTS

Differential roles of SphK1 and COX-2 in Ang II-induced acute hypertension in mice

We first analyzed the effects of the lack of SphK1 expression on Ang II-induced acute hypertension (Fig. 1 and Table 1) in mice. We monitored SBP for 35 min using a computerized noninvasive tail cuff system. All male animals, including C57BL/6 WT, SphK1-KO, and SphK2-KO mice at 8-12 wk of age were pretreated with celecoxib at 100 mg/kg BW or vehicle (polyethylene glycol 400) by oral gavage 1 h prior to SBP measurement. There was no significant difference in the SBP levels at baseline among the WT control (118.6±3.0 mmHg), SphK1-KO (108.5±3.9 mmHg), SphK2-KO (100.8±7.4 mmHg), and WT mice treated with celecoxib (97.9±4.0 mmHg). Ang II injection rapidly increased SBP in all groups of mice, as reported previously (Fig. 1, Table 1, and ref. 34). However, responses to Ang II injection were different among the groups. Interestingly, the results clearly indicated two distinct groups. Control WT mice (Fig. 1, blue) and SphK1-KO mice (Fig. 1, green) showed peak SBP at 4 min after Ang II injection (181 and 188 mmHg, respectively); thereafter, SBP gradually decreased toward basal levels at 30 min after Ang II injection. In contrast, celecoxib-treated WT mice (Fig. 1, red) and SphK2-KO mice (Fig. 1, orange) showed peak at 6 min 30 s at 192 and 197 mmHg, respectively, and the SBP remained high for the entire experiment (35 min). As expected, celecoxib-treated WT mice showed significantly higher SBP than control WT mice after 25 min until the end of the experiment. SphK2-KO mice showed the same responses to Ang II-induced hypertension as those in celecoxib-treated WT mice. SphK1-KO mice showed the same response to Ang II-induced hypertension as control WT mice, indicating that lack of SphK1 expression does not exacerbate Ang II-induced acute hypertension and suggesting that SphK1 is independent of COX-2 in ECs regarding hypertension. There were no histological abnormalities in the heart or any other tissues in SphK1 or SphK2-KO mice.

Figure 1.

Effects of SphK1 and SphK2 deletion and celecoxib administration on Ang II-induced acute hypertension. SphK1-KO mice show the same responses to Ang II-induced acute hypertension as WT mice, whereas celecoxib-treated mice and SphK2-KO mice show higher BP and retain higher BP until the end of the experiment than WT and SphK1-KO mice. Data are quantified in Table 1. *P < 0.05, **P < 0.01 for control vs. celecoxib; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 for SphK1-KO vs. celecoxib; ^†P < 0.05, ^††P < 0.01, ^†††P < 0.001 for control vs. SphK2-KO; ^‡P < 0.05, ^‡‡‡P < 0.001 for SphK1-KO vs. SphK2-KO.

Table 1.

Systolic BP at progressive time points after Ang II injection

Group	Systolic BP (mmHg)
	0 min	15 min	20 min	25 min	30 min	35 min
Control	118.6 ± 3.0	146.1 ± 4.9	138.7 ± 3.3	120.3 ± 5.1	117.3 ± 1.9	112.3 ± 3.1
Celecoxib	97.9 ± 4.0	155.5 ± 5.9	151.8 ± 4.6#	148.5 ± 3.2**,##	140.0 ± 3.1**,###	133.0 ± 4.0*
SphK1-KO	108.5 ± 3.9	138.4 ± 3.8	129.3 ± 6.6	124.0 ± 6.1	112.4 ± 3.9	116.0 ± 5.9
SphK2-KO	100.8 ± 7.4	165.0 ± 10.9‡	155.8 ± 8.1‡	143.8 ± 5.3†,‡	147.4 ± 8.0†††,‡‡‡	137.4 ± 7.3††,‡

Quantification of data from Fig. 1.

^*P < 0.05,

^**P < 0.01 vs. control;

^#P < 0.05,

^##P < 0.01,

^###P < 0.001 vs. SphK1-KO;

^†P < 0.05,

^††P < 0.01,

^†††P < 0.001 vs. control;

^‡P < 0.05,

^‡‡‡P < 0.001 vs. SphK1-KO.

Differential PG productions in kidney and aorta induced by Ang II

To determine the mechanisms by which lack of SphK1 expression showed different responses to Ang II-induced hypertension from celecoxib-treated WT mice, we first analyzed the levels of PGE₂ (Fig. 2A, B) and 6-keto PGF_1α (Fig. 2C, D), an index of PGI_2, in kidney and aorta collected from control WT, SphK1-KO, SphK2-KO, and celecoxib-treated WT mice at 5 min after Ang II injection. Levels of PGE₂ were up-regulated by Ang II injection in both kidney (P<0.01; Fig. 2A) and aorta (P<0.05; Fig. 2B). In celecoxib-treated WT mice and SphK1-KO mice, the levels of PGE₂ in kidney (P<0.005 and P<0.05, respectively; Fig. 2A) and aorta (P<0.01 and P<0.05, respectively; Fig. 2B) were significantly lower than those in control WT mice when treated with Ang II. Levels of PGE₂ in SphK1-KO mice were almost the same as celecoxib-treated WT mice, indicating that SphK1-KO mice block PGE₂ production induced by Ang II.

Figure 2.

Levels of PGE₂ (A, B) and 6-keto PGF_1α (C, D) in kidney (A, C) and aorta (B, D) in control WT, celecoxib-treated WT, SphK1-KO, and SphK2-KO mice. SphK1-KO mice and celecoxib-treated WT mice show reduced PGE₂ production in both kidney and aorta. SphK1-KO mice show the same levels of 6-keto PGF_1α as WT mice, while celecoxib-treated WT mice show reduced 6-keto PGF_1α production. *P < 0.05, **P < 0.01, ***P < 0.005.

The levels of 6-keto PGF_1α in kidney and aorta collected from control WT mice were significantly up-regulated by Ang II (P<0.01; Fig. 2C, D). Celecoxib treatment significantly reduced 6-keto PGF_1α production induced by Ang II in both kidney and aorta (P<0.005 and P<0.01, respectively; Fig. 2C, D). In SphK1-KO mice, by contrast, the levels of 6-keto PGF_1α in kidney and aorta were almost same as those in WT control mice induced by Ang II, indicating that lack of SphK1 expression does not inhibit 6-keto PGF_1α production. These results indicate that celecoxib treatment indeed reduced both PGE₂ and 6-keto PGF_1α induced by Ang II, whereas lack of SphK1 expression preferentially inhibited the production of PGE₂, suggesting that there are different mechanisms by which SphK1 inhibition inhibits PGE₂ production from COX-2 inhibition. In contrast, SphK2-KO mice showed the same levels of PGE₂ production as control WT (Fig. 2A, B) and significantly lower 6-keto PGF_1α production in both kidney and aorta (P<0.05 and P<0.005, respectively; Fig. 2C, D). These results indicate that a role of SphK2 on prostaglandin production in both kidney and aorta is different from SphK1.

We next analyzed S1P levels in blood collected from untreated control WT, Ang II-treated WT, SphK1-KO, and celecoxib-treated WT mice at 5 and 25 min after Ang II injection. Both Ang II and celecoxib did not alter S1P levels in blood in WT mice (Supplemental Fig. S1). As expected, S1P levels in blood in SphK1-KO mice were significantly lower than those in WT mice (P<0.05). These results suggest that S1P in blood does not play a key role on Ang II-induced acute hypertension.

SphK1 regulates PGE₂ production induced by cytokines via mPGES-1 expression in ECs

In order to determine the mechanisms by which lack of SphK1 expression inhibits PGE₂ production but not 6-keto PGF_1α production, we employed HUVECs to investigate the effects of SphK1 down-regulation on cytokine-induced PGE₂ and 6-keto PGF_1α production (Fig. 3). As expected, both celecoxib and SphK1 down-regulation significantly decreased PGE₂ production induced by TNF-α and IL-1β (Fig. 3A). We also confirmed that celecoxib significantly decreased production of 6-keto PGF_1α (Fig. 3B, solid bars). In contrast, SphK1 down-regulation significantly increased production of 6-keto PGF_1α (Fig. 3B, shaded columns). These results suggest that this in vitro HUVEC system represents our in vivo data model, where both SphK1 deficiency and celecoxib treatment inhibited Ang II-induced PGE₂ production and only celecoxib inhibited Ang II-induced 6-keto PGF_1α production. Next, in order to clarify which enzymes were responsible for this discrepancy between COX-2 inhibition and SphK1 deficiency on these prostanoid productions, we analyzed the mRNA expression of COX-2, mPGES-1, and PGIS in HUVECs using quantitative real-time PCR. As expected of a direct inhibitor, celcoxib did not inhibit the expression of COX-2. Interestingly, SphK1 down-regulation did not alter COX-2 expression in HUVECs (Fig. 3C). SphK1 down-regulation, however, significantly attenuated mPGES-1 expression (Fig. 3D) and significantly up-regulated PGIS expression (Fig. 3E) induced by TNF-α and IL-1β. We also confirmed that celecoxib did not alter expression of these genes (Fig. 3C–E). These data suggest that SphK1 down-regulation reduces PGE₂ production induced by cytokines through selective inhibition of mPGES-1 expression, but not COX-2, in HUVECs.

Figure 3.

Effects of SphK1 down-regulation by siRNA and celecoxib treatment on production of PGE₂ (A) and 6-keto PGF_1α (B), and expression of COX-2 (C), mPGES-1 (D), and PGIS (E) induced by TNF-α and IL-1β in HUVECs. Celecoxib treatment reduces both PGE₂ and 6-keto PGF_1α induced by cytokines in HUVECs, while SphK1 down-regulation only reduces PGE₂, not 6-keto PGF_1α. SphK1 down-regulation inhibits mPGES-1, not PGIS expression levels, while celecoxib treatment inhibits PGIS expression levels, not mPGES-1 expression. Celecoxib was incubated at 1 μM with vehicle control, TNF-α at 5 nM or IL-1β at 2.5 ng/ml. Error bars = se. *P < 0.05, **P < 0.01, ***P < 0.005 vs. siSCR.

To confirm and extend the regulation of mPGES-1 expression by SphK1, we isolated lung EC from WT and SphK1-KO mice and analyzed the mRNA expression levels of mPGES-1 and PGIS using quantitative real-time PCR (Fig. 4). mPGES-1 expression in SphK1-KO mice was significantly lower than in WT mice (Fig. 4A), whereas levels of PGIS expression in SphK1-KO mice were significantly greater than in WT mice (Fig. 4B). Taken together, these data suggest that SphK1 regulates PGE₂ production induced by cytokines via mPGES-1 expression in ECs (see Fig. 6A).

Figure 4.

mRNA levels of mPGES-1 and PGIS expression in lung ECs isolated from WT and SphK1-KO mice. SphK1-KO mice show significantly lower (A; mPGES-1) and higher (B; PGIS) expression levels than WT mice (P<0.05). Data are expressed as MNE, normalized with β-actin as the reference gene. Error bars = se. *P < 0.05 vs. WT mice.

Figure 6.

Proposed signaling schemes for SphK1 down-regulation in endothelial cells. The mechanism by which SphK1 down-regulation decreases PGE₂ production, but not PGI₂, in HUVECs (A), and mechanism by which SphK1 regulate mPGES-1 expression (B).

SphK1 regulates mPGES-1 expression via MKK6 and Egr-1 in p38 MAPK pathway in HUVECs

To assess the mechanism by which SphK1 regulates mPGES-1 expression, we first employed the MAPK signaling pathway PCR array to analyze gene expression profiles in SphK1 down-regulated HUVECs treated with IL-1β. SphK1 down-regulation using siRNA markedly reduced the expression of Egr-1, a transcription factor that is related to mPGES-1 expression (Supplemental Table S1A). Although Egr-1 is regulated by both p38 MAPK and ERK1/2 pathways, SphK1 down-regulation decreased only MKK6 expression, which phosphorylates and activates p38 MAPK but did not affect the Ras/Raf-1/ERK1/2 pathway (Supplemental Table S1B, C). On the basis of these results, we further analyzed levels of mRNA expression of MKK6 and Egr-1 with more samples using quantitative real-time PCR. The results are summarized in Fig. 5. These mRNA expression levels in SCR negative control siRNA-treated HUVECs were increased by cytokines as well as mPGES-1; however, SphK1 down-regulation by siRNA inhibited the increase in these mRNA expression levels induced by cytokines (Fig. 5A, B). The suppression of MKK6 and Egr-1 expression by SphK1 down-regulation correlates with the reduced expression of mPGES-1 (Figs. 3D and 5), suggesting that SphK1 regulates mPGES-1 expression via MKK6 and Egr-1 in p38 MAPK pathway in HUVECs (Fig. 6B).

Figure 5.

Effect of SphK1 down-regulation on the levels of MKK6 and Egr-1 mRNA expression in HUVECs. Cytokines induce MKK6 (A) and Egr-1 (B) expression and SphK1 down-regulation significantly inhibits both MKK6 (A) and Egr-1 (B) expression in HUVECs. Data are expressed as MNE, normalized with β-actin as the reference gene. Error bars = se. *P < 0.05, **P < 0.01, and ***P < 0.005 vs. siSCR.

DISCUSSION

This study shows that SphK1-KO mice do not worsen Ang II-induced acute hypertension as compared to WT control mice. Lack of SphK1 expression inhibits only PGE₂ production, but not 6-keto PGF_1α production, whereas celecoxib treatment significantly inhibits both PGE₂ and 6-keto PGF_1α production induced by Ang II in kidney and aorta. We also show that down-regulation of SphK1 reduces only PGE₂ production via a decrease in mPGES-1 expression induced by cytokines, while celecoxib reduces both PGE₂ and PGI₂ production through an inhibition of COX-2 in HUVECs.

The major finding in this study is the role of SphK1 in acute hypertension. We first showed that Ang II injection into control WT mice causes a prompt increase in SBP, followed by a gradual decline. The mechanism of this fall in SBP may be related to the synthesis of compensatory vasodilators, especially vasodepressor prostanoid PGI₂ (35–39). The sustained elevation of SBP observed in celecoxib-treated WT mice may be due to the reduction of PGI₂ (40). It has been reported that both celecoxib and rofecoxib reduced urinary PGI₂ levels in healthy volunteers (41, 42), COX-2 inhibition by SC58236 or gene disruption in mice completely blocked the Ang II-induced increase in renal medullary 6-keto-PGF_1α (40), and celecoxib inhibited cytokine-induced 6-keto-PGF_1α production in HUVECs (43). Similarly, our results show that celecoxib blocked the Ang II-induced increase in both of PGE₂ and 6-keto-PGF_1α in kidney and aorta. On the other hand, SphK1-KO mice show the same response to Ang II-induced hypertension as control WT mice. Although there were concerns that SphK1 inhibition may abolish COX-2 signaling, interestingly, lack of SphK1 expression blocked only PGE₂ production (not 6-keto PGF_1α production). The results suggest that SphK1 mediates a different pathway from COX-2/PGE₂ pathway in ECs. In addition, we found that both Ang II and celecoxib did not affect the levels of S1P in blood. Taken together, our results suggest that SphK1 inhibition does not affect BP, unlike COX-2 inhibitors.

Notably, we found that SphK1 regulates PGE₂ production through mPGES-1 expression, not COX-2 expression, in HUVECs. Several studies demonstrated that SphK1 down-regulation by siRNA reduces COX-2 expression and PGE₂ production in several cell lines, including colon cancer cells (13–16). Therefore, we investigated whether SphK1 down-regulation reduces 6-keto PGF_1α production in HUVECs, as well as celecoxib. Interestingly, in contrast to celecoxib and consistent with the in vivo results, down-regulation of SphK1 significantly inhibited PGE₂ production, but not 6-keto PGF_1α production, induced by cytokines. We also show that SphK1 down-regulation does not affect COX-2 expression and significantly suppressed elevation of mPGES-1 expression. These observations suggest that SphK1 regulates mPGES-1 in HUVECs, whereas SphK1 regulates COX-2 in colon cancer cells. In addition, SphK1 down-regulation significantly inhibited the decline of PGIS expression induced by cytokines and enhanced PGI₂ production. This may be caused by accumulation of PGH₂ accrued by inhibition of mPGES-1 expression in response to SphK1 down-regulation. Furthermore, lung ECs isolated from WT and SphK1-KO mice showed consistent results with HUVECs. These results suggest that SphK1 regulates only PGE₂ production through regulating mPGES-1 expression, but not COX-2, in ECs.

Our results suggest that SphK1 deficiency mediates the induction of mPGES-1 and does not exacerbate Ang II-induced hypertension because of sustained PGI₂ production. However, the previous studies employing mPGES-1-KO mice have reported conflicting results: i.e., the Coffman group (44) and the FitzGerald group (45–47)reported that mPGES-1-KO mice showed the same level of BP as compared to WT mice in untreated, high-salt feeding, and Ang II-infused mice, whereas the Yang group (48–51) demonstrated that mPGES-1 deficiency exaggerated high-salt feeding and Ang II infusion-induced hypertension. Although Facemire et al. (52) recently reported that these conflicting results might be due to the genetic background of mice, further studies are required to clarify this point.

Finally, we showed that SphK1 regulates mPGES-1 expression via MKK6 and Egr-1 in p38 MAPK pathway in HUVECs. We first employed the MAPK signaling pathway PCR array to analyze gene expression profiling. We found that SphK1 down-regulation markedly reduced the expression of Egr-1 and MKK6. Although Egr-1 is regulated by both p38 MAPK and ERK1/2 pathway, SphK1 down-regulation reduced only MKK6 expression, which phosphorylates and activates p38 MAPK. To confirm these results, we further analyzed the expression of MKK6 and Egr-1 in HUVECs. SphK1 down-regulation inhibited the increase in these mRNA expression levels. Thus, these findings suggest that SphK1 regulates mPGES-1 expression via MKK6 and Egr-1 in p38 MAPK pathway in HUVECs (Fig. 6B).

In summary, the present study shows that SphK1 deficiency ameliorates Ang II-induced acute hypertension, and these mice fared better than WT mice treated with celecoxib. In addition, we demonstrate that SphK1 down-regulation reduced PGE₂ production via a decrease in mPGES-1 expression, whereas celecoxib inhibited both PGE₂ and 6-keto PGF_1α production through selective COX-2 inhibition. Moreover, SphK1 down-regulation decreased mPGES-1 expression via MKK6 and Egr-1 in p38 MAPK pathway. Taken together, these results suggest that SphK1 may be a potential target for cancer chemoprevention because of its safer role in cardiovascular functions.