OUP user menu

A sphingosine kinase 1 mutation sensitizes the myocardium to ischemia/reperfusion injury

Zhu-Qiu Jin, Jianqing Zhang, Yong Huang, Holly E. Hoover, Donald A. Vessey, Joel S. Karliner
DOI: http://dx.doi.org/10.1016/j.cardiores.2007.05.029 41-50 First published online: 1 February 2008

Abstract

Objective Sphingosine kinase (SphK) is a key enzyme in the synthesis of sphingosine 1-phosphate (S1P), a bioactive sphingolipid. SphK is involved in ischemic preconditioning (IPC). To date no studies in genetically altered animals have examined the role of SphK1 in myocardial ischemia/reperfusion (IR) injury and IPC.

Methods and results Wild-type and SphK1 null mouse hearts were subjected to IR (50 min global ischemia and 40 min reperfusion) in a Langendorff apparatus. IPC consisted of 2 min of global ischemia and 2 min of reperfusion for two cycles. At baseline, there were no differences in left ventricular developed pressure (LVDP), ±dP/dtmax, and LV end-diastolic pressure (EDP) between SphK1 mutant and wild-type (WT) mouse hearts. In the mutants, total SphK enzyme activity was reduced by 44% and S1P levels were decreased by 41%. SphK1 null hearts subjected to IR exhibited more cardiac damage compared with WT: LVDP and ±dP/dtmax decreased, LVEDP increased, and infarct size increased (n=6, P<0.05). Apoptosis was markedly enhanced in SphK1 mutant IR mouse hearts. IPC was cardioprotective in WT hearts, but this protection appeared to be ineffective in SphK1 null hearts. There was no change in infarct size in the IPC+IR group compared to the IR group in the null hearts (50.1±5.0% vs 45.0±3.8%, n=6, P=NS). IPC remained ineffective in the null hearts even when the index ischemia time was shortened by 10 min.

Conclusions Deletion of the SphK1 gene sensitizes the myocardium to IR injury and appears to impair the protective effect of IPC. These data provide the first genetic evidence that the SphK1-S1P pathway is a critical mediator of IPC and cell survival.

Keywords
  • Ischemia
  • Myocardial infarction
  • Sphingosine kinase
  • Sphingosine 1-phosphate
  • Cardioprotection
  • Signal transduction

Time for primary review 23 days

This article is referred to in the Editorial by Nishino, Webb and Marber (pages 3–4) in this issue.

1 Introduction

Myocardial injury during reperfusion after ischemia results from a complex cascade of events that involves free radical generation, calcium overload and cytokine activation [1]. Recent studies have indicated that sphingolipid metabolites also contribute to myocardial ischemia/reperfusion (IR) injury [2]. Thus, activation of neutral sphingomyelinase and generation of ceramide and sphingosine occur early during IR injury [3]. It has been suggested that ceramide, which is increased in myocardial IR injury [4], and sphingosine mediate the actions of cytokines such as TNFα and interleukin-1 on the heart [5,6]. Sphingosine-1-phosphate (S1P), a product of sphingosine kinase (SphK) activation, is an important intracellular signaling molecule that regulates diverse cellular events, including survival, growth, motility, differentiation, cytoskeletal reorganization, and calcium mobilization [7]. S1P is anti-apoptotic and the ratio of S1P to ceramide has been proposed as a major determinant of cell survival [8]. S1P is a ligand for G-protein coupled S1P receptors [9]. It is cardioprotective in cultured cardiac myocytes [10,11] and in isolated hearts via a PKCϵ-independent pathway [12,13].

Ischemic preconditioning (IPC) is a short period of IR that rescues hearts from subsequent long term IR injury. Reported triggers are adenosine, bradykinin, opioids, and α1-agonists coupled to G-proteins, mainly Gi [14]. Among subsequent mediators are activated PKCϵ and KATP channel opening followed by activation of prosurvival pathways [15]. Recent studies have shown that sphingolipids are involved in IPC. Thus, IR injury increased ceramide content, whereas IPC had the opposite effect [16]. In contrast, S1P content was enhanced after IPC [16]. We have recently reported that SphK mediates an alternative or parallel pathway of myocardial IPC [17]. Increased S1P formation was observed after IPC while dimethylsphingosine, a selective SphK inhibitor, abolished IPC-induced cardioprotection [17].

SphK is the key enzyme responsible for the formation of S1P [7] and exhibits two isoforms in mouse heart [18,19]. SphK1 appears to be protective, as its overexpression increases intracellular S1P content and promotes cell growth and survival [20,21]. Increased cell viability induced by SphK1 and subsequent S1P synthesis results from activation of a prosurvival pathway that includes PI-3K/Akt and increased bcl-2 expression followed by reduced cytochrome C release and caspase activation [11,12,20,22]. In contrast, SphK2 is considered to be pro-apoptotic (23). To date, no studies of cardioprotection have been performed in a model in which any non-receptor member of the sphingolipid pathway has been genetically altered. Thus, the primary objective of the present study was to utilize mouse hearts harboring an inactivated SphK1 gene to definitively determine the role of SphK1 in myocardial IR injury and in cardioprotection conferred by IPC.

2 Materials and methods

This study was conducted in accordance with the Guide for the Care and Use of Laboratory Animals (National Academic Press, Washington DC, 1996). All procedures were approved by the Animal Care Subcommittee of the San Francisco VA Medical Center.

2.1 SphK1 null mice

SphK1 null (KO) mice in which exons 3–6 of the SphK1 gene had been deleted were obtained from Drs. Shaun Coughlin and Rajita Pappu (Cardiovascular Research Institute, University of California, San Francisco). These mice along with their wild-type littermates were used for all studies reported herein. Male homozygous null (SphK1−/−) and wild-type (WT) mice were generated by breeding heterozygous (SphK1 +/−) mice. Genotyping using PCR to confirm the absence of exons 3–6 of SphK1 DNA was routinely performed on tail biopsies of 3–4-week-old mice. Fig. 1 shows a typical PCR analysis.

Fig. 1

PCR analysis showing amplification of SphK1 from wild-type (+/+), heterozygous (+/−), and null (−/−) mice. A=primer for identification of the wild-type gene; B=primer for the detection of the null genotype.

2.2 Langendorff isolated perfused heart preparation

Wild-type and SphK1 null mice (2–3 months old, weighing 28–32 g) were heparinized (500 U/kg, IP) and anesthetized with sodium pentobarbital (60 mg/kg, IP). Hearts were rapidly excised, washed in ice-cold arresting solution (NaCl 120 mmol/L, KCl 30 mmol/L), and cannulated via the aorta on a 20 gauge stainless steel blunt needle. Hearts were perfused at 70 mmHg on a modified Langendorff apparatus using Krebs–Henseleit solution at 37 °C as previously described in our laboratory [12,17]. Platinum electrodes connected to a stimulus generator (Grass Instruments, West Warwick, RI) were used to pace hearts at 360 bpm.

2.3 Ischemia–reperfusion (IR) and ischemic preconditioning (IPC) protocols

For IR experiments, the protocol consisted of a 20 min equilibration period, followed by 50 min of global ischemia and 40 min of reperfusion. In some studies, the index ischemia time was reduced to 40 min. For IPC, hearts were equilibrated for 16 min and then subjected to two short cycles of IR, each consisting of 2 min of global ischemia and 2 min of reperfusion, followed immediately by prolonged IR as described above. Hemodynamics were recorded as previously described [12,17].

2.4 Infarct size measurement

After IR±IPC, a subset of wild-type and SphK1 null hearts was infused with 15 ml 1% triphenyltetrazolium chloride (Sigma) in phosphate-buffed saline at a rate of 1.5 ml/min as previously described [24]. Hearts were then removed from the cannula, weighed, and fixed overnight in 10% formalin. Hearts were removed from formalin and stored at −20 °C until sectioning for analysis of LV infarct size as previously described in our laboratory [12,17]. The infarct size of each section was expressed as a fraction of the area at risk defined as the total area of the left ventricle in this global ischemia model.

2.5 Creatine kinase (CK) determination

CK measurements were performed as previously described in our laboratory (12). CK reagent was from Stanbio Laboratory, Boerne, TX.

2.6 DNA fragmentation (DNA ladder)

IR hearts were placed in liquid nitrogen and stored at −80 °C. Hearts were minced on ice in digestion buffer (NaCl 100 mM, Tris (pH 8.0) 10 mM, EDTA 25 mM, SDS 0.5%) and proteinase K (400 μg/ml) added. After overnight incubation at 55 °C, 25 μl of 10 mg/ml DNA-free RNase was added (0.7 ml volume) and incubated for 30 min at 37 °C. 1/20 volume of 5 mol/L NaCl was added. DNA was extracted in a 1: 1 volume ratio of phenol: chloroform: isoamyl alcohol (25: 24: 1) 3 times, and precipitated in isopropranol. DNA was centrifuged and washed with 80% ethanol. Twenty μg of DNA was loaded onto a 1.8% agarose gel, and DNA ladder formation was detected under UV light.

2.7 Terminal deoxynucleotidyl transferase-mediated dUTP-biotin in situ nick-end labeling (TUNEL)

The TUNEL assay was performed using In Situ Cell Death Detection Kit, Fluorescein (Roche, Mannheim, Germany) according to the manufacturer's instructions. Randomly selected cardiomyocytes from 4 sections/heart were evaluated with fluorescence microscopy to determine the number and percentage of cells exhibiting apoptosis. Nuclei were stained blue with Hoechst 33258 (Molecular Probes, Eugene, OR). Actin was stained red with fluorescent phalloidin (Molecular Probes, Eugene, OR). Apoptotic nuceli stained green. For each section 10 fields were randomly chosen and counted. The proportion of apoptotic nuclei was determined from a total of 40 fields/heart.

2.8 Sphingosine kinase activity

Mice were anesthetized by intraperitoneal injection of pentobarbital sodium (60 mg/ml). Hearts and spleens were excised and washed in cold buffer A (0.13 M KCl, 20 mM Hepes, pH 7.4, 1 mM EGTA, 1 μg/L leupeptin, 0.25 μg/L each of chymostatin and pepstatin A). Tissue was minced, homogenized and centrifuged for 6 min at 350 ×g to remove cell debris and nuclei. The supernatant was centrifuged for 50 min at 100,000 ×g, decanted and designated as the cytosolic fraction.

The assay uses a chloroform/methanol/aqueous trisodium EDTA extraction system to separate reactant ([3H]sphingosine) from product ([3H]S1P) as previously described in our laboratory [25]. A standard assay contains Triton X-100 (0.05%), 250 mM KCl, 1 μM [3H]sphingosine (300–400 cpm/pmol), 5 mM ATP, 10 mM Mg2+, 100 mM Tris, pH 8.0 (30 °C), and enzyme protein in a volume of 0.5 ml. A 0.1 ml aliquot is removed into 0.3 ml of methanol and 0.6 ml of chloroform. The mixture is separated into two phases by the addition of 0.3 ml of trisodium EDTA (pH 9) followed by vortexing and centrifugation. The upper aqueous phase containing the S1P is removed for liquid scintillation counting. SphK activity is assayed by conversion of [3H]-sphingosine to [3H]-S1P and is reported as pmol/min/g tissue.

2.9 Sphingosine-1-phosphate measurement

Hearts were removed from the Langendorff apparatus, plunged into cold perfusion buffer and cut into small pieces. After chilling, they were rapidly blotted dry and weighed. Minced heart was then homogenized in 1 ml ice-cold chloroform: methanol (1:2) and incubated at −20 °C overnight. Samples were then centrifuged at 15,000 rpm for 10 min, the supernatant decanted and 0.95 ml CHCl3 and 0.95 ml 0.1 N NH4OH added. The mixture was vortexed and centrifuged at 15,000 rpm. The supernatant was removed and 0.1 μg S1P was added to each tube. Samples were lyophilized and resuspended in 0.3 ml of 70% acetonitrile for analysis by combined liquid chromatography/dual tandem mass spectrometry. The system was a Micromass Quatro Ultima equipped with an electrospray source, Shimadzu LC-10 AD pumps and Waters Intelligent Sample Processor 717 Plus. The LC column was BDS C18 (4.6×50 mm, 5 μM particle size, Keystone). Conditions for liquid chromatography and mass spectrometry have been described elsewhere [26].

2.10 Western Blot analysis

SphK protein was measured using standard SDS-PAGE Western blotting as previously described in our laboratory (17). Primary antibodies (Cell Signaling Technology, Inc. Beverly, MA and Santa Cruz Biotechnology, Inc, Santa Cruz, CA) were used to measure Akt phosphorylation (Ser 473), total Akt, cytochrome C, and SphK2. Immunoreactive bands were detected by enhanced chemiluminescence (ECL) (Amersham Bioscience, Piscataway, NJ) and quantitated by densitometric analysis of digitized autoradiograms with NIH Image 1.61 software.

2.11 Statistics

Data are mean±SEM. The significance of the differences in mean values for hemodynamics, infarct size, and CK release between groups was evaluated by one-way ANOVA, followed by post-hoc testing (Newman–Keuls). Differences in SphK activity and S1P levels between control and mutant tissues were evaluated by Student's t test. P<0.05 was considered significant.

3 Results

3.1 Baseline characteristics

There were no baseline differences in heart weight, heart weight/body weight ratio, and left ventricular end-diastolic pressure (LVEDP) between WT and KO mice. As shown in Table 1, the left ventricular developed pressure (LVDP), +dP/dtmax, and −dP/dtmax were slightly higher in KO mouse hearts, but these differences did not reach statistical significance.

View this table:
Table 1

Baseline cardiac functional parameters in SphK-null and wild-type littermate (WT) mice

WT-IRWT-IPCKO-IRKO-IPC
Body weight (g)31.1±0.333.6±2.831.2±0.633.1±0.9
Heart weight (mg)176.8±7.3166.4±14.7156.2±13.9161.0±5.8
Heart/body wt (mg/g)5.7±0.24.6±0.45.1±0.44.8±0.1
LVDP (mmHg)120.1±7.8117.6±3.8129.1±4.8137.1±7.8
LVEDP (mmHg)3.3±1.72.9±0.41.9±0.33.1±0.9
±dP/dt (mmHg/s)3031±2993172±2793373±2203638±317
−dP/dt (mmHg/s)2670±2572319±622758±2902975±306
CF (ml/min)3.0±0.33.2±0.43.7±0.43.3±0.1
  • WT: wild-type; KO: knockout; IR: ischemia/reperfusion; IPC: ischemic preconditioning followed by IR. Data are expressed as mean±SEM of at least 5 animals per group. There are no statistically significant differences in any of the above parameters between groups as assessed by one-way ANOVA. Hemodynamic changes after IR and IPC followed by IR are shown in Fig. 2.

3.2 Deletion of SphK1 impairs cardiac function and myocardial survival during IR injury and ischemic preconditioning

Both WT and KO hearts were subjected to 50 min of global ischemia and 40 min of reperfusion. As shown in Fig. 2A, recovery of LVDP at the end of IR was significantly lower in KO mouse hearts than in WT hearts. +dP/dtmax and −dP/dtmax also were lower in KO hearts (Fig. 2B and C). Moreover, LVEDP was significantly higher in KO hearts than in WT hearts (Fig. 2D). Thus, IR caused more serious impairment of both cardiac systolic and diastolic function in KO hearts compared to the dysfunction observed in hearts from WT mice.

Fig. 2

A. Panel A, left ventricular developed pressure (LVDP), its maximum increase (Panel B) and decrease (Panel C) of velocity (*P<0.05 vs WT-IR. #P<0.05 vs WT-IPC. IR: ischemia/reperfusion; IPC: ischemic preconditioning followed by ischemic preconditioning. n=5–7/group. Panel E. Representative cross-sections of wild-type (WT) and sphingosine kinase1 (SphK1) knockout (KO) mouse hearts after perfusion with 2, 3, 5-triphenyltetrazolium chloride. Visual inspection identifies larger infarcts in WT-IR, KO-IR, and KO-IPC hearts compared to WT-IPC. Abbreviations as in Panel A. Infarct size (Panel F) expressed as percentage of risk area and creatine kinase (CK) release (Panel G) in wild-type (WT) and sphingosine kinase1 (SphK1) knockout (KO) mouse hearts subjected to 50 min of global myocardial ischemia and 40 min of reperfusion. Data are expressed as mean±s.e.m. *P<0.05 vs WT-IR. #P<0.05 vs WT-IPC. IR: ischemia/reperfusion; IPC: ischemic preconditioning followed by IR. n=6–7/group.

As shown in Fig. 2A–D, ischemic preconditioning (IPC) increased cardiac performance in WT hearts. At the end of IR, WT-IPC hearts exhibited improved LVDP and +dP/dtmax and −dP/dtmax compared with the WT control group. In WT hearts LVEDP was less after IPC+IR compared with IR alone, but this difference was not significant (Fig. 2D).

A prominent hemodynamic finding was that the cardioprotective effect of IPC is apparently ineffective in KO hearts. The recovery of LVDP at the end of IR was not different between KO-IR hearts and KO-IPC hearts (Fig. 2A). LVEDP in KO hearts was markedly increased (Fig. 2D). There were no differences in coronary flow between WT and KO hearts subjected to IR or IR+IPC (data not shown).

As shown in Fig. 2E and F, infarct size, expressed as percentage of risk area, was significantly larger in KO hearts than in WT hearts. CK release at the end of IR also was significantly higher in KO hearts than in WT hearts (Fig. 2G).

IPC reduced CK release in WT hearts compared with WT-IR group. IPC also decreased infarct size in WT hearts. Both of these cardioprotective effects of IPC were significant (Fig. 2E–G). In striking contrast, the cardioprotection induced by IPC was not evident in SphK1 null hearts. There was no difference in CK release at the end of IR between KO-IPC hearts and KO-IR hearts (Fig. 2G). Infarct size was unchanged in KO-IPC hearts compared with KO-IR hearts (Fig. 2F).

When the index ischemia time was reduced to 40 min, infarct size in the KO hearts declined to the level seen in the WT-IR group (28.5±2.7% of the risk area). At this reduced level of injury, IPC was still ineffective in producing cardioprotection in the KO hearts (infarct size=29.3±1.6%, P=NS vs IR alone, n=4/group).

3.3 SphK1 deletion sensitizes cardiac myocytes to apoptosis induced by IR injury

At baseline, myocardial tissue from sham WT and SphK1 null mouse hearts exhibited virtually no TUNEL-positive staining (Fig. 3A and B). To determine if there are differences in the rate and extent of apoptosis between WT and KO hearts, we extended the reperfusion time to 120 min. This maneuver induced the appearance of TUNEL-positive nuclei which were much more prevalent in SphK1 null IR hearts compared with WT hearts (Fig. 3A and B). Similarly, in both WT and KO hearts no DNA laddering was present at baseline. Consistent with the induction of TUNEL-positivity, prolonged reperfusion induced formation of nucleosome ladders SphK1 null IR hearts, compared with WT-IR hearts (Fig. 3C).

Fig. 3

Panel A. Representative photomicrographs of in situ detection of DNA fragments in SphK1 knockout (KO) and wild-type (WT) mouse hearts. Panel B: Percent TUNEL-positive nuclei. *P<0.05 vs all other groups, n=3/per group. Panel C. Electrophoretic analysis of internuclosomal DNA extracted from two SphK1 knockout hearts (lanes 1 and 2) and two separate wild-type (lanes 3 and 4) mouse hearts exposed to 50 min of ischemia and 120 min of reperfusion. Lane 0 is a DNA size marker. One additional set of hearts showed identical results.

3.4 Exogenous S1P is cardioprotective in SphK1 null mouse hearts

A 2 min infusion of 10 nM S1P given as pretreatment [12,17] protected SphK1 null mouse hearts against IR injury. As shown in Fig. 4, S1P significantly improved LVDP recovery and reduced infarct size, consistent with previous results from our laboratory in WT hearts [2,17].

Fig. 4

Sphingosine-1-phosphate improved left ventricular developed pressure (LVDP) (A) and reduced infarction size (B) in SphK1 mutant mouse hearts subjected to ischemia/reperfusion injury. See Methods and Results for details. *P<0.05 vs Control. n=4/group. RA=risk area.

3.5 Deletion of SphK1 depresses SphK activity and S1P content

SphK activity was detected in both cytosolic and membrane fractions in WT hearts, with the majority of enzyme activity detected in the cytosol. We also measured SphK activity in spleen of WT mice. As shown in Fig. 5A and B, SphK activity was higher in spleen vs heart tissue. In the SphK1 null hearts, SphK activity in heart tissue was decreased in the cytosol by 44%. It was also decreased in the membrane fraction. SphK activity was also reduced by 75% in the spleen of mutant mice. After IR, SphK activity in the WT hearts was decreased from 256±8 pmol/min/g tissue in the normal control hearts to 144±39 pmol/min/g tissue in the IR injured hearts (n=4, P<0.05). SphK activity in KO hearts remained the same (140±17 pmol/min/g tissue vs 126±51 pmol/min/g tissue, n=4, P=NS).

Fig. 5

Sphingosine kinase (SphK) activity and sphingosine-1-phosphate (S1P) content in wild-type (WT) and sphingosine kinase1 (SphK1) knockout (KO) mice. (A). SphK activity in the normal control WT and KO hearts. (B). SphK activity in the normal control WT and KO spleens. (C). S1P content in WT and KO hearts. Data are expressed as mean±SEM; n=3–7/group.

As shown in Fig. 5C, S1P content detected by mass spectrometry was significantly decreased in the KO mouse hearts.

3.6 Effect of SphK1 deletion on cardiac SphK2 protein expression

Since SphK activity was still detected in SphK1 null mouse hearts, we wondered whether residual SphK activity was the result of an increase in SphK2 protein. As shown in Fig. 6, SphK2 expression was significantly higher in SphK1 mutant mouse hearts than in WT hearts. To test the hypothesis that the increase in SphK2 in the KO hearts was responsible for impaired hemodynamics and increased infarct size after IR injury, we used dimethysphingosine (DMS), a known SphK inhibitor. DMS (3 μM) was perfused for 5 min before IR injury in both WT and KO mouse hearts. We found that there was no difference between WT and KO groups with regard to infarct size (39.0±4.5%% in the DMS-WT group vs 38.3±5.4% in the DMS-KO group, n=3/group) and LVDP recovery (8.1±2.0% in the DMS-WT group vs 8.0±5.7% in the DMS-KO group). This result suggests that raised SphK2 in the SphK1 KO mouse hearts is not associated with alterations in infarct size or hemodynamic recovery after IR injury.

Fig. 6

Above: Representative western blots of sphingosine kinase (SphK) type 2 protein expression in wild-type (WT) and sphingosine kinase1 (SphK1) knockout (KO) mouse hearts. Below: bar graphs quantitating the data from WT and KO hearts, n=4/group. Data are expressed as mean±s.e.m. *P<0.05 vs WT. P=positive control.

3.7 Phosphorylation of Akt (Ser 473) and cytochrome C release after IPC and IR injury in wild-type and SphK1 null mouse hearts

Phosphorylation of the prosurvival signaling molecule Akt (p-Akt, Ser 473) was measured in WT and KO hearts during IPC and IR. At baseline, p-Akt was increased in KO hearts (Fig. 7A). After IPC, p-Akt was enhanced in WT hearts. No alteration of p-Akt was observed in KO hearts (Fig. 7B). After IR injury, cytochrome C release into the cytosol was increased by over 4-fold in KO hearts as determined by western blotting and subsequent densitometry from 12.8±2.6 to 57.8±7.6 arbitrary units (n=4, P<0.05).

Fig. 7

(A). Representative immunoblots showing phosphorylation of Akt (p-Akt, Ser 473) in wild-type (WT) and SphK1-knockout (KO) mouse hearts. (B). p-Akt and total Akt expression in whole tissue homogenates of normal control (NC) WT and KO hearts and hearts subjected to ischemic preconditioning (IPC). N=4/group. *P<0.05 vs WT-NC.

4 Discussion

In these experiments, we sought additional evidence for the role of SphK in cardioprotection by testing the hypothesis that genetic elimination of SphK1 alters the cardiac response to IR injury and IPC. We used a newly available mouse model in which deletion of exons 3–6 of the SphK1 gene leads to reduction of total SphK activity and a concurrent decrease in serum and tissue S1P levels. These mice exhibit no evident phenotype, breed normally, have normal vascular development, and live a normal lifespan. We found no significant difference in baseline cardiac function between wild-type and SphK1 null mouse hearts. However, after 50 min of global ischemia and 40 min of reperfusion, left ventricular developed pressure and ±dP/dtmax were significantly decreased in KO hearts. Infarct size, CK release, and cytochrome C content in the cytosol were significantly increased in the KO hearts after IR injury. Strikingly, IPC-induced cardiac protection was apparently ineffective in the KO hearts. Extensive apoptosis was present in the KO hearts following prolongation of reperfusion injury. When the extent of IR injury in the KO hearts was reduced by a shorter index ischemia time, IPC was still ineffective. These results provide the first evidence in a genetically modified animal that SphK1 is an important lipid kinase mediating cell survival and that SphK1 appears to be required for IPC in the heart. We also recognize that these data, although highly suggestive, do not establish an absolute causal relation between the SphK1 gene mutation and impaired IPC.

Recently, Allende et al. reported that mice deficient in SphK1 were still rendered lymphopenic by S1P1 receptor agonist FTY720 [27]. Like us, they found mutant mice to be fertile, long-lived, and also reported no histologic abnormalities in major organs. Although they performed no functional cardiac studies, Allende et al. reported a greater than 60% reduction in serum S1P levels, but surprisingly in contrast to our observations, no significant decrease in tissue levels, including brain, heart, kidney, liver, spleen and testis. A possible explanation for this apparent discrepancy could be related to tissue sample preparation and methods of lipid extraction. However, substantial reductions in SphK1 enzyme activity were observed in all of these organs, suggesting a compensatory increase in SphK2 activity (see below). Of note, in their SphK1 null mice, Allende et al. reported no changes in the mRNA levels for the genes encoding enzymes known to regulate S1P levels, including S1P lyase, S1P phosphatase, or ceramidase [27].

Recent reports have further elucidated the role and mechanisms of SphK1 in maintaining cell viability. In MCF-7 breast cancer cells, the DNA-damaging agent actinomycin D downregulated SphK1 protein and activity and reduced the percentage of viable cells, an effect reversed by S1P treatment [28]. These observations suggested that decreased SphK1 may be necessary for induction of cell death [28]. Knockdown of SphK1 by siRNA caused cell cycle arrest and induced apoptosis via effector caspase activation, cytochrome C release and Bax oligomerization in the mitochondrial membrane [28]. The considerable residual SphK activity that both we and Allende et al. [27] found is likely attributable to SphK2. Thus, we noted a substantial increase in SphK2 protein in the SphK1 null mice. SphK2 is a BH3-only protein [23], and has been implicated mitochondria-mediated apoptotic pathways [29]. Maceyka et al. also reported that targeting SphK1 to the endoplasmic reticulum converted it from anti-apoptotic to pro-apoptotic [29]. They concluded that the location of the S1P generated by SphK isoforms is critical to cell function [29]. It should be emphasized that these studies were done in vitro in cell lines, and may not apply in vivo.

Our recent work has revealed additional possibilities. In acutely ischemic hearts, SphK activity declines markedly and remains depressed during recovery, while in hearts that have been preconditioned, recovery of enzyme activity is much more robust [30]. S1P levels are altered in parallel [30], and there may be a threshold concentration below which prosurvival pathways are either not activated or are suppressed. In isolated adult cardiac myocytes we have found that internally generated S1P must be exported to act in an autocrine or paracrine manner, and that this process is impaired during simulated ischemia [31]. Thus many factors, which require further study, may contribute to the inability of SphK2 to promote cardiac myocyte survival during IR injury [32].

Acute Akt activation is cardioprotective both in vitro and in vivo [33,34]. In contrast, chronic activation is deleterious [35]. Enhanced chronic Akt phosphorylation is found in the hearts of patients with advanced heart failure [36] and restoration of PI3K rescues the deleterious effects of chronic Akt activation in the heart during IR injury [37]. In this study, we found that the increased phosphorylation of Akt occurred in SphK1 null mouse hearts at baseline. This was associated with depressed cardiac function, increased infarct size, and enhanced cytochrome C release in SphK1 KO mouse hearts after IR injury.

Despite an abundance of residual SphK activity and serum and tissue levels of S1P that would seem to be adequate to sustain viability, especially after IPC, cardioprotection was not evident in the SphK1 null mice even when the index ischemia time was shortened. Such a result might be predicted, presumably because all of the residual SphK activity can be attributed to SphK2, which generates S1P that does not transactivate S1P receptors [29]. Our observation (Fig. 4) that exogenously supplied S1P improves hemodynamics and reduces infarct size, even in the SphK1 null mouse, is consistent with this hypothesis. Of note is that specific inhibition of SphK1 by siRNA or the compound SKI results in a proportional decrease of S1P and a concomitant increase of ceramide in cardiac myoblasts that leads to apoptosis [38]. In isolated adult mouse cardiac myocytes subjected to hypoxic stress, we have previously shown that S1P can trigger a prosurvival pathway involving the S1P1 receptor, Gi, and phosphorylation of Akt, Bad, and GSK-3β, resulting in reduced mitochondrial cytochrome C release [11]. Moreover, experiments in isolated adult mouse cardiac myocytes indicate that VPC 23019, the selective S1P1 receptor antagonist, abolished the protective effect induced by S1P and by a SphK activator, the monoganglioside GM-1 [31].

As noted in the Results and Fig. 5, we measured SphK activity in WT and SphK1 knockout Langendorff hearts immediately prior to and after IR. SphK activity was reduced in the KO hearts and was unchanged by IR injury. In contrast this intervention reduced SphK activity by an average of 44% in the WT hearts. In addition, we have reported elsewhere that SphK activity declined 61% during ischemia and did not recover upon reperfusion [30]. Preconditioning reduced the decrease in SphK activity during ischemia by half and upon reperfusion activity returned to normal [30]. Measurements of S1P content followed a similar pattern [30].

In summary, our data are the first to show in an ex vivo whole organ cardiac preparation that myocardial damage is enhanced after IR injury and that the cardioprotective intervention of preconditioning appears to be impaired by deletion of the SphK1 gene. These observations are consistent with accumulating evidence, derived heretofore exclusively by the in vitro study of cell lines, that SphK1 and SphK2 orchestrate opposing functions that regulate cell fate. In addition, our data using this genetically modified model provide the strongest evidence to date for the critical role of SphK, specifically SphK1 activation, in ischemic preconditioning.

Acknowledgements

This work was supported by NIH grant 1P01 HL068738-01A1 (JSK).

The authors thank Drs. Shaun Coughlin and Rajita Pappu for supplying the SphK1 null mice and Mr. Michael Kelley for technical assistance.

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
View Abstract