Effect of Myriocin on Plasma Sphingolipid Metabolism and Atherosclerosis in apoE-deficient Mice*

Sphingolipids play a very important role in cell membrane formation, signal transduction, and plasma lipoprotein metabolism, all of which may well have an impact on the development of atherosclerosis. To investigate the relationship between sphingolipid metabolism and atherosclerosis, we utilized myriocin to inhibit mouse serine palmitoyl-CoA transferase (SPT), the key enzyme for sphingolipid biosynthesis. We injected 8-week-old apoE-deficient mice with myriocin (0.3 mg/kg/every other day, intraperitoneal) for 60 days. On a chow diet, myriocin treatment caused a significant decrease (50%) in liver SPT activity (p < 0.001), significant decreases in plasma sphingomyelin, ceramide, and sphingosine-1-phosphate levels (54, 32, and 73%, respectively) (p < 0.0001), and a significant increase in plasma phosphatidylcholine levels (91%) (p < 0.0001). Plasma total cholesterol and triglyceride levels demonstrated no significant changes, but there was a significant decrease in atherosclerotic lesion area (42% in root and 36% in en face assays) (p < 0.01). On a high fat diet, myriocin treatment caused marked decreases in plasma sphingomyelin, ceramide, and sphingosine-1-phosphate levels (59, 66, and 81%, respectively) (p < 0.0001), and a marked increase in plasma phosphatidylcholine levels (100%) (p < 0.0001). Total cholesterol and triglyceride demonstrated no significant changes, but there was a significant decrease in atherosclerotic lesion area (39% in root and 37% in en face assays) (p < 0.01). These results indicate that, apart from cholesterol levels, sphingolipids have an effect on atherosclerotic development and that SPT has proatherogenic properties. Thus, inhibition of SPT activity could be an alternative treatment for atherosclerosis.

Sphingolipids play a very important role in cell membrane formation, signal transduction, and plasma lipoprotein metabolism, all of which may well have an impact on the development of atherosclerosis. To investigate the relationship between sphingolipid metabolism and atherosclerosis, we utilized myriocin to inhibit mouse serine palmitoyl-CoA transferase (SPT), the key enzyme for sphingolipid biosynthesis. We injected 8-week-old apoE-deficient mice with myriocin (0.3 mg/ kg/every other day, intraperitoneal) for 60 days. On a chow diet, myriocin treatment caused a significant decrease (50%) in liver SPT activity (p < 0.001), significant decreases in plasma sphingomyelin, ceramide, and sphingosine-1-phosphate levels (54, 32, and 73%, respectively) (p < 0.0001), and a significant increase in plasma phosphatidylcholine levels (91%) (p < 0.0001). Plasma total cholesterol and triglyceride levels demonstrated no significant changes, but there was a significant decrease in atherosclerotic lesion area (42% in root and 36% in en face assays) (p < 0.01). On a high fat diet, myriocin treatment caused marked decreases in plasma sphingomyelin, ceramide, and sphingosine-1-phosphate levels (59, 66, and 81%, respectively) (p < 0.0001), and a marked increase in plasma phosphatidylcholine levels (100%) (p < 0.0001). Total cholesterol and triglyceride demonstrated no significant changes, but there was a significant decrease in atherosclerotic lesion area (39% in root and 37% in en face assays) (p < 0.01). These results indicate that, apart from cholesterol levels, sphingolipids have an effect on atherosclerotic development and that SPT has proatherogenic properties. Thus, inhibition of SPT activity could be an alternative treatment for atherosclerosis.
Sphingolipids have many biological functions, including cell membrane formation, signal transduction, and lipid metabolism, and all of these may be related to the development of atherosclerosis. Serine palmitoyl-CoA transferase (SPT) 1 is the rate-limiting enzyme in the biosynthesis of sphingolipids (1). It has long been known that SPT plays an important role in the metabolism of sphingolipids, but its role in other lipid metabolisms and atherosclerosis has not been unequivocally determined. When SPT activity is increased in rat liver (2) and lung (3), sphingolipid formation is likewise increased. The activity of SPT is heightened in the aortas of rabbits fed a high cholesterol diet (4).
Two candidate cDNAs for yeast SPT, termed LCB1 and LCB2, have been cloned (5,6), and the translated sequences indicate that their gene products have a 21% amino acid sequence identity (6). The lack of SPT activity in a yeast strain defective in LCB1 or LCB2, together with the protein similarity data, suggest that the two genes encode subunits of SPT (6). Mouse and human LCB1 and LCB2 cDNA homologues have also been cloned (7,8). In mouse, the two mRNAs have the same tissue distribution (lung, kidney Ͼ brain Ͼ cartilage, skin Ͼ heart Ͼ liver Ͼ muscle), and the ratio of the amounts of the two transcripts remains approximately constant in all tissues (8). The tissue distribution of LCB2 mRNA parallels the distribution of SPT activity (9).
Isaria sinclairii is a fungus traditionally used in Chinese medicine in an effort to attain eternal youth (10). From it, a specific SPT inhibitor named myriocin has been isolated (10) and proven to have a molecular structure similar to that of sphingosine (11). Using myriocin-based affinity chromatography, two proteins, LCB1 and LCB2, can be purified from an interleukin-2-dependent mouse cytotoxic T cell line (CTLL-2) (12). This result indicates that LCB1 and LCB2 are myriocinbinding proteins and confirms the fact that they are responsible for SPT activity (12).
In this study, we utilized myriocin to investigate the impact of SPT inhibition on lipid metabolism and atherosclerosis development in the apoE-deficient mouse, a well known atherosclerosis animal model. We found that myriocin administration caused a decrease in plasma sphingomyelin (SM), ceramide (Cer), sphingosine (Sph), and sphingosine-1-phosphate (S1P) levels, caused an increase in plasma phosphatidylcholine (PC) levels, and caused a decrease in atherosclerotic lesions in apoE knock-out (apoE KO) mice on both chow and high fat, high cholesterol diets.

EXPERIMENTAL PROCEDURES
Animals and Myriocin Treatment-Eight-week-old apoE KO mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Myriocin (0.3 mg/kg) (Biomol Research Laboratories Inc.) or phosphatebuffered saline was injected intraperitoneally every other day for 8 weeks. The animals were on Purina Rodent Chow (catalog number 5001) or a high fat, high cholesterol diet (20% milk fat and 0.15% cholesterol; Harlan Teklad, Madison, WI).
Lipid and Lipoprotein Measurements-Fasting plasma was collected for fast protein liquid chromatography (FPLC) separation and lipid * This work was supported by NIH HL-64735 and HL-69817. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. measurement. Total cholesterol, phospholipids and triglyceride in plasma, and lipoproteins were assayed by enzymatic methods (Wako Pure Chemical Industries Ltd., Osaka, Japan). Plasma sphingomyelin was measured as described previously (13). PC concentration was obtained by subtracting SM from total phospholipid concentration. Apolipoprotein analysis using SDS-PAGE was also done as described previously (14).
Sphingolipid Analysis by Mass Spectrometry-Plasma sphingosine bases, sphingoid base-1-phosphates, and ceramide species were performed on a Thermo Finnigan TSQ 7000 triple quadrupole mass spectrometer operating in a multiple reaction monitoring, positive ionization mode at the Department of Biochemistry and Molecular Biology, Medical University of South Carolina, on a fee-for-service basis.
Briefly, 250 l of mouse plasma was fortified with the internal standards (IC 17 base D-erythro-sphingosine (17C-Sph), C 17 sphingosine-1-phosphate (17C-S1P), N-palmitoyl-D-erythro-C 13 sphingosine (13C-Cer), and heptadecanoyl-D-erythro-sphingosine (C17-Cer)) and extracted with ethyl acetate/iso-propanol/water (60:30:10) (v/v) solvent system. After evaporation and reconstitution in 100 l of methanol, samples were injected onto the Surveyor/TSQ 7000 liquid chromatography/mass spectrometry system, and gradient was eluted from a BDS Hypersil C8, 150 ϫ 3.2-mm, 3-m particle size column with a 1 mM methanolic ammonium formate, 2 mM aqueous ammonium formate mobile phase system. Peaks corresponding to the target analytes and internal standards were collected and processed using the Xcalibur software system. Quantitative analysis was based on the calibration curves generated by spiking an artificial matrix with the known amounts of the target analyte synthetic standards and an equal amount of the internal standards. The target analyte/internal standard peak area ratios were plotted against analyte concentration. The target analyte/internal standard peak area ratios from the samples were similarly normalized to their respective internal standards and compared with the calibration curves using a linear regression model.
Atherosclerosis-At the end of the myriocin treatment period, the mice were sacrificed, and the hearts and proximal aortas as well as the whole aortas were removed, dissected, and photographed. An aorta root   assay and an en face assay were performed as described previously (15,16). Statistical Analysis-Differences between groups were tested by Student's t test. Data are presented as mean Ϯ S.D. A p value of Ͻ0.05 was considered significant.

RESULTS
Two groups of 8-week-old apoE KO mice were utilized. Group 1 (n ϭ 7) and group 2 (n ϭ 7) animals were injected with 100 l of myriocin (0.3 mg/kg) or phosphate-buffered saline, respectively, every other day for 8 weeks. As expected, myriocintreated mice had 50% less SPT activity in the liver than the controls.
As shown in Table I, plasma SM levels were significantly decreased (54%) (p Ͻ 0.001) and plasma PC levels were significantly increased (91%) (p Ͻ 0.0001) after myriocin administration, whereas total cholesterol and triglyceride levels were not significantly changed. It should be emphasized that the PC/SM ratio was dramatically increased (317%) (p Ͻ 0.0001) in the myriocin-treated group as compared with control, indicating that lipoprotein composition was changed.
To investigate the lipid distribution among the lipoproteins with or without myriocin treatment, we utilized FPLC to fractionize lipoproteins and measured SM, PL, and cholesterol in each fraction. We found that myriocin significantly decreased SM and increased PC levels but had no significant effect on cholesterol (Fig. 1). SDS-PAGE revealed that there were no significant changes of the levels of apolipoproteins, including apoB100, apoB48, and apoA-I (data not shown).
To investigate whether myriocin treatment has any impact on other sphingolipid levels, including Cer, Sph, and S1P, mass spectrometry was utilized. We found that after myriocin treatment Cer, Sph, and S1P were significantly decreased (Table II), indicating that myriocin treatment not only influences plasma SM levels but also those of Cer, Sph, and S1P, three important second messengers in signal transduction. The following two findings are also worth noting. 1) The major ceramides in apoE KO mouse plasma are Cer24:0, Cer24:1, Cer18:0, and C16:0 (Table II). 2) The S1P and Sph concentrations in apoE KO mice are ϳ200 nM (Table II).
For further evaluation of the myriocin effect on plasma lipid levels, 2-month-old mice were challenged with a high fat, high cholesterol (Western type) diet for 8 weeks with or without   myriocin treatment. As shown in Table III, plasma SM levels were dramatically decreased (59%), whereas plasma PC levels and the PC/SM ratio were dramatically increased (100% and 380%, respectively) (p Ͻ 0.0001) after myriocin administration. Total cholesterol and triglyceride levels were not significantly changed, with FPLC administration producing the same results (Fig. 2). Again, SDS-PAGE revealed that there were no significant changes of the levels of apolipoproteins, including apoB100, apoB48, and apoA-I (data not shown). We also measured other sphingolipid levels, finding that Cer, Sph, and S1P were dramatically decreased after myriocin treatment (Table  IV). Basically, a profound myriocin effect was observed when a high fat, high cholesterol diet was used. It is reported that myriocin treatment (1 mg/kg but not 0.3 mg/kg) reduces T-lymphocyte populations in mice (17). We utilized FAS to evaluate myriocin effect on T cell counts in the   FIG. 3. Myriocin treatment dramatically reduced atherosclerosis in apoE KO aorta. A, mice were euthanized and the aortae dissected and photographed. This set of pictures is representative of seven sets. B and C, quantification of atherosclerotic lesions in the proximal aorta by root assay in mice fed a chow diet (B) or a high fat, high cholesterol diet (C). The procedure for root assay was described previously (15). D and E, quantification of atherosclerotic lesions in whole aorta by en face analysis in mice fed a chow diet (D) or a high fat, high cholesterol diet (E). The procedure for en face analysis was described previously (16). Values are mean Ϯ S.D. *, p Ͻ 0.001; n ϭ 7. circulation and did not found any difference (data not shown).
To evaluate the effect of myriocin on atherogenesis, we dissected mouse aortas and photographed them. We also measured proximal and whole aortic lesion areas. After 2 months of myriocin administration on a chow diet, we found a reduction of lesions in the aortas (Fig. 3A). We also found a 42% (p Ͻ 0.01) reduction in mean proximal aortic lesion areas and a 36% (p Ͻ 0.01) reduction in mean whole aortic lesion areas compared with controls (Fig. 3, B and C). After 2 months of myriocin administration on a Western type diet, we also found a reduction of lesions in the aortas (Fig. 3A). There was a 39% (p Ͻ 0.01) reduction in mean proximal aortic lesion areas and a 37% (p Ͻ 0.01) reduction in mean whole aortic lesion areas compared with controls (Fig. 3, D and E).
These results indicate that myriocin possesses important anti-atherosclerotic properties.

DISCUSSION
In this study we demonstrated for the first time that intraperitoneal myriocin administration in apoE KO mice caused the following: 1) dramatic decreases in plasma SM, Cer, S1P, and Sph levels; 2) dramatic increases in plasma PC levels, thus increasing the PC/SM ratio; and 3) significant decreases in atherosclerotic lesions.
There are two methods of myriocin delivery in vivo, intraperitoneal injection and oral administration. Because the latter was shown to inflict serious gastrointestinal toxicity (18) and may have had an impact on cholesterol absorption during the high fat, high cholesterol loading experiment, we chose the former, as have other investigators (19,20). Indeed, intraperitoneal injection of myriocin did not change mouse plasma cholesterol levels on the chow or high fat diets (Tables I and III). In a most recent report, Park et al. showed that oral myriocin administration causes significant reduction of plassma cholesterol and SM levels, thus causing a dramatic reduction of atherosclerotic lesions in apoE KO mice on a high cholesterol diet (21). The different outcome of that study and ours, in terms of plasma cholesterol levels, might be due to the different methods of myriocin delivery.
There was a profound induction of plasma PC levels after myriocin treatment (Tables I and III). This result was consistent with a previous report indicating that administration of L-cycloserine, another inhibitor of SPT, stimulated CTP:choline-phosphate cytidylyltransferase (CT; a key enzyme for PC biosynthesis) activity by 74% (22). This effect might have been due to the decrease of Sph (Tables II and IV), a specific inhibitor of CT activity (23).
There is some question as to why myriocin treatment caused fewer atherosclerotic lesions in apoE-deficient mice. The decrease of SM and the increase of PC contents in non-HDL particles might be one of the mechanisms. Substantial evidence now supports the role of lipoprotein SM and arterial SMase in atherogenesis. SM carried into the arterial wall on atherogenic lipoproteins is acted on by an arterial wall SMase, leading to an increase in Cer content and promoting lipoprotein aggregation (24). LDL extracted from human atherosclerotic lesions is highly enriched in SM as compared with plasma LDL (25,26). Moreover, a significant fraction of LDL extracted from fresh human lesions is aggregated and has a high content of Cer, indicating that the LDL has been modified by SMase, resulting in aggregation (24). The absolute and relative concentrations of plasma SM are both increased in atherosclerosis-susceptible animal models (26 -28). In vitro manipulation has shown that the relative SM concentration is an important determinant of susceptibility to SMase-induced aggregation (24,26,29). We have previously shown in a case-control study that plasma SM levels are an independent risk factor for coronary heart disease (13) and confirmed the result in another larger and more homogenous case-control trial. 2 The decrease of plasma Cer levels might be another mechanism for the reduction of atherosclerosis in apoE KO mice after myriocin treatment. However, this hypothesis seems to controvert existing reports. Cer is a well known second messenger involving apoptosis (30). Typically, strategies that elevate cellular Cer are used for therapies aimed at arresting growth or promoting apoptosis. Charles et al. found that Cer analogs, applied directly to damaged arteries, can be strongly antiproliferative (31). Proliferation of cultured vascular smooth muscle cells appears to involve the extracellular signal-regulated kinase (ERK) and AKT kinase cascades and to be inhibited by Cer (32).
In vivo, C 6 -Cer-coated balloon catheters prevent stretch-induced neointimal hyperplasia in rabbit carotid arteries (31) by inactivating ERK and AKT signaling and thereby inducing cell cycle arrest in stretch-injured vascular smooth muscle cells (31). Based on published reports, we should expect more atherosclerotic lesions in myriocin-treated apoE KO mice than in controls, but we found the opposite (Fig. 3). The impact of plasma Cer levels on atherosclerosis deserves further investigation.
The decrease of plasma S1P levels might be another mechanism for the reduction of atherosclerosis in apoE KO mice after myriocin treatment. In human plasma, 65% of S1P is associated with lipoproteins, where HDL is the major carrier (33). There is some debate as to whether plasma or serum S1P is an atherogenic or anti-atherogenic mediator. On the one hand, the S1P in HDL has been shown to bind to S1P/Edg receptors on human endothelial cells and, for this reason, probably mediates many of the anti-inflammatory actions of HDL on endothelial cells (34). On the other hand, serum S1P was found to be a remarkably strong predictor of both the occurrence and the severity of coronary stenosis in a recent case-control study (35). It should be noted that the S1P concentration in apoE KO mice is Ͼ200 nM (Table II), and the amount needed to activate S1P receptors on endothelial cells is ϳ100 nM (34,36). Thus, the reduction of S1P to Ͻ100 nM by myriocin treatment (Table II) might have pathological relevance to atherosclerosis development in the mouse model. The impact of plasma S1P levels on atherosclerosis also deserves further investigation.
It is reported that myriocin treatment (1 mg/kg but not 0.3 mg/kg) reduces T-lymphocyte populations in mice (37). We utilized phycoerythrin-labeled anti-CD3 antibodies and flow cytometry to evaluate the effect of myriocin on T cell counts in the circulation, and we did not find any difference (data not shown), confirming that 0.3 mg/kg myriocin administration has no effect on T cell populations (37).
In summary, SPT inhibition mediated by myriocin dramatically decreases plasma SM, Cer, S1P, and Sph levels and has anti-atherogenic properties. Because the treatment has no or little effect on cholesterol metabolism, the inhibition of SPT activity could well be an important alternative treatment for atherosclerosis.