Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

Objectives

Sphingolipids have been demonstrated to be involved in many human diseases. However, measurement of sphingolipids, especially of sphingosine 1-phosphate (S1P) and dihydro-sphingosine 1-phosphate (dhS1P), in blood samples requires strict sampling, since blood cells easily secrete these substances during sampling and storage, making it difficult to introduce measurement of sphingolipids in clinical laboratory medicine. On the other hand, cerebrospinal fluid (CSF) contains few blood cells. Therefore, we attempted to establish a system based on liquid chromatography-tandem mass spectrometry (LC-MS/MS) for the measurement of sphingolipids in the CSF, and applied it for the diagnosis of carcinomatous meningitis.

Methods

We developed and validated a LC-MS/MS-based measurement system for S1P and dhS1P and for ceramides and sphingosines, used this system to measure the levels of these sphingolipids in the CSF collected from the subjects with cancerous meningitis, and compared the levels with those in normal routine CSF samples.

Results

Both the measurement systems for S1P/dhS1P and for ceramides/sphingosines provided precision with the coefficient of variation below 20% for sphingolipids in the CSF samples. We also confirmed that the levels of S1P, as well as ceramides/sphingosines, in the CSF samples did not increase after the sampling. In the CSF samples collected from patients with cancerous meningitis, we observed that the ratio of S1P to ceramides/sphingosine and that of dhS1P to dihydro-sphingosine were higher than those in control samples.

Conclusions

We established and validated a measurement system for sphingolipids in the CSF. The system offers promise for being introduced into clinical laboratory testing.

sphingosine 1-phosphate, ceramides, cerebrospinal fluid, carcinomatous meningitis

Introduction

A series of elegant basic studies have illustrated that sphingolipids are involved in the pathogenesis of many human diseases, such as cancer (1, 2,), immunological disorders (3,), cardiovascular disorders (4–6,), and neurological disorders (7, 8,). Among the sphingolipids, sphingosine 1-phosphate (S1P), dihydro-sphingosine 1-phosphate (dhS1P), and ceramides, in particular, are receiving attention as pharmacological targets for the treatment of human diseases due to their potent physiological properties. Therefore, it is an important task to confirm the involvement of these sphingolipids in human diseases, using human samples. However, as with most other bioactive lipids, it is difficult to measure the exact levels of blood sphingolipid Blood cells, such as platelets and erythrocytes, easily release the same substances during sampling and storage, and the levels of S1P and dhS1P can increase during storage, prior to centrifugation. Therefore, we analyzed the samples at 4°C during storage and centrifugation (9).

Impact statement

This study will help both clinical and basic researchers to investigate the involvement of sphingolipids, especially sphingosine 1-phosphate, in human diseases. In the present study using the cerebrospinal fluids collected from patients with cancerous meningitis, we determined that the S1P-ceramides rheostat hypothesis, which has been proposed mainly from the results of basic studies, is actually involved in the pathogenesis of human cancers. This report will also provide other clinical laboratories a guide of the precise measurement of sphingolipids in cerebrospinal fluids using LC-MS/MS.

In contrast to blood samples, cerebrospinal fluid (CSF) contains few blood cells in the healthy state, and even in pathological states, where the blood cells migrate into, or are released into the CSF by inflammation, infection, hemorrhage, or cancerous meningitis, much smaller numbers of blood cells are found in the CSF than in blood samples. We recently reported that lysophosphatidic acid, which is a bioactive lysophospholipid similar to S1P, can be more precisely measured in CSF samples than in blood samples (10). The levels of lysophosphatidic acid in blood samples increased after incubation at room temperature, whereas their levels in the CSF remained quite stable.

Considering this background, we hypothesized that a measurement system for sphingolipids in the CSF could be introduced into clinical laboratory testing, although since the levels of sphingolipids, especially of S1P and dhS1P, in the CSF are low, a measurement system with a high sensitivity is necessary. In the present study, we developed and validated a novel measurement system for sphingolipids based on liquid chromatography-tandem mass spectrometry (LC-MS/MS), which allowed precise measurement of the low concentrations of sphingolipids present in the CSF. Then, using this method, we measured the levels of S1P, dhS1P, ceramides, and sphingosines in cerebrospinal fluid samples collected from patients with carcinomatous meningitis, since S1P and ceramides are reported to be involved in the development of human cancers (11).

Materials and Methods

Subjects and Samples

Blood and urine samples were collected from the antecubital veins of healthy adult volunteers who were not receiving any medication, after obtaining their informed consent. CSF samples were obtained from the residual specimens from other laboratory examinations of specimens collected by lumbar puncture from patients diagnosed as having carcinomatous meningitis; the causes of the carcinomatous meningitis are listed in Table 1. As the control group, residual CSF samples after other laboratory examinations of CSF specimens collected from subjects who were not diagnosed as having any malignant disorders were used; no abnormalities were detected in these samples on routine CSF testing, as shown in Table 1. The collected samples were stored at −80°C until the measurements, and were handled on ice during the experiments except for the incubation experiments. This study was conducted with the approval of the Institutional Research Ethics Committee of the Faculty of Medicine, the University of Tokyo (11158 and 2602).

Table 1
Open in new tab

Characteristics of the subjects.

Carcinomatous meningitis−+P value
n2721
Age55.5 ± 19.759.5 ± 15.80.447
Gender
 male1480.393
 female1313
Cell count [µL]0 [0–2]13 [0–1183]<0.01
CSF-TP [mg/dL]31.3 ± 6.674.9 ± 68.1<0.01
Neoplasms
 Leukemia3
 Lymphoma11
 Gastric cancer3
 Lung cancer2
 Peritoneal cancer1
 Extra bile duct cancer1
Carcinomatous meningitis−+P value
n2721
Age55.5 ± 19.759.5 ± 15.80.447
Gender
 male1480.393
 female1313
Cell count [µL]0 [0–2]13 [0–1183]<0.01
CSF-TP [mg/dL]31.3 ± 6.674.9 ± 68.1<0.01
Neoplasms
 Leukemia3
 Lymphoma11
 Gastric cancer3
 Lung cancer2
 Peritoneal cancer1
 Extra bile duct cancer1
Table 1
Open in new tab

Characteristics of the subjects.

Carcinomatous meningitis−+P value
n2721
Age55.5 ± 19.759.5 ± 15.80.447
Gender
 male1480.393
 female1313
Cell count [µL]0 [0–2]13 [0–1183]<0.01
CSF-TP [mg/dL]31.3 ± 6.674.9 ± 68.1<0.01
Neoplasms
 Leukemia3
 Lymphoma11
 Gastric cancer3
 Lung cancer2
 Peritoneal cancer1
 Extra bile duct cancer1
Carcinomatous meningitis−+P value
n2721
Age55.5 ± 19.759.5 ± 15.80.447
Gender
 male1480.393
 female1313
Cell count [µL]0 [0–2]13 [0–1183]<0.01
CSF-TP [mg/dL]31.3 ± 6.674.9 ± 68.1<0.01
Neoplasms
 Leukemia3
 Lymphoma11
 Gastric cancer3
 Lung cancer2
 Peritoneal cancer1
 Extra bile duct cancer1

Sample Preparation

For the measurement of S1P and dhS1P, the plasma samples were diluted 20-fold with PBS, whereas the CSF samples were not diluted. Then, 10 μL of diluted plasma samples or CSF samples were mixed with 10 μL of C17S1P and C17dhS1P, internal standards, at 1 ng/mL (final concentration) and the S1P contents were extracted with 180 µL of 0.1% formic acid in methanol (Wako Pure Chemical Industries). For ceramides and sphingosines, 10 µL of plasma samples and CSF samples were each diluted by the addition of 10 μL of C17:1 Sphingosine (Sph) (Avanti Polar Lipids), C17:1 dihydrosphingosine (dhSph) (Avanti Polar Lipids), and d18:1/17:0 Ceramide (Avanti Polar Lipids) at 1 ng/mL (final concentration) as internal standards, were mixed with 180 µL of methanol acidified with 0.1% formic acid. The mixtures were sonicated for 3 min and then centrifuged at 16,400 g for 10 min at 4°C. The supernatants were then analyzed by the LC-MS/MS method

Measurement of S1P and dhS1P

S1P and dhS1P were measured using the LC8060 LC/MS-MS system, consisting of a quantum ultra-triple quadrupole mass spectrometer (SHIMAZU). Briefly, 3.0 µL samples were injected and the LC separation was performed using a reverse-phase column (InertSustainSwift C8 PEEK column: 150 mmL × 2.1 mm I.D., 3 μm; GL sciences) with a gradient elution of solvent A (0.3% formic acid) and solvent B (0.3% formic acid, 95% acetonitrile) at 400 µL/min. The conditions were as follows: An isocratic run was performed at 20% solvent B for 1 min, followed by an isocratic run at 80% solvent B for 2 min, 100% solvent B for 2 min, and 20% solvent B for 5 min. The mass spectrometer was operated in electrospray ionization-positive ion mode and the analytical conditions were as follows: the nebulizer gas flow was set at 3.0 L/min, the drying gas flow at 8.0 L/min, the heating gas flow at 8.0 L/min, the interface temperature at 250°C, and the heat block temperature at 350°C. The analyses were performed in multiple reaction monitoring (MRM) mode in the positive ion mode for S1P, dhS1P, C17S1P, and C17dhS1P. The MRM settings are described in Supplemental Table 1, A. The data were analyzed by the Lab Solution software (SHIMADZU), using standard curves.

Measurement of Ceramides and Sphingosines

A measurement system for ceramides and sphingosines was constructed referring to our previous publication (12), with some modification. Six ceramide species (Cer d18:1/16:0, Cer d18:1/18:0, Cer d18:1/18:1, Cer d18:1/20:0, Cer d18:1/22:0, Cer d18:1/24:0), Sph, and dhSph were also measured with LC8060 LC/MS-MS system consisting of a quantum ultra-triple quadrupole mass spectrometer. Briefly, 1.0 µL of samples were injected and LC separation was performed using the InertSustain Swift C8 PEEK column (150 mm, 2.1 mm I.D., 3 µm particle size; GL science, Japan) with a gradient elution of solvent A (0.3% formic acid) and solvent B (0.3% formic acid, 95% acetonitrile) at 400 µL/min. The conditions were as follows: after a 0.5 min isocratic run, the proportion of solvent B was increased over 3 min from 15% to 100%, followed by a run for 4 min at 100% solvent B and equilibration at 15% solvent B for the remaining 5 min. The mass spectrometer was operated in electrospray ionization-positive ion mode and the analytical conditions were as follows; the nebulizer gas flow was set at 3.0 L/min, the drying gas flow at 8.0 L/min, the heating gas flow at 8.0 L/min, the interface temperature at 150°C, and the heat block temperature at 250°C. The analyses were performed in MRM mode in positive ion mode for Cer d18:1/16:0, Cer d18:1/18:0, Cer d18:1/18:1, Cer d18:1/20:0, Cer d18:1/22:0, Cer d18:1/24:0, Sph, dhSph, C17:1 Sphingosine, C17:1 dihydrosphingosine, and Cer d18:1/17:0. The MRM settings are described in Supplemental Table 1, B. The data were analyzed by the Lab Solution software, using standard curves.

Method Validation

Five concentrations of standard mixtures of these sphingolipids (0.01, 0.1, 0.5, 1.0, 10 ng/mL) were prepared and evaluated for the linearity of the measurements. The lower limit of quantification was defined as the lowest concentration that was measurable with a signal-to-noise (S/N) ratio of >10. We evaluated the intra- and inter-day precisions by conducting quintuple (5 times) measurements of the plasma and CSF samples or internal standards and calculating the coefficient of variation (CV, %). We evaluated the performance of the measurement system, based on the criteria described in bioanalytical method validation guidance for industry, published from the Food and Drug Administration in 2018, which required the CVs below 15% for the precision, except 20% at lower limit of quantification.

Investigation for the Influences of Sample Incubation Duration after Sampling on the Levels of the Sphingolipids

CSF samples were incubated at 4°C or at room temperature for 0, 1, 3, 6 h and the samples were collected for measurement of the sphingolipids as described above.

Investigation for the Matrix Effects of Plasma, CSF, and Urine

The matrix effects for internal standards were examined by mixing 10 μL of internal standards at 1.0 ng/mL (final concentration) and 10 μL of PBS, plasma, CSF, or urine samples. The matrix effects for endogenous sphingolipids were examined by subtracting the peak area of intrinsic S1P, dhS1P, sphingosines, or ceramides from that of the samples to which standard S1P, dhS1P, sphingosines, or ceramides were added at 1.0 ng/mL. The peak areas of the internal standards were compared.

Investigation for the Influences of the Contamination with Blood Components on the Levels of the Sphingolipids

We mixed a CSF sample of low levels of innate sphingolipids with PBS, serum, or whole blood at 9:1 and measured the levels of sphingolipids.

Flow Injection Tests

We mixed 100 μL of standard mixtures of S1P and dhS1P or those of ceramides and sphingosines at 100 ng/mL with 100 μL of the extracts of pooled CSF samples prepared as described above or methanol. Then we injected the samples into the mass spectrometer without separation by the column. We monitored the mass spectrum by scan mode or SIM mode.

Recovery Test

We added 10 μL of the standard solutions of S1P and dhS1P or those of sphingosines and ceramides at the concentration of 1.0 ng/mL (final concentration) to 10 μL of PBS or CSF samples and measured the concentrations of sphingolipids. Then we calculated the recovery rates by subtracting the concentrations of sphingolipids in the samples alone from those in the samples added with standard solutions.

Investigation for the Influences of Protein Levels or Cell Counts in the CSF Samples on the Measurement of Sphingolipids

We investigated the recovery rates in the CSF samples with various protein levels and cell counts as described above.

Statistical Analysis

The results were expressed as means ± SD. A one-way analysis of variance (ANOVA) test followed by a post-hoc Dunnett’s test was used for the experiments performed to investigate the influences of incubation and the matrix. Differences between two groups were evaluated using the Mann-Whitney U test, since normality or equality of variance had been rejected by the Kolmogorov-Smirnov test or the Bartlett test for most of the parameters or analyses. The diagnostic properties were investigated with receiver-operating characteristic (ROC) analyses. P values of less than 0.05 were indicative of statistical significance.

Results

Development and Validation of the S1P and dhS1P Measurement System Based on LC-MS/MS

First, we developed the S1P and dhS1P measurement system based on LC-MS/MS. Extracted ion chromatograms obtained from the MRM transitions are shown in Fig. 1A. The lower limit of quantification was 0.01 ng/mL for both S1P and dhS1P. Linearity was confirmed over the entire range of the tested concentrations of standard solutions (0–10 ng/mL) (Fig. 1B and C). We could clearly determine the peaks of S1P and dhS1P in the CSF samples, as shown in Fig. 1D–K. The inter-day and intra-day precisions determined with the standard solution and the plasma and CSF samples are shown in Table 2. Although the CVs for CSF were higher than those for plasma, since the concentrations of S1P and dhS1P were lower in the CSF, the CVs for the CSF were still below 20% (most of the CVs were below 15%), suggesting that sufficient reproducibility was achieved to measure S1P and dhS1P levels in the CSF using this system We observed no obvious matrix effects for internal standards in the CSF samples, while the signals for C17S1P were augmented in the plasma samples (Supplemental Fig. 1A). For endogenous sphingolipids, we observed that the signals for S1P and dhS1P were significantly suppressed in the CSF samples, while those of S1P were augmented and those of dhS1P were suppressed in the plasma samples (Supplemental Fig. 2A). Although we could not identify target ions by scan mode since several impurities existed in the standard solution, we believe that ionization would not be a matter in our measurement systems for CSF samples, since we confirmed target ions when we analyze S1P and dhS1P by SIM mode (Supplemental Fig. 3). We also performed recovery tests to investigate whether ionization of S1P and dhS1P would be inhibited by CSF matrix and observed that the recovery rates were not different between in PBS and in CSF matrix (Supplemental Fig. 4A). We further investigated whether the protein levels and/or cell counts could influence the measurement for S1P and dhS1P. As shown in Supplemental Figs. 5A and 6A, we observed no specific influences of the protein levels and/or cell counts on the recovery rates for S1P and dhS1P.

Measurement system for S1P and dhS1P in cerebrospinal fluid specimens. (A) Retention time and chromatograms of sphingosine 1-phosphate (S1P) and dihydro-sphingosine 1-phosphate (dhS1P) and their internal standards. (B, C) The results of the calibration curves of S1P and dhS1P. (D–K) The chromatograms of S1P (D, H), C17S1P (E, I), dhS1P (F, J), and C17dhS1P (G, K) in the cerebrospinal fluid specimens (D–G) and standard solutions (H–K).
Fig. 1.

Measurement system for S1P and dhS1P in cerebrospinal fluid specimens. (A) Retention time and chromatograms of sphingosine 1-phosphate (S1P) and dihydro-sphingosine 1-phosphate (dhS1P) and their internal standards. (B, C) The results of the calibration curves of S1P and dhS1P. (D–K) The chromatograms of S1P (D, H), C17S1P (E, I), dhS1P (F, J), and C17dhS1P (G, K) in the cerebrospinal fluid specimens (D–G) and standard solutions (H–K).

Open in new tabDownload slide
Table 2
Open in new tab

Intra-day and inter-day precisions of the measurement of S1P and dhS1P.

Intra-day and inter-day measurement precisions of standard solutions
Concentration (ng/mL)Intra-day CV (%)Inter-day CV (%)
S1P0.017.08.8
0.052.27.6
0.16.92.5
11.13.1
101.56.5
dhS1P0.0112.312.9
0.055.36.7
0.16.06.4
11.61.9
101.87.3
Intra-day and inter-day measurement precisions in CSF samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 113.219.9
Sample 218.712.2
Sample 35.310.1
Sample 410.316.5
Sample 512.511.2
dhS1PSample 15.318.7
Sample 219.311.4
Sample 313.217.3
Sample 46.717.7
Sample 58.49.9
Intra-day and inter-day measurement precisions in plasma samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 16.77.7
Sample 25.49.8
dhS1PSample 11.12.7
Sample 20.81.5
Intra-day and inter-day measurement precisions of standard solutions
Concentration (ng/mL)Intra-day CV (%)Inter-day CV (%)
S1P0.017.08.8
0.052.27.6
0.16.92.5
11.13.1
101.56.5
dhS1P0.0112.312.9
0.055.36.7
0.16.06.4
11.61.9
101.87.3
Intra-day and inter-day measurement precisions in CSF samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 113.219.9
Sample 218.712.2
Sample 35.310.1
Sample 410.316.5
Sample 512.511.2
dhS1PSample 15.318.7
Sample 219.311.4
Sample 313.217.3
Sample 46.717.7
Sample 58.49.9
Intra-day and inter-day measurement precisions in plasma samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 16.77.7
Sample 25.49.8
dhS1PSample 11.12.7
Sample 20.81.5
Table 2
Open in new tab

Intra-day and inter-day precisions of the measurement of S1P and dhS1P.

Intra-day and inter-day measurement precisions of standard solutions
Concentration (ng/mL)Intra-day CV (%)Inter-day CV (%)
S1P0.017.08.8
0.052.27.6
0.16.92.5
11.13.1
101.56.5
dhS1P0.0112.312.9
0.055.36.7
0.16.06.4
11.61.9
101.87.3
Intra-day and inter-day measurement precisions in CSF samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 113.219.9
Sample 218.712.2
Sample 35.310.1
Sample 410.316.5
Sample 512.511.2
dhS1PSample 15.318.7
Sample 219.311.4
Sample 313.217.3
Sample 46.717.7
Sample 58.49.9
Intra-day and inter-day measurement precisions in plasma samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 16.77.7
Sample 25.49.8
dhS1PSample 11.12.7
Sample 20.81.5
Intra-day and inter-day measurement precisions of standard solutions
Concentration (ng/mL)Intra-day CV (%)Inter-day CV (%)
S1P0.017.08.8
0.052.27.6
0.16.92.5
11.13.1
101.56.5
dhS1P0.0112.312.9
0.055.36.7
0.16.06.4
11.61.9
101.87.3
Intra-day and inter-day measurement precisions in CSF samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 113.219.9
Sample 218.712.2
Sample 35.310.1
Sample 410.316.5
Sample 512.511.2
dhS1PSample 15.318.7
Sample 219.311.4
Sample 313.217.3
Sample 46.717.7
Sample 58.49.9
Intra-day and inter-day measurement precisions in plasma samples
Intra-day CV (%)Inter-day CV (%)
S1PSample 16.77.7
Sample 25.49.8
dhS1PSample 11.12.7
Sample 20.81.5

Validation of the Ceramides and Sphingosines Measurement System Based on LC-MS/MS in the CSF Samples

We developed a measurement system for ceramides and sphingosines similar to a previously described method (12). Extracted ion chromatograms obtained from the MRM transitions are shown in Supplemental Fig. 7. We confirmed the linearity of measurements of the diluted standard solutions (Supplemental Fig. 8) and could clearly determine the peaks of the ceramides and sphingosines in the CSF samples (Fig. 2). Good precision was observed in the intra-day and inter-day analyses, the values being <20% (most of the CVs being below 15%) for all the monitored molecules and samples (Supplemental Table 2). The CSF samples revealed almost no matrix effects, while the signals for Cer d18:1/17:0 were attenuated in the plasma samples (Supplemental Fig. 1B). For the endogenous sphingosines and ceramides, we observed no obvious matrix effects except Sph and Cer C18:0 in the CSF samples, while in the plasma samples the signals for dhSph, Cer C16:0, Cer C18:0, and Cer C18:1 were attenuated and those for Cer C20:0, Cer C22:0, and Cer C24:0 were augmented (Supplemental Fig. 2B and C). Although we could not identify target ions by scan mode, in CSF samples, they were confirmed in analysis of sphingosines and ceramides by SIM mode (Supplemental Figs. 9 and 10). We also confirmed that the recovery rates were not different between when we added the standards in CSF and PBS, except for Cer C18:0, Cer C18:1, and Cer C24:0 (Supplemental Fig. 4B). The levels of Cer C16:0 were positively influenced, while Cer C22:0 were negatively influenced in the CSF samples with extremely high protein levels and cell counts (Supplemental Figs. 5B–D and 6B–D).

Measurement system for sphingosines and ceramides in the cerebrospinal fluid specimens. The chromatograms of Sph (A), C17Sph (B), dhSph (C), C17dhSph (D), Cer d18:1/16:0 (E), Cer d18:1/18:0 (F), Cer d18:1/18:1 (G), Cer d18:1/20:0 (H), Cer d18:1/22:0 (I), Cer d18:1/24:0 (J), Cer d18:1/17:0 (K) in the cerebrospinal fluid are shown.
Fig. 2.

Measurement system for sphingosines and ceramides in the cerebrospinal fluid specimens. The chromatograms of Sph (A), C17Sph (B), dhSph (C), C17dhSph (D), Cer d18:1/16:0 (E), Cer d18:1/18:0 (F), Cer d18:1/18:1 (G), Cer d18:1/20:0 (H), Cer d18:1/22:0 (I), Cer d18:1/24:0 (J), Cer d18:1/17:0 (K) in the cerebrospinal fluid are shown.

Open in new tabDownload slide

No Elevation of the Sphingolipid Levels Was Observed after Sample Collection, during the Incubation of the CSF Samples at Room Temperature

As reported previously (9,), levels of S1P could increase when the samples are placed at room temperature without ultracentrifugation after sampling, mainly since S1P can be release by erythrocytes (13–15,) and platelets (16, 17). Therefore, we investigated whether the levels of sphingolipids measured in the present study, especially S1P, could increase with incubation of the sample at room temperature without ultracentrifugation. As shown in Fig. 3, no elevation of the levels of S1P and dhS1P were observed in the CSF samples, but were reduced by a slight degree when the CSF samples were placed at room temperature.

The effects of duration of incubation after sampling on the levels of sphingolipids in the cerebrospinal fluid specimens. Three independent cerebrospinal fluid samples were incubated at room temperature (A, C) or at 4 °C (B, D) for 0, 1, 3, 6 h and the samples were collected for the measurement of S1P and dhS1P (A, B) and sphingosines and ceramides (C, D) (n = 5). The ratios of the levels of sphingolipids after incubation to those at 0 h were calculated. *P < 0.05 vs. 0 h, †P < 0.01 vs. 0 h.
Fig. 3.

The effects of duration of incubation after sampling on the levels of sphingolipids in the cerebrospinal fluid specimens. Three independent cerebrospinal fluid samples were incubated at room temperature (A, C) or at 4 °C (B, D) for 0, 1, 3, 6 h and the samples were collected for the measurement of S1P and dhS1P (A, B) and sphingosines and ceramides (C, D) (n = 5). The ratios of the levels of sphingolipids after incubation to those at 0 h were calculated. *P < 0.05 vs. 0 h, †P < 0.01 vs. 0 h.

Open in new tabDownload slide

The Contamination with Blood Components Increased the Levels of Sphingolipids in CSF Samples

We investigated whether the contamination with blood components might affect the levels of sphingolipids in CSF samples. We observed that the concentrations of sphingolipids became higher when whole blood was added to CSF samples (Supplemental Fig. 11), suggesting that it is important to avoid the contamination with blood and to centrifuge the samples as quickly as possible to measure sphingolipids in CSF precisely.

The S1P/Ceramides and S1P/Sph Ratios Were Higher in the CSF Samples Collected from the Subjects with Carcinomatous Meningitis

Lastly, using this validated method to measure the levels of S1P, dhS1P, ceramides, and sphingosines, we measured the concentrations of these substances in the CSF samples subjects with carcinomatous meningitis. For the samples with the levels of sphingolipids more than 10 ng/mL, we diluted them with PBS and measured again. Although the S1P levels in the CSF tended to be higher in the samples obtained from patients with carcinomatous meningitis as compared to normal control samples, the differences were not significant (Fig. 4A). However, the Sph and dhSph levels in the CSF samples were lower in carcinomatous meningitis, while the S1P/Sph and dhS1P/dhSph ratios were significantly higher in these samples (Fig. 4A and B). The ceramide levels of Cer d18:1/18:0, Cer d18:1/18:1, and Cer d18:1/22:0 were significantly lower in the samples from patients with carcinomatous meningitis, and that the ratios of S1P to Cer d18:1/18:0, Cer d18:1/18:1, Cer d18:1/20:0, or Cer d18:1/22:0 were higher in these samples (Fig. 4C and D). When we performed ROC analyses, we observed that Sph, S1P/Sph, S1P/Cer C18, and S1P/Cer C18:1 possessed a similar area under the curve to total protein levels (Supplemental Table 3 and Supplemental Fig. 12).

The levels of sphingolipids in cerebrospinal fluid specimens collected from patients with cancerous meningitis. The levels of S1P, dhS1P, sphingosines, and ceramides were measured in cerebrospinal fluid specimens collected from patientswith cancerous meningitis (n = 21) and control subjects (n = 27). (A) The levels of S1P, dhS1P, Sph, and dhSph. (B) The S1P or dhS1P to Sph or dhSph ratios. (C) The levels of Cer d18:1/16:0 (C16:0), Cer d18:1/18:0 (C18:0), Cer d18:1/18:1 (C18:1), Cer d18:1/20:0 (C20:0), Cer d18:1/22:0 (C22:0), and Cer d18:1/24:0 (C24:0). (D) The levels of S1P for each ceramide species.
Fig. 4.

The levels of sphingolipids in cerebrospinal fluid specimens collected from patients with cancerous meningitis. The levels of S1P, dhS1P, sphingosines, and ceramides were measured in cerebrospinal fluid specimens collected from patientswith cancerous meningitis (n = 21) and control subjects (n = 27). (A) The levels of S1P, dhS1P, Sph, and dhSph. (B) The S1P or dhS1P to Sph or dhSph ratios. (C) The levels of Cer d18:1/16:0 (C16:0), Cer d18:1/18:0 (C18:0), Cer d18:1/18:1 (C18:1), Cer d18:1/20:0 (C20:0), Cer d18:1/22:0 (C22:0), and Cer d18:1/24:0 (C24:0). (D) The levels of S1P for each ceramide species.

Open in new tabDownload slide

Discussion

A method that was developed to measure the levels of S1P, dhS1P, ceramides, and sphingosines in human CSF samples was well-validated and showed high sensitivity (Figs. 1 and 2, Table 2, Supplemental Fig. 8, Supplemental Table 2). As compared to the measurement systems for S1P reported previously (12, 18–24), our system showed sufficient sensitivity to measure low levels of S1P, dhS1P, sphingosines, and ceramides in the CSF within a short time. Although we cannot measure S1P, dhS1P, sphingosines, and ceramides simultaneously, since we used the same solvents and column for both systems, we can perform analyses of both S1P and, dhS1P and of sphingosines and ceramides without changing the mobile solvents or column. The main limitation of the present method is the issues of matrix effects. As shown in Supplemental Figs. 1 and 2, the signals of internal standards and some endogenous sphingolipids were differently influenced by the CSF and plasma matrixes. Since the signals of S1P, dhS1P, and Sph were suppressed in the matrix of CSF samples, their concentrations could be underestimated. In plasma samples, the levels of long chain ceramides would be overestimated. Considering the results of the recovery tests in the CSF samples with various protein levels and cell counts (Supplemental Figs. 5 and 6), however, we believe that the levels of sphingolipids in CSF samples can be compared with those of other CSF samples, except the concentrations of long-chain ceramides in CSF samples with extremely high levels of total protein levels and cell counts. Further studies on the issues of matrix effects are necessary to improve the present methods.

Using this method, we found the possible involvement of S1P and ceramides in the pathogenesis of human cancer. The S1P-ceramides rheostat hypothesis is believed to be involved in the pathogenesis of human cancers (Supplemental Fig. 13) (25,); however, this hypothesis remains to be fully verified in clinical studies conducted with human samples. In the present study, we found that the ratio of S1P to Sph or ceramides, as well as the ratio of dhS1P to dhSph, was generally elevated in the CSF samples collected from patients with cancerous meningitis. These results suggest that the S1P-ceramides rheostat hypothesis must certainly be considered in the pathogenesis of human cancer, at least in the advanced stage, such as at the stage of development of cancerous meningitis. In the demonstration of the involvement of the S1P-rheostat theory in humans, many studies have reported upregulation of Sph kinase, a limiting enzyme for the production of S1P, in colon cancer (26, 27,), hepatocellular carcinoma (28,), breast cancer (29, 30,), and acute leukemia (31). However, there are a few reports demonstrating that ceramides are actually converted to S1P in human cancer.

One reason for the lack of research on the measurement of sphingolipids, especially S1P, is the difficulty in measuring S1P in human blood. The levels of S1P in samples can easily increase during/after sampling, as S1P is released from erythrocytes and platelets (9,), and can be degraded both inside and outside the cells by enzymes such as S1P lyase, S1P phosphatase, and lipid phosphate phosphatase (11, 32). To overcome these obstacles, we measured the sphingolipid levels in CSF, which contain few blood cells. No increase in the levels of S1P in CSF was found, similar to the case for other sphingolipids, after sampling (Fig. 3). Therefore, we propose that CSF samples are more amenable to measurement of S1P as compared to blood samples and tissues.

Although limited reports have thus far been published on the modulation of sphingolipid levels, especially of S1P, in CSF samples in human diseases, several studies have proposed the potential usefulness of the measurement system in diseases other than cancerous meningitis. The S1P levels in the CSF have been reported to be lower in subjects with Alzheimer disease than in those with idiopathic normal-pressure hydrocephalus (33,), whereas the CSF levels of S1P are elevated in the subjects with multiple sclerosis (34,). Several studies have demonstrated modulation of the CSF levels of ceramides in human diseases, such as subarachnoid hemorrhage (35, 36,), psychosis (37,), amyotrophic lateral sclerosis (38,), and multiple sclerosis (39). Although these studies provide important evidence for the involvement of ceramides in the pathogenesis of human diseases, the measurement of S1P in addition would allow a clearer elucidation of the underlying pathological mechanisms. Moreover, precise measurement of sphingolipids would make it possible to introduce the testing into clinical laboratory medicine. Although the present methods might not be applied to diagnose cancerous meningitis, since the area under the curves of sphingolipids were not obviously larger than those of total protein levels (Supplemental Table 3), we believe that the methods could pave the way to obtain important evidence of the involvement of sphingolipids in human diseases and introduce the biology of sphingolipids into clinical practice in the future.

In conclusion, we have developed and validated a measurement system for S1P, dhS1P, ceramides, and sphingosines, which allows precise measurement of sphingolipids in human CSF samples. Using this method, we verified the involvement of the S1P-ceramides rheostat theory in the pathogenesis of human cancerous meningitis.

Supplemental Material

Supplemental material is available at The Journal of Applied Laboratory Medicine online.

Nonstandard Abbreviations

S1P, sphingosine 1-phosphate; dhS1P, dihydro-sphingosine 1-phosphate; CSF, cerebrospinal fluid; LC-MS/MS, liquid chromatography-tandem mass spectrometry; Sph, sphingosine;dhSph, dihydrosphingosine; MRM, multiple reaction monitoring; Cer, ceramide; ROC, receiver-operating characteristic.

Author Contributions

All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved.

M. Kurano and Y. Yatomi designed the research; E. Sakai and M. Kurano analyzed the data; E. Sakai, M. Kurano, and Y. Morita performed the research; M. Kurano, J. Aoki, and Y. Yatomi wrote the paper. All authors have read and approved the final manuscript.

Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the author disclosure form. Authors' disclosures and/or potential conflicts of interest. Employment or Leadership: None declared. Consultant or Advisory Role: None declared. Stock Ownership: None declared. Honoraria: None declared. Expert Testimony: LEAP from the AMED. M. Kurano, JSPS KAKENHI Grant Numbers and 16H06236; Y. Yatomia, Grant-in-Aid for Scientific Research on Innovative Areas 15H05906. Patents: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, preparation of manuscript, or final approval of manuscript.

REFERENCES

1

Kunkel
GT
,
Maceyka
M
,
Milstien
S
,
Spiegel
S.
Targeting the sphingosine-1-phosphate axis in cancer, inflammation and beyond
.
Nat Rev Drug Discov
2013
;
12
:
688
–
702
.

Google Scholar

Crossref
Search ADS
PubMed

 
2

Brachtendorf
S
,
El-Hindi
K
,
Grosch
S.
Ceramide synthases in cancer therapy and chemoresistance
.
Prog Lipid Res
2019
;
74
:
160
–
85
.

Google Scholar

Crossref
Search ADS
PubMed

 
3

Spiegel
S
,
Milstien
S.
The outs and the ins of sphingosine-1-phosphate in immunity
.
Nat Rev Immunol
2011
;
11
:
403
–
15
.

Google Scholar

Crossref
Search ADS
PubMed

 
4

Kurano
M
,
Yatomi
Y.
Sphingosine 1-phosphate and atherosclerosis
.
J Atheroscler Thromb
2018
;
25
:
16
–
26
.

Google Scholar

Crossref
Search ADS
PubMed

 
5

Mihanfar
A
,
Nejabati
HR
,
Fattahi
A
,
Latifi
Z
,
Pezeshkian
M
,
Afrasiabi
A
, et al.
The role of sphingosine 1 phosphate in coronary artery disease and ischemia reperfusion injury
.
J Cell Physiol
2019
;
234
:
2083
–
94
.

Google Scholar

Crossref
Search ADS
PubMed

 
6

Yu
Z
,
Peng
Q
,
Huang
Y.
Potential therapeutic targets for atherosclerosis in sphingolipid metabolism
.
Clin Sci (Lond)
2019
;
133
:
763
–
76
.

Google Scholar

Crossref
Search ADS
PubMed

 
7

Sabourdy
F
,
Astudillo
L
,
Colacios
C
,
Dubot
P
,
Mrad
M
,
Segui
B
, et al.
Monogenic neurological disorders of sphingolipid metabolism
.
Biochim Biophys Acta
2015
;
1851
:
1040
–
51
.

Google Scholar

Crossref
Search ADS
PubMed

 
8

Healy
LM
,
Antel
JP.
Sphingosine-1-phosphate receptors in the central nervous and immune systems
. Curr Drug Targets
2016
;
17
:
1841
–
50
.

9

Ohkawa
R
,
Nakamura
K
,
Okubo
S
,
Hosogaya
S
,
Ozaki
Y
,
Tozuka
M
, et al.
Plasma sphingosine-1-phosphate measurement in healthy subjects: close correlation with red blood cell parameters
.
Ann Clin Biochem
2008
;
45
:
356
–
63
.

Google Scholar

Crossref
Search ADS
PubMed

 
10

Morita
Y
,
Kurano
M
,
Sakai
E
,
Nishikawa
M
,
Sawabe
M
,
Aoki
J
,
Yatomi
Y.
Evaluation of lysophospholipid measurement in cerebrospinal fluid samples using liquid chromatography-tandem mass spectrometry
.
Lipids
2019
;
54
:
487
–
500
.

Google Scholar

Crossref
Search ADS
PubMed

 
11

Le Stunff
H
,
Milstien
S
,
Spiegel
S.
Generation and metabolism of bioactive sphingosine-1-phosphate
.
J Cell Biochem
2004
;
92
:
882
–
99
.

Google Scholar

Crossref
Search ADS
PubMed

 
12

Saigusa
D
,
Okudaira
M
,
Wang
J
,
Kano
K
,
Kurano
M
,
Uranbileg
B
, et al.
Simultaneous quantification of sphingolipids in small quantities of liver by LC-MS/MS
.
Mass Spectrom (Tokyo)
2014
;
3
:
S0046
.

Google Scholar

Crossref
Search ADS
PubMed

 
13

Pappu
R
,
Schwab
SR
,
Cornelissen
I
,
Pereira
JP
,
Regard
JB
,
Xu
Y
, et al.
Promotion of lymphocyte egress into blood and lymph by distinct sources of sphingosine-1-phosphate
.
Science
2007
;
316
:
295
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
14

Vu
TM
,
Ishizu
AN
,
Foo
JC
,
Toh
XR
,
Zhang
F
,
Whee
DM
, et al.
Mfsd2b is essential for the sphingosine-1-phosphate export in erythrocytes and platelets
.
Nature
2017
;
550
:
524
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
15

Kurano
M
,
Nishikawa
M
,
Kuma
H
,
Jona
M
,
Yatomi
Y.
Involvement of band3 in the efflux of sphingosine 1-phosphate from erythrocytes
.
PLoS One
2017
;
12
:
e0177543
.

Google Scholar

Crossref
Search ADS
PubMed

 
16

Yatomi
Y
,
Ruan
F
,
Hakomori
S
,
Igarashi
Y.
Sphingosine-1-phosphate: a platelet-activating sphingolipid released from agonist-stimulated human platelets
.
Blood
1995
;
86
:
193
–
202
.

Google Scholar

Crossref
Search ADS
PubMed

 
17

Mishima
Y
,
Kurano
M
,
Kobayashi
T
,
Nishikawa
M
,
Ohkawa
R
,
Tozuka
M
,
Yatomi
Y.
Dihydro-sphingosine 1-phosphate interacts with carrier proteins in a manner distinct from that of sphingosine 1-phosphate
.
Biosci Rep
2018
;
38
.

Google Scholar

OpenURL Placeholder Text

 
18

Lan
T
,
Bi
H
,
Liu
W
,
Xie
X
,
Xu
S
,
Huang
H.
Simultaneous determination of sphingosine and sphingosine 1-phosphate in biological samples by liquid chromatography-tandem mass spectrometry
.
J Chromatogr B Analyt Technol Biomed Life Sci
2011
;
879
:
520
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
19

Berdyshev
EV
,
Gorshkova
IA
,
Garcia
JG
,
Natarajan
V
,
Hubbard
WC.
Quantitative analysis of sphingoid base-1-phosphates as bisacetylated derivatives by liquid chromatography-tandem mass spectrometry
.
Anal Biochem
2005
;
339
:
129
–
36
.

Google Scholar

Crossref
Search ADS
PubMed

 
20

Schmidt
H
,
Schmidt
R
,
Geisslinger
G.
LC-MS/MS-analysis of sphingosine-1-phosphate and related compounds in plasma samples
.
Prostaglandins Other Lipid Mediat
2006
;
81
:
162
–
70
.

Google Scholar

Crossref
Search ADS
PubMed

 
21

Scherer
M
,
Schmitz
G
,
Liebisch
G.
High-throughput analysis of sphingosine 1-phosphate, sphinganine 1-phosphate, and lysophosphatidic acid in plasma samples by liquid chromatography-tandem mass spectrometry
.
Clin Chem
2009
;
55
:
1218
–
22
.

Google Scholar

Crossref
Search ADS
PubMed

 
22

Frej
C
,
Andersson
A
,
Larsson
B
,
Guo
LJ
,
Norström
E
,
Happonen
KE
,
Dahlbäck
B.
Quantification of sphingosine 1-phosphate by validated lc-ms/ms method revealing strong correlation with apolipoprotein m in plasma but not in serum due to platelet activation during blood coagulation
.
Anal Bioanal Chem
2015
;
407
:
8533
–
42
.

Google Scholar

Crossref
Search ADS
PubMed

 
23

Chipeaux
C
,
de Person
M
,
Burguet
N
,
Billette de Villemeur
T
,
Rose
C
,
Belmatoug
N
, et al.
Optimization of ultra-high pressure liquid chromatography—tandem mass spectrometry determination in plasma and red blood cells of four sphingolipids and their evaluation as biomarker candidates of Gaucher’s disease
.
J Chromatogr A
2017
;
1525
:
116
–
25
.

Google Scholar

Crossref
Search ADS
PubMed

 
24

Bielawski
J
,
Szulc
ZM
,
Hannun
YA
,
Bielawska
A.
Simultaneous quantitative analysis of bioactive sphingolipids by high-performance liquid chromatography-tandem mass spectrometry
.
Methods
2006
;
39
:
82
–
91
.

Google Scholar

Crossref
Search ADS
PubMed

 
25

Ogretmen
B.
Sphingolipid metabolism in cancer signalling and therapy
.
Nat Rev Cancer
2018
;
18
:
33
–
50
.

Google Scholar

Crossref
Search ADS
PubMed

 
26

Tan
SS
,
Khin
LW
,
Wong
L
,
Yan
B
,
Ong
CW
,
Datta
A
, et al.
Sphingosine kinase 1 promotes malignant progression in colon cancer and independently predicts survival of patients with colon cancer by competing risk approach in South Asian population
.
Clin Transl Gastroenterol
2014
;
5
:
e51
.

Google Scholar

Crossref
Search ADS
PubMed

 
27

Uranbileg
B
,
Nishikawa
T
,
Ikeda
H
,
Kurano
M
,
Sato
M
,
Saigusa
D
, et al.
Evidence suggests sphingosine 1-phosphate might be actively generated, degraded, and transported to extracellular spaces with increased S1P2 and S1P3 expression in colon cancer
.
Clin Colorectal Cancer
2018
;
17
:
e171
–
82
.

Google Scholar

Crossref
Search ADS
PubMed

 
28

Uranbileg
B
,
Ikeda
H
,
Kurano
M
,
Enooku
K
,
Sato
M
,
Saigusa
D
, et al.
Increased mRNA levels of sphingosine kinases and S1P lyase and reduced levels of S1P were observed in hepatocellular carcinoma in association with poorer differentiation and earlier recurrence
.
PLoS One
2016
;
11
:
e0149462
.

Google Scholar

Crossref
Search ADS
PubMed

 
29

Nagahashi
M
,
Ramachandran
S
,
Kim
EY
,
Allegood
JC
,
Rashid
OM
,
Yamada
A
, et al.
Sphingosine-1-phosphate produced by sphingosine kinase 1 promotes breast cancer progression by stimulating angiogenesis and lymphangiogenesis
.
Cancer Res
2012
;
72
:
726
–
35
.

Google Scholar

Crossref
Search ADS
PubMed

 
30

Tsuchida
J
,
Nagahashi
M
,
Nakajima
M
,
Moro
K
,
Tatsuda
K
,
Ramanathan
R
, et al.
Breast cancer sphingosine-1-phosphate is associated with phospho-sphingosine kinase 1 and lymphatic metastasis
.
J Surg Res
2016
;
205
:
85
–
94
.

Google Scholar

Crossref
Search ADS
PubMed

 
31

Sobue
S
,
Iwasaki
T
,
Sugisaki
C
,
Nagata
K
,
Kikuchi
R
,
Murakami
M
, et al.
Quantitative RT-PCR analysis of sphingolipid metabolic enzymes in acute leukemia and myelodysplastic syndromes
.
Leukemia
2006
;
20
:
2042
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
32

Le Stunff
H
,
Peterson
C
,
Liu
H
,
Milstien
S
,
Spiegel
S.
Sphingosine-1-phosphate and lipid phosphohydrolases
.
Biochim Biophys Acta
2002
;
1582
:
8
–
17
.

Google Scholar

Crossref
Search ADS
PubMed

 
33

Torretta
E
,
Arosio
B
,
Barbacini
P
,
Casati
M
,
Capitanio
D
,
Mancuso
R
, et al.
Particular CSF sphingolipid patterns identify inph and ad patients
.
Sci Rep
2018
;
8
:
13639
.

Google Scholar

Crossref
Search ADS
PubMed

 
34

Kulakowska
A
,
Zendzian-Piotrowska
M
,
Baranowski
M
,
Kononczuk
T
,
Drozdowski
W
,
Gorski
J
,
Bucki
R.
Intrathecal increase of sphingosine 1-phosphate at early stage multiple sclerosis
.
Neurosci Lett
2010
;
477
:
149
–
52
.

Google Scholar

Crossref
Search ADS
PubMed

 
35

Testai
FD
,
Hillmann
M
,
Amin-Hanjani
S
,
Gorshkova
I
,
Berdyshev
E
,
Gorelick
PB
,
Dawson
G.
Changes in the cerebrospinal fluid ceramide profile after subarachnoid hemorrhage
.
Stroke
2012
;
43
:
2066
–
70
.

Google Scholar

Crossref
Search ADS
PubMed

 
36

Testai
FD
,
Xu
HL
,
Kilkus
J
,
Suryadevara
V
,
Gorshkova
I
,
Berdyshev
E
, et al.
Changes in the metabolism of sphingolipids after subarachnoid hemorrhage
.
J Neurosci Res
2015
;
93
:
796
–
805
.

Google Scholar

Crossref
Search ADS
PubMed

 
37

Checa
A
,
Malmqvist
A
,
Flyckt
L
,
Schwieler
L
,
Samuelsson
M
,
Skogh
E
, et al.
Cerebrospinal fluid levels of sphingolipids associate with disease severity in first episode psychosis patients
.
Schizophr Res
2018
;
199
:
438
–
41
.

Google Scholar

Crossref
Search ADS
PubMed

 
38

Blasco
H
,
Veyrat-Durebex
C
,
Bocca
C
,
Patin
F
,
Vourc’h
P
,
Kouassi Nzoughet
J
, et al.
Lipidomics reveals cerebrospinal-fluid signatures of ALS
.
Sci Rep
2017
;
7
:
17652
.

Google Scholar

Crossref
Search ADS
PubMed

 
39

Vidaurre
OG
,
Haines
JD
,
Katz Sand
I
,
Adula
KP
,
Huynh
JL
,
McGraw
CA
, et al.
Cerebrospinal fluid ceramides from patients with multiple sclerosis impair neuronal bioenergetics
.
Brain
2014
;
137
:
2271
–
86
.

Google Scholar

Crossref
Search ADS
PubMed

 

Nonstandard Abbreviations

     
  • S1P

    sphingosine 1-phosphate

    •  
  • dhS1P

    dihydro-sphingosine 1-phosphate

    •  
  • CSF

    cerebrospinal fluid

    •  
  • LC-MS/MS

    liquid chromatography-tandem mass spectrometry

    •  
  • Sph

    sphingosine

    •  
  • dhSph

    dihydrosphingosine

    •  
  • MRM

    multiple reaction monitoring

    •  
  • Cer

    ceramide

    •  
  • ROC

    receiver-operating characteristic

© American Association for Clinical Chemistry 2020. All rights reserved. For permissions, please email: [email protected]
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Issue Section:
Articles > Lipids, Lipoproteins, and Cardiovascular Risk Factors
Download all slides
  • Supplementary data

    • Supplementary data

    jfaa022_Supplementary_Data - zip file