Background/Purpose: SLE patients are at risk for atherosclerotic cardiovascular disease (ASCVD).  In some SLE patients, high density lipoprotein (HDL) has impaired vasoprotective effects, and this “proinflammatory” HDL (piHDL) is more prevalent in those with ASCVD.  The protective effect of HDL in atherosclerosis is attributable, in part, to its ability to deliver to S1P to receptors on endothelial cells. Apolipoprotein M (ApoM) is a component of some HDLs and serves as a chaperone of sphingosine 1-phosphate (S1P), a critical mediator of vascular homeostasis. S1P interacts with its receptors on endothelial cells to prevent vascular injury and inflammation. We hypothesized that the ApoM/S1P axis is deregulated in SLE.  

Methods: We performed a cross-sectional study to measure ApoM and S1P levels in SLE patients. Plasma samples were obtained from 52 SLE patients who were part of a University of California Los Angeles cohort followed for ASCVD and met at least four of eleven 1982 ACR SLE Classification Criteria (Table 1). Patients on statins or with renal failure (Cr > 2.0) were excluded. Plasma lipid levels were measured using standard methods. Measurement of pro-inflammatory HDL was performed using a cell free LDL oxidation assay. ApoM was measured using a standardized ELISA and S1P levels by mass spectrometry. Statistics were performed using GraphPad Prism 6.0 and SPSS using appropriate non-parametric tests.

Results: Total plasma ApoM and S1P levels were significantly lower in the SLE patients compared to previously published levels in healthy controls (HC). Mean ApoM level was 0.75 ± 0.28 mmol/L in lupus patients versus 0.92 ± 0.32 mmol/L in HC (p<0.0001) and mean S1P level was 47.08 ± 17.08 ng/ml in lupus patients versus 221.7 ± 84.25 ng/ml in HC (p=0.004). There was a positive correlation between levels of ApoM and S1P (r = 0.34, p=0.013).  ApoM levels negatively correlated with body mass index (BMI) (r = -0.54, p<0.0001) and SLE disease duration (r = -0.33, p=0.017). Level of S1P was negatively correlated with BMI (r = -0.45, p=0.002) and hs-CRP (r = -0.46, p=0.008). There was no statistically significant association of ApoM or S1P levels with piHDL status, plaque status, intimal medial thickness (IMT), smoking, age, hypertension, lipid levels, ethnicity or lifetime prednisone use. In a multivariate regression combining these variables, only BMI and disease duration was significantly associated with ApoM levels.

Conclusion: Although total plasma levels of ApoM and S1P in circulation appeared to be significantly lower in SLE patients, they were neither associated with cardiovascular risk factors nor established plaque or piHDL.  Studies comparing ApoM/S1P levels in SLE patients with age and sex matched healthy controls and ApoM/S1P levels in lipid sub-fractions of SLE patients are ongoing. We postulate that levels of ApoM and S1P may be influenced by inflammation status of SLE patients as this has been previously reported in patients with sepsis.