Role of alkaline ceramidases in the generation of sphingosine and its phosphate in erythrocytes

Abstract

Plasma sphingosine-1-phosphate (S1P) has been suggested to mainly originate from erythrocytes; however, within the erythrocyte, how sphingosine (SPH) generation--the precursor to S1P--is controlled is unknown. SPH is only generated from the hydrolysis of ceramides via ceramidases. Five human ceramidases have been identified: 1 acid, 1 neutral, and 3 alkaline ceramidases (ACER1, ACER2, and ACER3). Here, we demonstrate that only alkaline ceramidase activity is expressed in erythrocytes and that it is instrumental for SPH generation. Erythrocytes have alkaline but not acid or neutral ceramidase activity on D-e-C(18:1)-ceramide, a common substrate of ceramidases. Not only alkaline ceramidase activity but also the generation of SPH and S1P are increased during erythroid differentiation in K562 erythroleukemic cells. Such SPH and S1P increases were inhibited by the alkaline ceramidase inhibitor D-e-MAPP, suggesting that alkaline ceramidases have a role in the generation of SPH and S1P in erythroid cells. Alkaline ceramidase activity is highly expressed in mouse erythrocytes, and intravenous administration of D-e-MAPP decreased both SPH and S1P in erythrocytes and plasma. Collectively, these results suggest that alkaline ceramidase activity is important for the generation of SPH, the S1P precursor in erythrocytes.

Figure 1.

Alkaline but not acid or neutral ceramidase is expressed in human erythrocytes. A) Total membranes were prepared from human erythrocytes and subjected to ceramidase activity assays using D-e-C_18:1-ceramide as a substrate at pH 5.0 or 7.0 in the absence of Ca²⁺ or at pH 9.4 in the presence of 1 mM CaCl₂. B) Total membranes of human erythrocytes were assayed for alkaline ceramidase activity on D-e-C₁₂-NBD-ceramide, D-e-C₁₆-ceramide, or D-e-C_24:1-ceramide at pH 9.4 and in the presence of 1 mM Ca²⁺. Data represent means ± sd of 3 experiments performed in duplicate.

Figure 2.

Both alkaline ceramidase activity and SPH and S1P levels are increased during erythroid differentiation. A) K562 cells were treated with Ara-C or H₂O (vehicle control) for 72 h before erythroid differentiation was determined by benzidine staining as described in Materials and Methods. Benzidine-positive cells were counted using a microscope. B) Microsomes were isolated from K562 cells treated with Ara-C as in A, and alkaline ceramidase activity on different substrates was measured as in Fig. 1B. C) Total RNA was isolated from K562 cells treated with Ara-C or H₂O and subjected to qPCR analysis to measure ACER1, ACER2, or ACER3 mRNA. D) K562 cells treated with Ara-C or H₂O were harvested by centrifugation, washed with 50 mM Tris-HCl (pH 7.4) containing 150 mM NaCl, and subjected to ESI/MS/MS for SPH and S1P measurement. Data represent means ± sd of 3 experiments performed in duplicate. *P < 0.05.

Figure 3.

Inhibition of alkaline ceramidase activity suppresses SPH and S1P increases in response to erythroid differentiation. A) K562 cells were transduced with Mission lentiviral particles expressing a control nontargeting shRNA (shCON) or a shRNA specific for ACER1 (shACER1), ACER2 (shACER2), or ACER3 (shACER3) at a MOI of 5. At 72 h after lentiviral transduction, ACER1 or ACER2 mRNA levels were determined by qPCR analysis as described in Materials and Methods. B, C) ACER1-TET-ON cells (B) or ACER2-TET-ON cells (C) were grown in the presence of TET (10 ng/ml) or ET (vehicle control) for 24 h before being treated with D-e-MAPP (10 μM) or DMSO (vehicle control) for 4 h. SPH in these cells was measured by HPLC. D) ACER3-TET-ON cells, created as described in Materials and Methods, were grown in the presence of ET and TET for 48 h before expression of the FLAG-tagged ACER3 (FLAG-ACER3) was analyzed by Western blot with anti-FLAG antibody. E) ACER3-TET-ON cells were grown in the presence of ET or TET for 24 h before being treated with D-e-MAPP or DMSO for 4 h. SPH was measured by HPL. F) AC-TET-ON cells were grown in the presence of ET or TET for 24 h before being treated with D-e-MAPP or DMSO for 4 h. SPH was measured by HPLC. G) K562 cells were treated with Ara-C or H₂O in the presence of D-e-MAPP or DMSO for 72 h before SPH and S1P were measured by ESI/MS/MS. Data represent means ± sd of 3 experiments performed in duplicate. *P < 0.05.

Figure 4.

Inhibition of alkaline ceramidase activity blocks S1P accumulation in response to knockdown of SPL in erythroid cells. A, B) K562 cells were transduced with Mission lentiviral particles expressing a SPL-specific shRNA (shSPL) or the control siRNA (shCON) for 72 h before being analyzed by Western blot with anti-SPL antibody (A) or by in vitro activity assays (B). C) K562 cells were transduced with shCON or shSPL Mission lentiviral particles in the presence of Ara-C or H₂O for 72 h before S1P was measured by ESI/MS/MS. D) K562 cells transduced with shSPL lentiviruses were treated with Ara-C or H₂O in the presence of D-e-MAPP or DMSO for 72 h before S1P was measured. Data represent means ± sd of 3 experiments performed in duplicate. *P < 0.05.

Figure 5.

Increasing alkaline ceramidase activity results in S1P accumulation in cells with SPL knockdown. A, B) ACER1-TET-ON cells grown in the presence of ET or TET were transfected with a SPL-specific siRNA (siSPL) or a control siRNA (siCON) for 72 h before SPL activity (A) and S1P levels (B) were measured. C, D) ACER2-TET-ON cells grown in the presence of ET or TET were transfected with siSPL or siCON for 72 h before SPL activity (C), and the levels of S1P (D) were determined. E, F) ACER3-TET-ON cells grown in the presence of ET or TET were transfected with siSPL or siCON for 72 h before SPL activity (E), and the levels of S1P (F) were determined. Data represent means ± sd of 3 experiments performed in duplicate. *P < 0.05.

Figure 6.

Inhibition of alkaline ceramidase activity decreases mouse plasma S1P and DHS1P. A) Blood was drawn from mice, and blood cells were separated from plasma. Erythrocytes were separated from other blood cells by Ficoll gradient and subjected to alkaline ceramidase activity assays with indicated substrates. B, C) D-e-MAPP (2.5 or 5.0 nmol/g) or vehicle control DMSO was injected into mice through tail veins. At 12 h postinjection, blood was drawn from mice, and erythrocytes and plasma were subjected to ESI/MS/MS to measure SPH (B) and S1P (C). Data represent means ± sd in 3 mice. *P < 0.05.