Prosaposin: a new player in cell death prevention of U937 monocytic cells

www.elsevier.com/locate/yexcr

Experimental Cell Research 298 (2004) 38–47

Prosaposin: a new player in cell death prevention of U937 monocytic cells

Roberta Misasi,a,* Tina Garofalo,a Luisa Di Marzio,b Vincenzo Mattei,a Chiara Gizzi,a

Masao Hiraiwa,c Antonio Pavan,d Maria Grazia Cifone,d and Maurizio Soricea

aDipartimento di Medicina Sperimentale e Patologia, Universita ‘‘La Sapienza’’ Roma, Rome, ItalybDipartimento di Scienze del Farmaco, Universita G. D’Annunzio, Chieti Scalo, Italy

cDepartment of Neurosciences, University of California at San Diego, La Jolla, CA 92093, USAdDipartimento di Medicina Sperimentale, Universita di L’Aquila, Italy

Received 11 August 2003, revised version received 2 April 2004

Available online 13 May 2004

Abstract

We report that prosaposin binds to U937 and is active as a protective factor on tumor necrosis factor a (TNFa)-induced cell death. The

prosaposin-derived saposin C binds to U937 cells in a concentration-dependent manner, suggesting that prosaposin behaves similarly.

Prosaposin binding induces U937 cell death prevention, reducing both necrosis and apoptosis. This effect was inhibited by mitogen-activated

protein ERK kinase (MEK) and sphingosine kinase (SK) inhibitors, indicating that prosaposin prevents cell apoptosis by activation of

extracellular signal-regulated kinases (ERKs) and sphingosine kinase. Prosaposin led to rapid ERK phosphorylation in U937 cells as detected

by anti-phospho-p44/42 mitogen-activated protein (MAP) kinase and anti-phosphotyrosine reactivity on ERK immunoprecipitates. It was

partially prevented by apo B-100 and pertussis toxin (PT), suggesting that both lipoprotein receptor-related protein (LRP) receptor and Go-

coupled receptor may play a role in the prosaposin-triggered pathway. Moreover, sphingosine kinase activity was increased by prosaposin

treatment as demonstrated by the enhanced intracellular formation of sphingosine-1-phosphate (S-1-P). The observation that the

phosphatidylinositol 3-kinase (PI3K) inhibitor wortmannin prevented the prosaposin effect on cell apoptosis suggests that sphingosine kinase

exerts its anti-apoptotic activity by the PI3K–Akt pathway.

Thus, cell apoptosis prevention by prosaposin occurs through ERK phosphorylation and sphingosine kinase. The biological effect

triggered by prosaposin might be extended to primary cells because it triggers Erk phosphorylation in peripheral blood mononuclear cells

(PBMCs). This is the first evidence of a biological effect consequent to a signal transduction pathway triggered by prosaposin in cells of non-

neurological origin.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Prosaposin; Apoptosis; TNFa; Sphingosine kinase; ERKs; Lipid microdomains; Sphingosine-1-phosphate

0014-4827/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.yexcr.2004.04.011

Abbreviations: MEK, mitogen-activated protein ERK kinase; ERKs,

extracellular signal-regulated kinase; MAP, mitogen-activated protein; LRP,

lipoprotein receptor-related protein; PI3K, phosphatidylinositol 3-kinase;

SK, sphingosine kinase; S-1-P, sphingosine-1-phosphate; TNFa, tumor

necrosis factor a; PBMC, peripheral blood mononuclear cells; PBS,

phosphate-buffered saline; HPLC, high pressure liquid chromatography;

SDS-PAGE, sodium dodecyl sulphate polyacrilamide gel electrophoresis;

LDL, low-density lipoproteins; DMS, N-N dimethylsphingosine; PT,

pertussis toxin; PMA, phorbol ester myristate acetate; HRP, horseradish

peroxidase; ECL, enhanced chemiluminescence.

* Corresponding author. Dipartimento di Medicina Sperimentale e

Patologia, Universita ‘‘La Sapienza’’ Roma, Policlinico Umberto I, Viale

Regina Elena 324, Rome, Italy. Fax: +39-6-4454820.

E-mail address: [email protected] (R. Misasi).

Introduction

Prosaposin, a glycoprotein encoded by a single locus

on human chromosome 10 [1], is the precursor of four

sphingolipid activator proteins named saposins A, B, C,

and D that are localized within lysosomes and that

activate the hydrolysis of sphingolipids by lysosomal

hydrolases [2]. Beside the precursor function of prosapo-

sin and the lysosomal distribution of mature saposins,

prosaposin has been found in body fluids [3,4] and as a

plasma membrane constituent [5,6]. This distribution of

prosaposin suggests a certain specific function for extra-

cellular prosaposin. Prosaposin was identified as a neuro-

trophic factor [7], and a neurotrophic sequence has been

recognized in the amino terminal portion of the saposin C

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–47 39

domain [8–10]; peptides encompassing this region have

been called Prosaptidesk.

Prosaposin and prosaptides were shown to stimulate

neurite outgrowth and to trigger a signal cascade after

binding to a putative Go-coupled cell surface receptor

[11] or to the low-density lipoprotein (LDL) receptor-

related protein (LRP) [12,13]. In PC12 pheochromocyto-

ma cells, prosaposin was able to activate extracellular

signal-regulated protein kinase (ERKs) and sphingosine

kinase (SK) with sphingosine-1-phosphate (S-1-P) pro-

duction, thus eliciting an effect of proliferation and cell

death prevention [14]. Sphingolipid turnover was impli-

cated in signal transduction by the observation that

external stimuli regulate the activity of sphingomyelinase,

ceramidase, and sphingosine kinase. Prosaposin may be

involved in such stimulation; sphingolipid metabolites,

including ceramide, sphingosine, and S-1-P, play essential

roles in cell growth, survival, and death: the balance

between cellular levels of ceramide that favor cell death

and levels of S-1-P that inhibit death is critical. More-

over, prosaposin has been involved in ERK phosphory-

lation, inducing an increase of sulfatide synthesis in

Schwann cells and oligodendrocytes, and preventing cell

death [15,16]. In addition, prosaposin and prosaptides

were shown to act as myelinotrophic factors [17,18].

The signaling pathway triggered by prosaposin and pro-

saptide in Schwann cell survival has been identified in

phosphatidylinositol 3-kinase (PI3K)-dependent ERK ac-

tivation; phosphorylation of the PI3K signaling target Akt

highly increased after prosaptide treatment of cells [19].

In other studies, it has been reported that the anti-

apoptotic effect of S-1-P was dependent on the PI3K–

Akt pathway [20].

Tumor necrosis factor a (TNFa) is known to have a

cytotoxic effect on a variety of cells [21,22]. In few cases

[20,23], cells may be not sensitive to TNFa cytotoxicity,

and this resistance seems to be in part due to the

activation of intracellular pathways such as SK and

PI3K–Akt, which protect human hepatocytes and endo-

thelial cells from the apoptotic action of TNFa and

probably FasL.

Although prosaposin is present in body fluids, such

as cerebrospinal liquor, milk, seminal fluid, and blood,

at present, it is not known whether it may exert its

biological effects only on neuronal derived cells or can

bind to other cell types, thus activating the signaling

cascade pathway. In this investigation, we analyzed the

binding of prosaposin with U937 cells, a histiocytic cell

line that represents a very useful model for analyzing

the signaling cascade pathway triggered by prosaposin.

We demonstrated that prosaposin treatment prevented

cell apoptosis by activation of ERKs and sphingosine

kinase. This leads us to suggest that such a mechanism

may play a key role in the regulation of the apoptotic

signal transduction pathway in cells of non-neurological

origin.

Materials and methods

Cells

Human histiocytic U937 cells [24] were maintained in

RPMI 1640 medium (Gibco-BRL, Life Technologies Italia

srl, Milan, Italy) containing 10% fetal calf serum plus

100 units/ml penicillin, 100 Ag/ml streptomicin, at 37jC in

a humidified 5% CO2 atmosphere. Peripheral blood mono-

nuclear cells (PBMCs) were isolated from fresh heparinized

blood by Lymphoprep (Nycomed AS Pharma Diagnostic

Div., Oslo, Norway) density-gradient centrifugation and

washed three times in phosphate-buffered saline (PBS),

pH 7.4.

Proteins and inhibitors

Milk prosaposin was prepared as described [3]. Briefly,

human milk was fractionated by ion exchange chromato-

graphy on DEAE-cellulose DE-52 followed by lectin affinity

chromatography on concanavalin A-Sepharose to obtain

glycoprotein fraction. The glycoprotein fraction was further

fractionated by immuno-affinity chromatography utilizing

monoclonal anti-saposin C antibody beads. Purified prosa-

posin preparation gave a single protein band with a molecular

weight of 66 kDa. Saposin C was purified from Gaucher’s

spleen [25] by procedures involving chloroform–methanol

extraction, preparative C4 reverse-phase high pressure liquid

chromatography (HPLC), DEAE-cellulose, and second C4

reverse-phase HPLC utilizing an analytical grade column.

The purities of purified prosaposin and saposin C (95–100%)

were assessed by sodium dodecyl sulphate polyacrilamide

gel electrophoresis (SDS-PAGE), immunoblotting, and N-

terminal analysis. Human saposin C was iodinated as already

described [7], radiolabeled saposin C showed specific acti-

vity of 72 cpm/pg. Nonradioactive iodide was used to label

saposin C by the same method. Iodinated milk prosaposin

was obtained using Iodo-Beads (Pierce) and following the

manufacturer’s instructions.

Low-density lipoproteins (LDLs) were purified from

plasma of healthy donors according to Frostegard et al.

[26]. Venous blood was drawn after overnight fasting into

precooled vacutainer tubes containing Na2EDTA (1 mg/ml).

Plasma was recovered with low-speed centrifugation

(1400 � g for 20 min) at 1jC and kept at this temperature

throughout the separation procedures. LDL was isolated

from plasma in the density interval 1.060–1.065 kg/l by

sequential preparative ultracentrifugation in a TLA-110

Beckman fixed-angle rotor (100,000 rpm, Beckman TL

100 ultracentrifuge) for 4 h at 4jC. The total protein content

of the LDL preparation was determined by the Lowry

technique. The LDL preparation was then desalted using

PD10 Desalting Columns (Amersham Pharmacia Biotech,

Sweden), following the manufacturer’s instructions. Apoli-

poprotein B-100 was isolated and quantified according to

Sparks [27].

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–4740

Mitogen-activated protein ERK kinase (MEK) inhibitor

PD98059 (2V-Amino-3V-methoxyflavone) [28], sphingosine

kinase inhibitor N-N dimethylsphingosine (DMS) [29],

PI3K specific inhibitor wortmannin [30], and pertussis

toxin (PT) were purchased from Calbiochem (La Jolla,

CA, USA).

125I-saposin C and prosaposin binding

U937 cells were rinsed twice with PBS, pH 7.4; binding

reactions were performed with 2 ml of 2 � 105 cells/ml cell

suspension in serum-free RPMI 1640 medium supple-

mented with insulin-transferrin (5mg/l), and the appropriate

amount of 125I-saposin C or 125I-prosaposin. The nonspe-

cific binding was in the presence of 100-fold molar excess

of unlabeled iodinated (nonradioactive) saposin C or prosa-

posin. After incubation for 3 h at 37jC, each tube was

centrifuged. An aliquot (100 Al) was taken from the super-

natant for the determination of free ligand concentration,

and the pellet was rinsed once with PBS. After lysis of the

cells performed by resuspending the pellet in NaOH 1 N, the

radioactivity was measured in a Beckmann g counter

(model 5500). Specific binding was determined as total

binding minus aspecific binding.

This experiment was repeated three times in duplicate.

Evaluation of cell death

Subconfluent U937 cells, washed in serum-free RPMI

1640 medium and incubated in the presence or absence of

prosaposin 1, 5, 10, or 50 nM in serum-free RPMI 1640

medium for 30 min, were treated with TNFa (Genzyme

Diagnostics, Cambridge MA, USA), 1000 IU/ml for 4 h at

37jC. In parallel experiments, cells, pretreated for 30 min

with PT (Recombinant holotoxin 100 ng/ml), were incubat-

ed with 50 nM prosaposin and then stimulated with TNFa

for 4 h. A Trypan blue dye exclusion test was performed to

evaluate the viability of the cultures [31]. A 0.2-ml aliquot

from each cell suspension was taken immediately after

treatment incubation and diluted 1:2 with 0.5% Trypan blue

solution. The viability of the cells was determined by

counting the number of stained or unstained cells on a

hemocytometer glass plate using an optical microscope.

Four hundred cells were scored for each sample; the

experiment was repeated five times in duplicate.

Apoptosis was measured by both morphologic and DNA

fragmentation analysis. DNA fragmentation was evaluated

according to Strauss [32], with slight modifications. U937

cells (5� 105) were suspended in 20 Al of 50 mM Tris–HCl,

pH 8.0, 10 mM EDTA, and 0.5 mg/ml of proteinase K

(Sigma, Saint Louis, USA). After incubation at 50jC for 2 h,

a 10-Al aliquot of 0.5 mg/ml RNase A solution was added

and the mixture was incubated for an additional 2 h. The

sample was mixed with 10 ml of preheated (70jC) 10 mM

EDTA solution, pH 8.0, containing 1% (w/v) low-melting-

point agarose (Sigma), 0.25% bromophenol blue, and 40%

sucrose. DNA was analyzed by electrophoresis in 2% aga-

rose gels followed by ethidium bromide staining and then

photographed on an ultraviolet (UV) illuminator.

In parallel experiments, cells were alternatively preincu-

bated in the presence of either 50 AM MEK inhibitor

PD98059 (for 30 min at 37jC), 3 AM sphingosine kinase

inhibitor DMS (for 30 min at 37jC), or 50 nM PI3K

inhibitor wortmannin (for 15 min at 37jC).Morphologic analysis of the nuclei was performed by

staining the cells with bisbenzimidetrihydrochloride

(Hoechst 33258, Sigma) [33], 5 Ag/ml in 30% glycerol/

PBS for 20 min. Cells were examined in an inverted

fluorescence microscope (320 nm UV excitation). Viable

cells were identified by their intact nuclei, and fragmented

or condensed nuclei were scored as apoptotic.

Analysis of ERKs activation

Cells were incubated with prosaposin (50 nM for 2 or

10 min at 37jC) or, as a positive control for ERK phosphor-

ylation [34], with phorbol ester myristate acetate (PMA)

(50 ng/ml for 2 min at 37jC) in serum-free RPMI 1640

medium. The cells were washed twice with ice-cold PBS. In

parallel experiments, the cells were incubated with 50 nM

prosaposin in serum-free RPMI 1640 medium in the presence

or absence of pretreatment with LDL (10 Ag/ml/106 cells),

apoB-100 (2 Ag/ml/106 cells) for 30 min at 4jC, or PT

(Recombinant holotoxin 100 ng/ml) for 30 min at 37jC.Cells were suspended in 1 ml of lysis buffer containing

1% Triton X-100, 10 mMTris–HCl (pH 8.0), 150 mMNaCl,

5 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM

sodium orthovanadate (NaVO4), and 75 U of aprotinin, and

allowed to stand for 20 min. The cell suspension was

mechanically disrupted by Dounce homogenization (10

strokes). Cell lysates, diluted in loading buffer, were resolved

in sodium dodecyl sulphate polyacrilamide gel electrophore-

sis (SDS-PAGE) under reducing conditions according to the

method of Laemmli [35] and proteins transferred electropho-

retically to nitrocellulosemembrane [36]. After blockingwith

PBS containing 3% albumin, the blots were incubated for 1

h with monoclonal anti-phospho-p44/42 mitogen-activated

protein (MAP) kinase [37] (New England Biolabs, Inc),

followed by horseradish peroxidase (HRP)-conjugated anti-

mouse IgG (Sigma). Immunoreactivity was assessed by

chemiluminescence using the enhanced chemiluminescence

(ECL) detection system (Amersham, Buckinghamshire, UK).

Manufacturer-specified protocols were used to strip the

membrane to reprobe with polyclonal anti-ERKs (K-23 Santa

Cruz Biotechnology), followed by HRP-conjugated anti-

mouse IgG.

In parallel samples, the lysate was centrifuged at

15,000 � g for 15 min at 4jC. After preclearing, cell-freelysates, normalized for proteins, were incubated overnight

with polyclonal anti-ERKs (K-23 Santa Cruz Biotechnolo-

gy) and then with protein G-sepharose beads. After centri-

fugation, the pellets resuspended in loading buffer were

Fig. 1. Binding of 125I-saposin C to U937 cells. (A) Binding reactions were

performed with 2 ml of 2 � 105 cells/ml cell suspension in serum-free

RPMI 1640 medium supplemented with insulin-transferrin (5 mg/l), and

0.2, 1, 2, 10, 50, 100 nM, or 1 AM 125I-saposin C. The nonspecific binding

was in the presence of 100-fold molar excess of unlabeled iodinated

(nonradioactive) saposin C. The radioactivity was measured in a g counter.

Specific binding was determined as total binding minus aspecific binding.

Binding of femtomoles of 125I-saposin C/106 cells is shown vs. free saposin

C concentration. These experiments were repeated three times in duplicate.

(B) Binding of 125I-saposin C (1, 2, 10, 50, 100 nM, or 1 AM) in a

concentration-dependent manner. (C) Scatchard analysis of the binding data

(1, 2, 10, 50, 100 nM, or 1 AM 125I-saposin C).

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–47 41

separated on 12% SDS-PAGE gels, under reducing condi-

tion, and proteins transferred electrophoretically to nitrocel-

lulose membrane. Nonspecific binding sites were blocked

with PBS containing 3% albumin for 1 h at room temper-

ature and the blots were incubated overnight with monoclo-

nal anti-phosphotyrosine antibody (Upstate Biotechnology,

Lake Placid, NY, USA), followed by horseradish peroxidase

(HRP)-conjugated anti-mouse IgG (Sigma). Immunoreac-

tivity was assessed by chemiluminescence using the ECL

detection system (Amersham). Manufacturer-specified pro-

tocols were used to strip the membrane to reprobe with

monoclonal anti-phospho-p44/42 MAP kinase (New Eng-

land Biolabs, Inc.), followed by HRP-conjugated anti-

mouse IgG. Densitometric scanning analysis was performed

by Mac OS 9.0 (Apple Computer International) using NIH

Image 1.62 software.

Sphingosine kinase assay

U937 cells (5 � 107), washed in serum-free RPMI 1640

medium, were incubated for 2, 5, 10, and 30 min in serum-

free RPMI 1640mediumwith or without 50 nM prosaposin at

37jC. In parallel experiments, cells, pretreated for 30 min

with PT (Recombinant holotoxin 100 ng/ml), were incubated

with 50 nM prosaposin. Then, the cells were washed with ice-

cold PBS and the cell sediment was lysed by freeze–thawing

in 20 mM MOPS (Sigma), pH 7.2, containing 200 mM

sucrose, 10 mM EDTA (Sigma), 10 mM EGTA (Sigma),

10 mM h-mercaptoethanol (Sigma), 1 mM phenylmethyl-

sulfonyl fluoride (Sigma), 0.0125% leupeptin (Sigma), and

0.5 mM 4-deoxypyridoxine. The cytosolic fractions were

prepared by ultracentrifugation at 105,000 � g for 60 min at

4jC. The protein concentration of supernatants was deter-

mined using Bio-Rad protein assay (Hercules, CA, USA).

The sphingosine kinase activity assay was performed as

previously described [14,38]. Briefly, U937 cytosolic

extracts (62 Ag) were incubated in 254-Al reaction buffer

containing 100 mM MOPS, pH 7.2, 60 mM MgCl2, 5%

glycerol, 5 mM h-mercaptoethanol, 51 mM h-octyl-gluco-side (Sigma), and 50 AM D-erythro-sphingosine (Sigma).

D-erythro-sphingosine was dried under a stream of nitrogen

from an ethanol solution and dissolved by sonication in

buffer (255 mM h-octyl-glucoside, 100 mM MOPS pH

7.2, 5% glycerol, and 5 mM h-mercaptoethanol). The

reaction was started by the addition of 10 Al of 5 mM

g-(32P)-ATP (Amersham, Bucks, UK), added to give a

specific activity 100,000 cpm/nmol. Assay tubes were

incubated at room temperature for 45 min and the reaction

was stopped by the addition of 1 ml of methanol/chloro-

form (2:1, v/v) containing 5% triethylamine (Sigma). S-1-P

was converted to N-caproyl-sphingosine-phosphate by the

addition of 20 Al of caproic anhydride (Sigma), followed

by incubation for 30 min at room temperature. Excess

caproic anhydride was removed by addition of 1 ml of 0.2

N methanolic NaOH for 30 min at room temperature. After

incubation, lipids were extracted by addition of 330 Al of

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–4742

methanol, 1.66 ml of choloform, 1 ml of 1% perchloric

acid solution, and 150 Al of 70% perchloric acid, and the

tubes were vortexed. After centrifuging, the lower phase

was washed twice with 2 ml of 1% perchloric acid

solution. The organic phases were dried under nitrogen

and resuspended in chloroform for thin layer chromatog-

raphy analysis. Sphingosine was resolved using Silica gel

60 F254 plates (Merck, Darmstadt, Germany) and butanol/

H2O/acetic acid (3:1:1, v/v/v) as a solvent system. N-

caproyl-sphingosine-1-phosphate (S-1-P) migrated with a

Rf = 0.47, and the corresponding radioactive spots were

visualized by autoradiography, scraped from plate, and

counted by liquid scintillation. Radioactive measurements

were converted to pmol product by using the specific

activity of g-(32P)-ATP.

Fig. 3. Prosaposin protective effect on TNFa-induced DNA laddering.

Subconfluent U937 cells, washed in serum-free RPMI 1640 medium and

incubated in the presence or absence of 50 nM prosaposin for 30 min, were

treated with TNFa, 1000 IU/ml for 4 h at 37jC. Electrophoresis was

performed in 2% agarose gels followed by ethidium bromide staining. (A)

Commercial DNA ladder; (B) untreated cells; (C) cells treated with 50 nM

prosaposin; (D) cells treated with TNFa; and (E) cells pretreated with

prosaposin and then with TNFa. The figure indicates a partial protection by

prosaposin of DNA degradation, although a DNA laddering was still

Results

Binding of 125I-saposin C to U937 cells

Because it is well-known that the neurotrophic activity of

prosaposin and its effect on the signaling transduction

pathway reside in the NH2-terminal sequence of saposin C

[8] and the putative prosaposin receptor was affinity purified

from brain using saposin C [11], we decided to perform the

binding studies of prosaposin to the cell surface using

saposin C, according to previous studies [7,11].

Fig. 2. Evaluation of cell death: a Trypan blue exclusion test was performed

to evaluate the viability of the cultures. Subconfluent U937 cells, washed in

serum-free RPMI 1640 medium, were treated with TNFa, 1000 IU/ml for

4 h at 37jC in the presence or absence of prosaposin (1, 5, 10, or 50 nM).

An additional sample pretreated with PT (Recombinant holotoxin 100 ng/

ml for 30 min at 37jC) was incubated with 50 nM prosaposin and then with

TNFa for 4 h. Mean of five experiments. *P < 0.01 when compared to cells

treated with TNFa for 4 h.

present. A representative example of three experiments.

Fig. 4. Prosaposin protective effect on TNFa-induced cell apoptosis,

detected by Hoechst 33258 staining. U937 cells, incubated in serum-free

RPMI 1640 medium in the presence or absence of 50 nM prosaposin for

30 min, were treated with TNFa, 1000 IU/ml for 4 h at 37jC. Morphological

analysis of U937 cell nuclei was stained with Hoechst 33258. The nuclei of

control cells were stained uniformly with this dye, indicating that the nuclei

were intact and the cells were viable (A). Treatment of cells with TNFa

caused nuclear fragmentation and condensation (B). Pretreatment of cells

with prosaposin prevented apoptosis (C). This effect was almost completely

inhibited by preincubation in the presence of 50 AMMEK inhibitor PD98059

(D), 3 AM sphingosine kinase inhibitor DMS (E), or 50 nM PI3K inhibitor

wortmannin (F). A representative example of five experiments.

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–47 43

After iodination, saposin C was used as a ligand to

perform binding studies. 125I-saposin C was found to bind

to U937 cell plasma membrane in a concentration-depen-

dent manner (Fig. 1A). Scatchard analysis of saposin C-

specific binding gave a straight line, indicating a single class

of specific binding sites (Figs. 1B and C), as previously

shown in other cell types [15], with an apparent Kd of 23.23

nM and a Bmax of 2199 fmol/106 cells.

The effects of incubation time and temperature on the

binding of prosaposin to U937 cells were determined. At

37jC, binding increased rapidly at 30 min compared to 0jCor 4jC, and reached a plateau between 30 and 60 min. As

expected, the binding of 125I-prosaposin was quite lower as

compared to that of 125I-saposin C (data not shown),

because prosaposin is not very stable toward protease

Fig. 5. Extracellular signal-regulated kinase (ERK) phosphorylation

induced by prosaposin in U937 cells. (A) Cells were treated for the

indicated times (2 and 10 min) with prosaposin (50 nM). The pellets of cell

lysates, resuspended in loading buffer, were resolved on 12% SDS-PAGE

under reducing conditions. The reactivity with monoclonal anti-phospho-

p44/42 MAP kinases was analyzed by immunoblotting. Bound antibodies

were visualized with HRP-conjugated anti-mouse IgG and immunoreac-

tivity assessed by chemiluminescence. Cell-free lysates from (a) control

cells; (b) cells stimulated with prosaposin for 2 min; (c) cells stimulated

with prosaposin for 10 min; and (d) cells stimulated with 50 ng/ml PMA for

2 min. A representative example of three experiments. (B) Cells were

treated for the indicated times (2 and 10 min) with prosaposin (50 nM). The

pellets of cell lysates, resuspended in loading buffer, were resolved on 12%

SDS-PAGE under reducing conditions. The reactivity with polyclonal anti-

ERKs was analyzed by immunoblotting. Bound antibodies were visualized

with HRP-conjugated anti-mouse IgG and immunoreactivity assessed by

chemiluminescence. Cell-free lysates from (a) control cells; (b) cells

stimulated with prosaposin for 2 min; (c) cells stimulated with prosaposin

for 10 min; and (d) cells stimulated with 50 ng/ml PMA for 2 min. A

representative example of three experiments. (C) Cells were treated with

prosaposin (50 nM) for 2 min in the presence or absence of pretreatment

with LDL (10 Ag/ml/106 cells), apoB-100 (2 Ag/ml/106 cells) for 30 min at

4jC, or PT (100 ng/ml) for 30 min at 37jC. The pellets of cell lysates,

resuspended in loading buffer, were resolved on 12% SDS-PAGE under

reducing conditions. The reactivity with monoclonal anti-phospho-p44/42

MAP kinases was analyzed by immunoblotting. Bound antibodies were

visualized with HRP-conjugated anti-mouse IgG and immunoreactivity

assessed by chemiluminescence. Cell-free lysates from (a) control cells; (b)

cells incubated with LDL; (c) cells stimulated with prosaposin for 2 min;

(d) cells incubated with LDL and then stimulated with prosaposin for 2 min;

(e) cells incubated with apoB-100; (f ) cells incubated with PT; (g) cells

pretreated with PT and then with prosaposin for 2 min; and (h) cells

pretreated with apoB-100 and then with prosaposin for 2 min. A

representative example of three experiments. (D) Cell-free lysates from

untreated or prosaposin-treated cells (2 and 10 min, 50 nM) were

immunoprecipitated with polyclonal anti-ERKs and then with protein G-

sepharose beads. The mixtures were centrifuged and washed three times

with 0.4 ml of the RIPA buffer. The pellets resuspended in loading buffer

were resolved on 12% SDS-PAGE under reducing conditions and

immunoreactivity with anti-phosphotyrosine MoAb was assessed as above.

(a) ERK immunoprecipitates from control cells; (b) ERK immunoprecipi-

tates from cells stimulated with prosaposin for 2 min; (c) ERK

immunoprecipitates from cells stimulated with prosaposin for 10 min;

and (d) immunoprecipitates by anti-mouse IgG with irrelevant specificity

from cells stimulated with prosaposin for 10 min. A representative example

of three experiments.

activities. There are nine tyrosine residues in the prosaposin

molecule that can be labeled with iodine, only one of which

is present in the domain for saposin C. The others locate far

away from the trophic sequence. Thus, the radioactivity of

prosaposin may be lost to the binding medium, causing a

low specific activity and apparently quite low binding.

Prosaposin effect on TNFa induced cell death

To determine whether prosaposin prevented cell death,

we used as a first approach a Trypan blue exclusion test.

U937 cells were incubated with TNFa for 4 h either in the

presence or absence of prosaposin; 400 cells were scored

for each sample. The results demonstrated that TNFa-

induced cell death was inhibited by incubation with 1, 5,

10, or 50 nM prosaposin for 30 min (Fig. 2). The prosaposin

death prevention was substantial at 50-nM concentration,

but this effect was partially inhibited by preincubation with

PT (Fig. 2).

To verify whether prosaposin prevented apoptosis in

these cells, the effect of TNFa on DNA fragmentation

was evaluated in the presence or absence of 50 nM

prosaposin for 30 min. As expected, electrophoresis in 2%

agarose gels followed by ethidium bromide staining

revealed that cell treatment with TNFa, 1000 IU/ml for 4

h, induced DNA fragmentation, consistent with apoptosis.

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–4744

In the presence of prosaposin, although a DNA laddering

was still present, a very consistent portion of native DNA

was detectable, compared to the TNFa-treated sample,

indicating a partial protection by prosaposin of DNA deg-

radation (Fig. 3).

To confirm these findings, cells were stained with

Hoechst 33258 (Fig. 4); nuclei of control U937 cells stained

uniformly with this dye, indicating that the nuclei were

intact and the cells were viable (Fig. 4A). As expected,

treatment of cells with TNFa caused nuclear fragmentation

and condensation (Fig. 4B). In cells treated with TNFa, in

the presence of prosaposin, a decrease of apoptotic cells was

observed (Fig. 4C). This protective effect of prosaposin was

partially inhibited by previous preincubation with 50 AMMEK inhibitor PD98059 (Fig. 4D), 3 AM sphingosine

kinase inhibitor DMS (Fig. 4E), or 50 nM PI3K inhibitor

wortmannin (Fig. 4F).

Prosaposin induces ERKs phosphorylation in U937 cells

To investigate whether activation of ERKs might be an

early event following prosaposin treatment, serum-starved,

subconfluent U937 cells were treated with 50 nM prosapo-

sin and then cell-free lysates were probed with anti-phos-

pho-p44/42 MAP kinase. The results clearly indicated that

prosaposin stimulated both ERK-1 and ERK-2 phosphory-

lation. It was evident as early as after 2 min of incubation

with prosaposin and the activity increased after 10 min

(about 5-fold above basal levels) (Fig. 5A). This prosaposin-

induced phosphorylation was completely abolished by pre-

vious incubation (30 min) with the synthetic MEK inhibitor

PD98059 (data not shown), which is known to specifically

prevent MEK-1 activation without affecting the activity of

other kinases [25]. After stripping of the membrane, the

polyclonal anti-ERKs, which is reactive with both ERK-1

Fig. 6. Prosaposin-induced activation of sphingosine kinase and S-1-P generation. C

The samples at 10 and 30 min were pretreated with PT (100 ng/ml) for 30 min at 3

in lysis buffer as described in Materials and methods. Cytosolic fractions were

incubating 50 AM sphingosine-h-octyl glucoside and g-32P-ATP for 60 min at roo

kinase induced by prosaposin. A representative example of three experiments. SD

and ERK-2, immunostained the 44–42 bands in both

prosaposin-treated and -untreated cells (Fig. 5B). In parallel

experiments, ERK phosphorylation by prosaposin was par-

tially prevented by previous incubation (30 min at 37jC) ofthe cells with LDL, as well as apoB-100, which binds to the

LRP [39], or with PT (Fig. 5C). This finding strongly

suggests that both the LRP receptor and the Go-coupled

receptor may play a role in the prosaposin-triggered path-

way leading to ERK phosphorylation.

ERK phosphorylation by prosaposin was confirmed by a

different approach. U937 cells were incubated for 2 or 10 min

in the presence or absence of prosaposin and then lysed as

reported above and immunoprecipitated with the polyclonal

anti-ERK antibody. Western blot analysis of these immuno-

precipitates, performed using anti-phosphotyrosine antibody,

demonstrated that prosaposin treatment induced a significant

ERK phosphorylation (Fig. 5D).

Sphingosine kinase activation and sphingosine-1-phosphate

generation after prosaposin treatment in U937 cells

Sphingosine kinase activity has been proposed to be the

rate-limiting step in the metabolism of sphingosine [40,41].

Thus, we analyzed the effect of prosaposin on sphingosine

kinase activation and sphingosine-1-phosphate generation

in U937 cells. Treatment with prosaposin increased sphin-

gosine kinase activity in these cells. When cells were

treated with prosaposin at different incubation times (0,

2, 5, 10, or 30 min), N-caproyl-S-1-P production was

evident as early as after 2 min of incubation with prosa-

posin and a peak of sphingosine kinase activation occurred

by 10 min (0.051 pmol of N-caproyl-S-1-P as compared to

0.02 in control cells). When cells were preincubated with

PT, stimulation of sphingosine kinase by prosaposin was

inhibited (Fig. 6).

ells were treated with prosaposin (50 nM) for 2, 5, 10, and 30 min at 37jC.7jC. After stimulation, the cells were washed and lysed by freeze– thawing

prepared and sphingosine kinase activity measured in the supernatant by

m temperature. Pretreatment with PT prevents the activation of sphingosine

< 1% mean; *P < 0.05 when compared to control cells.

Fig. 7. (A) Human PBMCs were treated for the indicated time (2 min) with

prosaposin (50 nM). The pellets of cell lysates, resuspended in loading

buffer, were resolved on 12% SDS-PAGE under reducing conditions. The

reactivity with monoclonal anti-phospho-p44/42 MAP kinases was

analyzed by immunoblotting. Bound antibodies were visualized with

HRP-conjugated anti-mouse IgG and immunoreactivity assessed by

chemiluminescence. Cell-free lysates from (a) control cells and (b) cells

stimulated with prosaposin for 2 min. A representative example of three

experiments. (B) Human PBMCs were treated for the indicated times

(2 min) with prosaposin (50 nM). The pellets of cell lysates, resuspended in

loading buffer, were resolved on 12% SDS-PAGE under reducing

conditions. The reactivity with polyclonal anti-ERKs was analyzed by

immunoblotting. Bound antibodies were visualized with HRP-conjugated

anti-mouse IgG and immunoreactivity assessed by chemiluminescence.

Cell-free lysates from (a) control cells and (b) cells stimulated with

prosaposin for 2 min. A representative example of three experiments.

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–47 45

Prosaposin induces ERKs phosphorylation in peripheral

blood mononuclear cells (PBMC)

To verify whether the biological effect triggered by

prosaposin might be extended to primary cells, we ana-

lyzed activation of ERKs in PBMC. Cells were treated

with 50 nM prosaposin and then cell-free lysates were

probed with anti-phospho-p44/42 MAP kinase. Again, the

results clearly indicated that prosaposin stimulated both

ERK-1 and ERK-2 phosphorylation as early as after 2 min

of incubation (Fig. 7A). The polyclonal anti-ERKs immu-

nostained the 44–42 bands in both prosaposin-treated and

-untreated cells (Fig. 7B).

Discussion

This study demonstrates that prosaposin is active as a

protective factor on TNFa-induced cell death in U937 cells.

Several studies suggested a pivotal role for prosaposin in

development not only in brain, but also in other cell systems

[42,43]. Until now, prosaposin has been considered a

trophic factor active specifically on neuronal-derived cells.

Although it was detected in a large variety of biological

fluids, as far as we know, no effects of this protein in other

cell systems have been demonstrated. Our study is the first

evidence of a biological effect consequent to a signal

transduction pathway triggered by prosaposin in cells of

non-neurological origin. It indicates that the hematopoietic

system may represent a key and unexplored model for

evaluating the meaning of prosaposin as a molecule in-

volved in signal transduction pathways leading to develop-

ment, cell proliferation, and apoptosis prevention.

Because in neuronal cells prosaposin exerts its biological

activity by binding to a putative high affinity receptor [11]

that is associated with a G-protein G0a, we primarily

evaluated the capacity of this protein to bind to U937 cells.

Our results revealed that prosaposin binds to U937 cell

plasma membrane in a concentration-dependent manner

with a lower number of sites per cell as compared to cells

of neurological origin [15].

To clarify the biological meaning of the binding of

prosaposin to U937 plasma membrane, we investigated its

effect on cell death induced by TNFa. Our findings,

obtained by DNA fragmentation and morphological analysis

after Hoechst 33258 staining, indicated that prosaposin

protected U937 cells from death. However, it is important

to consider that prosaposin is able to prevent only partially

the programmed cell death induced by TNFa, as shown by

DNA fragmentation analysis in which a significant portion

of native DNA was ‘‘rescued’’ in cells pretreated with

prosaposin. The prosaposin effect on U937 cells is consis-

tent with the observations that prosaposin addition rescues

cells from death after serum deprivation in neuroblastoma

cells [7,8], primary hippocampal neurons [44], and Schwann

cells [17], and rescues PC12 pheochromocytoma cells from

apoptosis induced by different agents [14]. One possible

molecular mechanism may be that prosaposin may activate

the ERK pathway [14–17].

Thus, it was of interest to analyze the molecular signals

triggered by prosaposin in U937 cells. Our findings indicated

that prosaposin treatment led to rapid ERK phosphorylation;

this effect was evident for both ERK-1 and ERK-2. Although

a direct stimulation of ERKs by prosaposin is unlikely, the

observation that cell incubation with the MEK-1 inhibitor

PD98059 prevented ERK phosphorylation, as well as pre-

vention of cell apoptosis, strongly suggests that prosaposin

stimulates a signaling cascade involving MEK-1 or a MAP

kinase kinase-related protein that subsequently activates

ERKs. To verify whether binding of prosaposin to the

putative Go protein-coupled receptor or to the LRP receptor

may be functional in these cells, we analyzed ERK phos-

phorylation after incubation of the cells with either PT or the

LRP ligand apoB-100. Interestingly, ERK phosphorylation

by prosaposin was partially prevented by previous incuba-

tion of the cells with both compounds, indicating that in

these cells, both receptors may play a role in the prosaposin-

triggered pathway leading to ERK phosphorylation. These

findings are fully in agreement with the observation of

Hiesberger et al. [12], who observed that the LRP, a

multifunctional endocytic receptor that is expressed in most

cells, can mediate cellular uptake and lysosomal delivery of

prosaposin. The binding of prosaposin with multiple unre-

lated cell surface receptors, such as LRP, may be explained

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–4746

with the multiple role of prosaposin as a signaling molecule

involved in different cell functions, including (neuro)trophic

activity.

Another possible mechanism may be modulation of the

ceramide–S-1-P pathway, which may play a regulatory

effect on mitogenic or apoptotic effects. Indeed, recent

evidence has suggested that branching pathways of sphin-

golipid metabolism may mediate either mitogenic or apo-

ptotic signaling cascade. It has been reported that ceramide

induced apoptosis in several cell lines [45,46], that sphin-

gosine and S-1-P are mitogenic [47], and that both

stimulate the activation of the ERKs pathway by a G-

protein-coupled receptor, the S-1-P receptor [48]. Thus,

both ceramide and S-1-P may be considered key signaling

molecules involved in cell fate [49,50]. In our cell system,

prosaposin enhanced sphingosine kinase and led to intra-

cellular formation of S-1-P. Our findings strongly suggest

that these pathways may be involved in the observed cell

death prevention by prosaposin. This hypothesis is sup-

ported by the observation that both the MEK-1 and

sphingosine kinase inhibitor partially prevented the protec-

tive prosaposin effect on TNFa-induced apoptosis in U937

cells. A hypothetical pathway by which sphingosine kinase

may exert its anti-apoptotic activity is represented by the

PI3K–Akt pathway, as suggested by the observation that

also the PI3K inhibitor wortmannin prevented the prosa-

posin effect on cell apoptosis. Indeed, S-1-P activates c-

Srk tyrosine kinases and promotes Grb2-PI3K complex

formation.

In conclusion, this paper deals with cell death prone-

ness. Hence, understanding the regulation pathways super-

vising both cell proliferation and, on the opposite side, cell

death by apoptosis could provide useful information on the

subcellular mechanisms influencing cell fate. It is well

known that apoptosis controls cell differentiation and cell

numbers homeostatically, thus having a role in the organ-

ogenesis during development and in the elimination of

autoreactive cells in the immune system. In this concern,

our work is the first evidence that prosaposin is an

additional molecule involved in the apoptotic machinery.

Because the biological effect triggered by prosaposin may

be extended to primary cells, as demonstrated by the

activation of ERKs in PBMC, prosaposin could be con-

sidered a new player in the regulation of the apoptotic

signal transduction pathway in cells of non-neurological

origin.

Acknowledgments

This paper is dedicated in memoriam to John S.

O’Brien, M.D., Professor of Neurosciences, University of

California San Diego. He planned the research reported in

this paper. We are greatly indebted to him for helping to

establish an outstanding scientific foundation for work in

this field.

We thank Prof. Roberto Strom for help in LDL

preparations and Dr Sergio Scaccianoce and Dr Paola Del

Bianco for precious suggestions.

References

[1] J.S. O’Brien, K.A. Kretz, N.N. Dewji, D.A. Wenger, F. Esch, A.L.

Fluharty, Coding of two sphingolipid activator proteins (SAP-1 and

SAP-2) by same genetic locus, Science 241 (1988) 1098–1101.

[2] J.S. O’Brien, Y. Kishimoto, Saposin proteins: structure, function and

role in human lysosomal storage disorders, FASEB J. 5 (1991)

301–308.

[3] M. Hiraiwa, J.S. O’Brien, Y. Kishimoto, M. Galdzicka, A.L. Fluharty,

E.I. Ginns, B.M. Martin, Isolation, characterization, and proteolysis of

human prosaposin, the precursor of saposins (Sphingolipid Activator

Proteins), Arch. Biochem. Biophys. 304 (1993) 110–116.

[4] T. Hineno, A. Sano, K. Kondoh, S. Ueno, Y. Kakimoto, K. Yoshida,

Secretion of sphingolipid hydrolase activator precursor, prosaposin,

Biochem. Biophys. Res. Commun. 176 (1991) 668–674.

[5] Q. Fu, G.F. Carson, M. Hiraiwa, M. Grafe, Y. Kishimoto, J.S.

O’Brien, Occurrence of prosaposin as a neuronal surface membrane

component, J. Mol. Neurosci. 5 (1994) 59–67.

[6] R. Misasi, M. Sorice, T. Garofalo, T. Griggi, W.M. Campana, M.

Giammatteo, A. Pavan, M. Hiraiwa, G.M. Pontieri, J.S. O’Brien,

Colocalization and complex formation between prosaposin and

monosialoganglioside GM3 in neural cells, J. Neurochem. 71

(1998) 2313–2321.

[7] J.S. O’Brien, G. Carson, H.C. Seo, M. Hiraiwa, Y. Kishimoto, Iden-

tification of prosaposin as a neurotrophic factor, Proc. Natl. Acad. Sci.

U. S. A. 91 (1994) 9593–9596.

[8] J.S. O’Brien, G.S. Carson, H.C. Seo, M. Hiraiwa, S. Weiler, J.M.

Tomich, J.A. Barranger, M. Kahn, N. Azuma, Y. Kishimoto, Identi-

fication of the neurotrophic factor sequence of prosaposin, FASEB J.

9 (1995) 681–685.

[9] X. Qi, W. Qin, Y. Sun, K. Kondoh, G. Grabowski, Functional

organization of saposin C. Definition of the neurotrophic and acid

h-glucosidase activation regions, J. Biol. Chem. 271 (1996)

6874–6880.

[10] Y. Kotani, S. Matsuda, M. Sakanaka, K. Kondoh, S. Ueno, A. Sano,

Prosaposin facilitates sciatic nerve regeneration in vivo, J. Neuro-

chem. 66 (1996) 2019–2025.

[11] M. Hiraiwa, W.M. Campana, B.M. Martin, J.S. O’Brien, Prosaposin

receptor: evidence for a G-protein-associated receptor, Biochem. Bio-

phys. Res. Commun. 240 (1997) 415–418.

[12] T. Hiesberger, S. Huttler, A. Rohlmann, W. Schneider, K. Sandoff,

J. Herz, Cellular uptake of saposin (SAP) precursor and lysosomal

delivery by the low density lipoprotein receptor-related protein

(LRP), EMBO J. 17 (1998) 4617–4625.

[13] V. Laurent-Matha, A. Lucas, S. Huttler, K. Sandhoff, M. Garcia, H.

Rochefort, Procathepsin D interacts with prosaposin in cancer cells

but its internalization has not mediated by LDL-receptor mediated

protein, Exp. Cell Res. 277 (2002) 210–219.

[14] R. Misasi, M. Sorice, L. Di Marzio, W.M. Campana, S. Molinari,

M.G. Cifone, A. Pavan, G.M. Pontieri, J.S. O’Brien, Prosaposin treat-

ment induces PC12 entry in the S phase of the cell cycle and prevents

apoptosis: activation of ERK’s and sphingosine kinase, FASEB J. 15

(2001) 467–474.

[15] W.M. Campana, M. Hiraiwa, K.C. Addison, J.S. O’Brien, Induction

of MAPK phosphorylation by prosaposin and prosaptide in PC12

cells, Biochem. Biophys. Res. Commun. 229 (1996) 706–712.

[16] W.M. Campana, M. Hiraiwa, J.S. O’Brien, Prosaptide activates the

MAPK pathway by a G-protein-dependent mechanism essential for

enhanced sulfatide synthesis by Schwann cells, FASEB J. 12 (1998)

307–314.

[17] M. Hiraiwa, E.M. Taylor, W.M. Campana, S.J. Darin, J.S. O’Brien,

R. Misasi et al. / Experimental Cell Research 298 (2004) 38–47 47

Cell death prevention, mitogen-activated protein kinase stimulation,

and increased sulfatide concentrations in Schwann cells and oligoden-

drocytes by prosaposin and prosaptides, Proc. Natl. Acad. Sci. U. S. A.

94 (1997) 4778–4781.

[18] M. Hiraiwa, W.M. Campana, A.P. Mizisin, L. Mohiuddin, J.S.

O’Brien, Prosaposin: a myelinotrophic protein that promotes ex-

pression of myelin constituents and is secreted after nerve in-

jury, Glia 26 (1999) 353–360.

[19] W.M. Campana, S.J. Darin, J.S. O’Brien, Phosphatidylinositol 3-

kinase and Akt protein kinase mediate IGF-I- and prosaptide-induced

survival in Schwann cells, J. Neurosci. Res. 57 (1999) 332–341.

[20] Y. Osawa, Y. Banno, M. Nagaki, D.A. Brenner, T. Naiki, Y. Nozawa,

S. Nakashima, H. Moriwaki, TNFa-induced sphingosine-1-phosphate

inhibits apoptosis through a phosphatidylinositol 3-kinase/Akt path-

way in human hepatocytes, J. Immunol. 167 (2001) 173–180.

[21] K. Schulze-Osthoff, P.H. Krammer, W. Droge, Divergent signalling

via APO-1/Fas and the TNF receptor, two homologous molecules

involved in physiological cell death, EMBO J. 13 (1994) 4587–4596.

[22] S. Nagata, Apoptosis by death factor, Cell 88 (1997) 355–365.

[23] P. Xia, L. Wang, J.R. Gamble, M.A. Vadas, Activation of sphingosine

kinase by tumor necrosis factor a inhibits apoptosis in human endo-

thelial cells, J. Biol. Chem. 274 (1999) 34499–34505.

[24] C. Sundstromm, K. Nilsson, Establishment and characterization of a

human histiocytic lymphoma cell line (U-937), Int. J. Cancer 17

(1976) 565–576.

[25] S. Morimoto, Y. Yamamoto, J.S. O’Brien, Y. Kishimoto, Distribution

of saposin proteins (sphingolipid activator proteins) in lysosomal sto-

rage and other diseases, Proc. Natl. Acad. Sci. U. S. A. 87 (1990)

3493–3497.

[26] J. Frostegard, J. Nilsson, A. Haegerstrand, A. Hamsten, H. Wigzell,

M. Gidlund, Oxidated low density lipoprotein induces differentiation

and adhesion of human monocytes and the monocytic cell line U937,

Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 904–908.

[27] J.D. Sparks, C.E. Sparks, Chromatographic method for isolation and

quantification of apolipoproteins B-100 and B-48, Methods Enzymol.

263 (1996) 104–120.

[28] D.R. Alessi, A. Cuenda, P. Cohen, D.T. Dudley, A.R. Saltiel,

PD098059 is a specific inhibitor of the activation of mitogen activat-

ed protein kinase in vitro and in vivo, J. Biol. Chem. 270 (1995)

27489–27494.

[29] L.C. Edsall, J.R. Van Brocklyn, O. Cuvillier, B. Kleuser, S. Spiegel,

N,N-dimethylsphingosine is a potent competitive inhibitor of sphin-

gosine kinase but not of protein kinase C: modulation of cellular

levels of sphingosine-1-phosphate and ceramide, Biochemistry 37

(1998) 12892–12898.

[30] H. Yano, S. Nakanishi, K. Kimura, N. Hanai, Y. Saitoh, Y. Fukui, Y.

Nonomura, Y. Matsuda, Inhibition of histamine secretion by wort-

mannin through the blockade of phosphatidylinositol 3-kinase in

RBL-2H3 cells, J. Biol. Chem. 268 (1993) 25846–25856.

[31] J.E. Coligan, A.M. Kruisbeek, D.H. Margulies, E.M. Shevach, W.

Strober, Trypan blue exclusion test of cell viability, in: John Wiley

and Sons (Eds.), Current Protocols in Immunology, Greene Publi-

shing, Associates and Wiley-Interscience, USA, vol. 2, A.3.3.

[32] W.M. Strauss, Preparation of genomic DNA from mammalian tissue,

in: F. Ausundel et al. (Eds.), Curr. Protoc. Mol. Biol., vol. 2, Wiley,

New York, NY, 1992, pp. 1–3, section 2.

[33] O. Cuivillier, D.S. Rosenthal, M.E. Smulson, S. Spiegel, Sphingosine

1-phosphate inhibits activation of caspases that cleave poly(ADP-

ribose)polymerase and lamins during Fas- and ceramide-mediated

apoptosis in Jurkat T lymphocytes, J. Biol. Chem. 273 (1998)

2910–2916.

[34] B. Marquardt, D. Frith, S. Stabel, Signalling from TPA to MAP kinase

requires protein kinase C, raf and MEK: reconstitution of the signal-

ling pathway in vitro, Oncogene 9 (1994) 3213–3218.

[35] U.K. Laemmli, Cleavage of structural proteins during the assembly of

the head of bacteriophage T4, Nature 227 (1970) 680–685.

[36] H. Towbin, T. Staehelin, J. Gordon, Electrophoretic transfer of pro-

teins from polyacrilamide gels to nitrocellulose sheets: procedure

and some applications, Proc. Natl. Acad. Sci. U. S. A. 76 (1979)

4350–4354.

[37] C.J. Marshall, Specificity of receptor tyrosine kinase signaling: tran-

sient versus sustained extracellular signal-regulated kinase activation,

Cell 80 (1995) 179–185.

[38] B.M. Buehrer, R.M. Bell, Inhibition of sphingosine kinase in vitro

and in platelets. Implications for signal transduction pathways, J. Biol.

Chem. 267 (1992) 3154–3159.

[39] M. Krieger, J. Herz, Structures and functions of multiligand

lipoprotein receptors: macrophage scavenger receptors and LDL

receptor-related protein (LRP), Annu. Rev. Biochem. 63 (1994)

601–637.

[40] A. Olivera, S. Spiegel, Sphingosine kinase: a mediator of vital cellular

function, Prostaglandins Other Lipid Mediators 64 (2001) 123–134.

[41] A. Olivera, T. Kohama, L. Edsall, V. Nava, O. Cuvillier, S. Poulton, S.

Spiegel, Sphingosine kinase expression increases intracellular sphin-

gosine-1-phosphate and promotes cell growth and survival, J. Cell

Biol. 147 (1999) 545–558.

[42] M.W. Collard, S.R. Sylvester, J.K. Tsuruta, M.D. Griswold, Biosyn-

thesis and molecular cloning of sulfated glycoprotein 1 secreted by rat

Sertoli cells: sequence similarity with the 70-kilodalton precursor to

sulfatide/GM1 activator, Biochemistry 27 (1988) 4557–4564.

[43] A. Sano, T. Hineno, T. Mizuno, K. Kondoh, S. Ueno, Y. Kakimoto, K.

Inui, Sphingolipid hydrolase activator proteins and their precursors,

Biochem. Biophys. Res. Commun. 165 (1989) 1191–1197.

[44] A. Sano, S. Matsuda, T.C. Wen, Y. Kotani, K. Kondoh, S. Ueno, Y.

Kakimoto, H. Yoshimura, N. Sakanaka, Protection by prosaposin

against ischemia-induced learning disability and neuronal loss, Bio-

chem. Biophys. Res. Commun. 204 (1994) 994–1000.

[45] Y.A. Hannun, L.M. Obeid, Ceramide: an intracellular signal for apo-

ptosis, Trends Biochem. Sci. 20 (1995) 73–77.

[46] S. Spiegel, D. Foster, R. Kolesnick, Signal transduction through lipid

second messengers, Curr. Opin. Cell Biol. 8 (1996) 159–167.

[47] S. Spiegel, A.H. Merril Jr., Sphingolipid metabolism and cell growth

regulation, FASEB J. 10 (1996) 1388–1397.

[48] S. An, Y. Zheng, T. Bleu, Sphingosine 1-phosphate-induced cell pro-

liferation, survival, and related signaling events mediated by G pro-

tein-coupled receptors Edg3 and Edg5, J. Biol. Chem. 275 (2000)

288–296.

[49] Y.A. Hannun, L.M. Obeid, The ceramide-centric universe of lipid-

mediated cell regulation: stress encounters of the lipid kind, J. Biol.

Chem. 277 (2002) 25847–25850.

[50] S. Spiegel, S.J. Milstien, Sphingosine 1-phosphate, a key cell signal-

ing molecule, J. Biol. Chem. 277 (2002) 25851–25854.