Portal de Periódicos da CAPES

Calcium Influx and Signaling in Yeast Stimulated by Intracellular Sphingosine 1-Phosphate Accumulation

2001; Elsevier BV; Volume: 276; Issue: 15 Linguagem: Inglês

10.1074/jbc.m010221200

1083-351X

Christine J. Birchwood, Julie D. Saba, Robert C. Dickson, Kyle W. Cunningham,

Endoplasmic Reticulum Stress and Disease

In mammalian cells, intracellular sphingosine 1-phosphate (S1P) can stimulate calcium release from intracellular organelles, resulting in the activation of downstream signaling pathways. The budding yeast Saccharomyces cerevisiaeexpresses enzymes that can synthesize and degrade S1P and related molecules, but their possible role in calcium signaling has not yet been tested. Here we examine the effects of S1P accumulation on calcium signaling using a variety of yeast mutants. Treatment of yeast cells with exogenous sphingosine stimulated Ca2+accumulation through two distinct pathways. The first pathway required the Cch1p and Mid1p subunits of a Ca2+ influx channel, depended upon the function of sphingosine kinases (Lcb4p and Lcb5p), and was inhibited by the functions of S1P lyase (Dpl1p) and the S1P phosphatase (Lcb3p). The biologically inactive stereoisomer of sphingosine did not activate this Ca2+ influx pathway, suggesting that the active S1P isomer specifically stimulates a calcium-signaling mechanism in yeast. The second Ca2+influx pathway stimulated by the addition of sphingosine was not stereospecific, was not dependent on the sphingosine kinases, occurred only at higher doses of added sphingosine, and therefore was likely to be nonspecific. Mutants lacking both S1P lyase and phosphatase (dpl1 lcb3 double mutants) exhibited constitutively high Ca2+ accumulation and signaling in the absence of added sphingosine, and these effects were dependent on the sphingosine kinases. These results show that endogenous S1P-related molecules can also trigger Ca2+ accumulation and signaling. Several stimuli previously shown to evoke calcium signaling in wild-type cells were examined in lcb4 lcb5 double mutants. All of the stimuli produced calcium signals independent of sphingosine kinase activity, suggesting that phosphorylated sphingoid bases might serve as messengers of calcium signaling in yeast during an unknown cellular response. In mammalian cells, intracellular sphingosine 1-phosphate (S1P) can stimulate calcium release from intracellular organelles, resulting in the activation of downstream signaling pathways. The budding yeast Saccharomyces cerevisiaeexpresses enzymes that can synthesize and degrade S1P and related molecules, but their possible role in calcium signaling has not yet been tested. Here we examine the effects of S1P accumulation on calcium signaling using a variety of yeast mutants. Treatment of yeast cells with exogenous sphingosine stimulated Ca2+accumulation through two distinct pathways. The first pathway required the Cch1p and Mid1p subunits of a Ca2+ influx channel, depended upon the function of sphingosine kinases (Lcb4p and Lcb5p), and was inhibited by the functions of S1P lyase (Dpl1p) and the S1P phosphatase (Lcb3p). The biologically inactive stereoisomer of sphingosine did not activate this Ca2+ influx pathway, suggesting that the active S1P isomer specifically stimulates a calcium-signaling mechanism in yeast. The second Ca2+influx pathway stimulated by the addition of sphingosine was not stereospecific, was not dependent on the sphingosine kinases, occurred only at higher doses of added sphingosine, and therefore was likely to be nonspecific. Mutants lacking both S1P lyase and phosphatase (dpl1 lcb3 double mutants) exhibited constitutively high Ca2+ accumulation and signaling in the absence of added sphingosine, and these effects were dependent on the sphingosine kinases. These results show that endogenous S1P-related molecules can also trigger Ca2+ accumulation and signaling. Several stimuli previously shown to evoke calcium signaling in wild-type cells were examined in lcb4 lcb5 double mutants. All of the stimuli produced calcium signals independent of sphingosine kinase activity, suggesting that phosphorylated sphingoid bases might serve as messengers of calcium signaling in yeast during an unknown cellular response. Sphingolipid metabolites, such as ceramide, sphingosine, and S1P, 1The abbreviations used are: S1Psphingosine 1-phosphate[Ca2+]ccytosolic-free Ca2+ concentrationCCEcapacitative calcium entrySCsynthetic completeIP3inositol 1,4,5-trisphosphateYPDyeast extract/peptone/dextroseBAPTA1,2-bis(2-aminophenoxy)ethane-N, N,N′, N′-tetraacetic acid function as important second messengers in mammalian cells mediating processes such as cell proliferation and motility, differentiation, senescence, stress responses, and apoptosis (1Spiegel S. J. Leukocyte Biol. 1999; 65: 341-344Crossref PubMed Scopus (160) Google Scholar). S1P accumulates in response to various physiological stimuli in mammals. In RBL-2H3 cells, for example, the clustering of the IgE receptor FcεRI activates sphingosine kinase, resulting in S1P production. Inhibitors of sphingosine kinases block the agonist-stimulated accumulation of S1P and also suppress the normal mobilization of Ca2+ stored in the endoplasmic reticulum (2Choi O.H. Kim J.H. Kinet J.P. Nature. 1996; 380: 634-636Crossref PubMed Scopus (384) Google Scholar). In permeabilized cells, S1P also triggers the release of Ca2+ through a mechanism independent of the known Ca2+ release pathways, suggesting that S1P activates a novel type of a Ca2+ release channel (3Mattie M. Brooker G. Spiegel S. J. Biol. Chem. 1994; 269: 3181-3188Abstract Full Text PDF PubMed Google Scholar, 4Ghosh T.K. Bian J. Gill D.L. J. Biol. Chem. 1994; 269: 22628-22635Abstract Full Text PDF PubMed Google Scholar, 5Ghosh T.K. Bian J. Gill D.L. Science. 1990; 248: 1653-1656Crossref PubMed Scopus (327) Google Scholar) capable of elevating cytosolic-free Ca2+concentrations ([Ca2+]c) and stimulating capacitative Ca2+ entry (CCE) mechanisms. The hypothetical intracellular S1P receptor and/or Ca2+ release channel has not yet been identified. sphingosine 1-phosphate cytosolic-free Ca2+ concentration capacitative calcium entry synthetic complete inositol 1,4,5-trisphosphate yeast extract/peptone/dextrose 1,2-bis(2-aminophenoxy)ethane-N, N,N′, N′-tetraacetic acid Sphingolipids are abundant components of the plasma membrane in yeast, comprising 30% of total membrane phospholipids (6Patton J.L. Lester R.L. J. Bacteriol. 1991; 173: 3101-3108Crossref PubMed Google Scholar). They differ slightly from mammalian sphingolipids, using a derivative of sphingosine known as phytosphingosine. The enzymes in yeast responsible for phosphorylation of endogenous phytosphingosine and exogenous long chain bases such as sphingosine have recently been identified. Two related sphingosine kinases were identified in yeast as the products of the LCB4 and LCB5 genes (7Nagiec M.M. Skrzypek M. Nagiec E.E. Lester R.L. Dickson R.C. J. Biol. Chem. 1998; 273: 19437-19442Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar, 8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar). Mutants lacking both sphingosine kinases accumulate no detectable S1P-related molecules but, nevertheless, are viable and exhibit no obvious phenotypes (7Nagiec M.M. Skrzypek M. Nagiec E.E. Lester R.L. Dickson R.C. J. Biol. Chem. 1998; 273: 19437-19442Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar, 9Kim S. Fyrst H. Saba J. Genetics. 2000; 156: 1519-1529Crossref PubMed Google Scholar). Yeast also expresses two related S1P phosphatases encoded by theLCB3/YSR2 and YSR3 genes, the former being the major enzyme (10Qie L. Nagiec M.M. Baltisberger J.A. Lester R.L. Dickson R.C. J. Biol. Chem. 1997; 272: 16110-16117Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar, 11Mandala S.M. Thornton R. Tu Z. Kurtz M.B. Nickels J. Broach J. Menzeleev R. Spiegel S. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 150-155Crossref PubMed Scopus (235) Google Scholar, 12Mao C. Wadleigh M. Jenkins G.M. Hannun Y.A. Obeid L.M. J. Biol. Chem. 1997; 272: 28690-28694Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar). Additionally, yeast cells express S1P lyase encoded by DPL1 (formerly BST1), which cleaves S1P to yield ethanolamine-1-phosphate and hexadecanal (13Saba J.D. Nara F. Bielawska A. Garrett S. Hannun Y.A. J. Biol. Chem. 1997; 272: 26087-26090Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar). Mutants lacking S1P lyase (dpl1/bst1 mutants) accumulate phyto-S1P and dihydro-S1P at slightly elevated levels (9Kim S. Fyrst H. Saba J. Genetics. 2000; 156: 1519-1529Crossref PubMed Google Scholar, 14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar) and reach maximal S1P accumulation levels within 60 min of the addition of sphingosine to the culture medium (13Saba J.D. Nara F. Bielawska A. Garrett S. Hannun Y.A. J. Biol. Chem. 1997; 272: 26087-26090Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar). These effects are further exacerbated in dpl1 lcb3 double mutants lacking both S1P lyase and the major S1P phosphatase with 500 times greater levels of phyto-S1P and dihydro-S1P relative to wild type (7Nagiec M.M. Skrzypek M. Nagiec E.E. Lester R.L. Dickson R.C. J. Biol. Chem. 1998; 273: 19437-19442Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar, 14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar). Although the functions of S1Ps in yeast are largely unknown, recent evidence suggests a role for these lipids in resistance to heat stress in the regulation of cell proliferation and in the shift from fermentative to respiratory growth (8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar, 14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar, 15Mao C. Saba J.D. Obeid L.M. Biochem. J. 1999; 342: 667-675Crossref PubMed Scopus (107) Google Scholar, 16Gottlieb D. Heideman W. Saba J.D. Mol. Cell. Biol. Res. Commun. 1999; 1: 66-71Crossref PubMed Scopus (63) Google Scholar). The possibility that S1Ps regulate calcium signaling in yeast cells has not been examined. In yeast, calcium signals are generated in response to a variety of external stimuli including mating pheromones, salt stress, glucose-1-phosphate accumulation, and depletion of Ca2+from secretory organelles (17Fu L. Miseta A. Hunton D. Marchase R.B. Bedwell D.M. J. Biol. Chem. 2000; 275: 5431-5440Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar, 18Ohsumi Y. Anraku Y. J. Biol. Chem. 1985; 260: 10482-10486Abstract Full Text PDF PubMed Google Scholar, 19Mendoza I. Rubio F. Rodriguez-Navarro A. Pardo J.M. J. Biol. Chem. 1994; 269: 8792-8796Abstract Full Text PDF PubMed Google Scholar, 20Halachmi D. Eilam Y. FEBS Lett. 1996; 392: 194-200Crossref PubMed Scopus (60) Google Scholar, 21Strayle J. Pozzan T. Rudolph H.K. EMBO J. 1999; 18: 4733-4743Crossref PubMed Scopus (109) Google Scholar). Yeast expresses a Ca2+influx channel related to voltage-gated Ca2+ channels of animals (22Paidhungat M. Garrett S. Mol. Cell. Biol. 1997; 17: 6339-6347Crossref PubMed Scopus (161) Google Scholar, 23Fischer M. Schnell N. Chattaway J. Davies P. Dixon G. Sanders D. FEBS Lett. 1997; 419: 259-262Crossref PubMed Scopus (149) Google Scholar, 24Locke E.G. Bonilla M. Liang L. Takita Y. Cunningham K.W. Mol. Cell. Biol. 2000; 20: 6686-6694Crossref PubMed Scopus (180) Google Scholar) in addition to various intracellular Ca2+pumps and exchangers related to animal enzymes (25Cunningham K.W. Fink G.R. J. Cell Biol. 1994; 124: 351-363Crossref PubMed Scopus (363) Google Scholar, 26Cunningham K.W. Fink G.R. Mol. Cell. Biol. 1996; 16: 2226-2237Crossref PubMed Scopus (380) Google Scholar, 27Pozos T.C. Sekler I. Cyert M.S. Mol. Cell. Biol. 1996; 16: 3730-3741Crossref PubMed Scopus (129) Google Scholar, 28Antebi A. Fink G.R. Mol. Biol. Cell. 1992; 3: 633-654Crossref PubMed Scopus (377) Google Scholar, 29Sorin A. Rosas G. Rao R. J. Biol. Chem. 1997; 272: 9895-9901Abstract Full Text Full Text PDF PubMed Scopus (218) Google Scholar). However, homologs of the sarcoendoplasmic reticulum calcium ATPase-type Ca2+ pump and the inositol 1,4,5-trisphosphate (IP3) or ryanodine receptors, which supply and release Ca2+ from the endoplasmic reticulum in animal cells, are not evident in the yeast genome. Nevertheless, the yeast endoplasmic reticulum accumulates sufficient Ca2+ to facilitate protein folding and secretion, in part through Pmr1p, a member of the secretory pathway calcium ATPase family (21Strayle J. Pozzan T. Rudolph H.K. EMBO J. 1999; 18: 4733-4743Crossref PubMed Scopus (109) Google Scholar, 29Sorin A. Rosas G. Rao R. J. Biol. Chem. 1997; 272: 9895-9901Abstract Full Text Full Text PDF PubMed Scopus (218) Google Scholar). Sequences of the mammalian S1P-receptor involved in Ca2+ release from microsomes have not been reported. Here we show that conversion of exogenously added sphingosine to sphingosine 1-phosphate stimulates Ca2+ influx, accumulation, and signaling in yeast. Similar to the CCE-like mechanism of yeast (15Mao C. Saba J.D. Obeid L.M. Biochem. J. 1999; 342: 667-675Crossref PubMed Scopus (107) Google Scholar), the calcium channel subunit Cch1p was required for the majority of S1P-stimulated Ca2+ accumulation. Therefore, yeast may retain S1P-regulated calcium-signaling mechanisms analogous to those of mammalian cells. All yeast strains (TableI) were maintained on either YPD medium or synthetic complete medium (SC) lacking leucine or uracil. Strains of the JK9-3d background CBY31, CBY32, CBY33, and CBY34 were constructed from MSS200, MSS204, MSS205, and MSS207 (14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar), respectively, by curing the TPS2-lacZ::URA3 plasmid after growth in 5-fluoroorotic acid. CBY79 (dpl1 cch1) was constructed by transformation using thecch1::TRP1 disruption plasmid pKC289 (24Locke E.G. Bonilla M. Liang L. Takita Y. Cunningham K.W. Mol. Cell. Biol. 2000; 20: 6686-6694Crossref PubMed Scopus (180) Google Scholar). Similarly, CBY224–229 were constructed by transformation using the pmr1::LEU2 disruptant plasmid pL119 (30Rudolph H.K. Antebi A. Fink G.R. Buckley C.M. Dorman T.E. LeVitre J. Davidow L.S. Mao J.I. Moir D.T. Cell. 1989; 58: 133-145Abstract Full Text PDF PubMed Scopus (436) Google Scholar). Finally, pgm2::LEU2 mutant strains were constructed by transformation of CBY31 and CBY106 using pDB419 (17Fu L. Miseta A. Hunton D. Marchase R.B. Bedwell D.M. J. Biol. Chem. 2000; 275: 5431-5440Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar). All disruptions were confirmed by polymerase chain reaction and/or phenotypic analyses. All other strains were constructed by isogenic crosses.Table IList of yeast strains used in this studyStrainGenotypeSourceCBY31MATa leu2–3,112 ura3–52 his4 trp1 rme1This studyCBY32MATa leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷LEU2This studyCBY33MATa leu2–3,112 ura3–52 his4 trp1 rme1 lcb3∷LEU2This studyCBY34MATa leu2–3,112 ura3–52 his4 trp1 rme1 lcb3∷LEU2 dpl1∷leu2∷TRP1This studyCBY47MATα leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷LEU2 lcb4∷G418R lcb5∷G418RThis studyCBY106MATa leu2–3,112 ura3–52 his4 trp1 rme1 lcb4∷G418R lcb5∷G418RThis studyCBY79MATa leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷LEU2 cch1∷TRP1This studyCBY200MATα leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷LEU2 cch1∷TRP1 lcb4∷G418R lcb5∷G418RThis studyCBY224–226MATα leu2–3,112 ura3–52 his4 trp1 rme1 pmr1∷LEU2This studyCBY227–229MATα leu2–3,112 ura3–52 his4 trp1 rme1 pmr1∷LEU2 lcb4∷G418R lcb5∷G418RThis studyCBY230MATa leu2–3,112 ura3–52 his4 trp1 rme1 pgm2∷LEU2This studyCBY231MATa leu2–3,112 ura3–52 his4 trp1 rme1 pgm2∷LEU2 lcb4∷G418R lcb5∷G418RThis studyCBY169MATa leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷TRP1 lcb3∷LEU2 lcb4∷G418R lcb5∷G418RThis studyMSS200MATa leu2–3,112 ura3–52 his4 trp1 rme1 TPS2-lacZ∷URA3(14)MSS204MATa leu2–3,112 ura3–52 his4 trp1 rme1 lcb3∷LEU2 TPS2-lacZ∷URA3(14)MSS205MATa leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷LEU2 TPS2-lacZ∷URA3(14)MSS207MATa leu2–3,112 ura3–52 his4 trp1 rme1 dpl1∷leu2∷TRP1 lcb3∷LEU2 TPS2-lacZ∷URA3(14) Open table in a new tab d-Erythro-sphingosine andl-erythro-sphingosine, phytosphingosine, and dihydrosphingosine were purchased from Sigma. FK506 was generously provided by Fujisawa USA, Inc. (Tokyo, Japan). Coelenterazine was obtained from Molecular Probes, Inc. 45CaCl2was obtained from Amersham Pharmacia Biotech. Cells were grown overnight at 30 °C in SC medium. Log phase cells were harvested, resuspended in fresh SC medium to an A 600 of 1–2, and then diluted 2-fold into medium containing45Ca2+, phytosphingosine, dihydrosphingosine, sphingosine, chlorpromazine, and/or FK506, as described in the text. Cells were harvested by filtration, washed, and processed for determination of total associated Ca2+ as described previously (25Cunningham K.W. Fink G.R. J. Cell Biol. 1994; 124: 351-363Crossref PubMed Scopus (363) Google Scholar). Assays were performed as described previously (31Nakajima-Shimada J. Iida H. Tsuji F.I. Anraku Y. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6878-6882Crossref PubMed Scopus (100) Google Scholar). Cells expressing pKC146 2E. Muller E. Locke, and K. W. Cunningham, submitted for publication. apoaequorin were grown overnight at 30 °C in SC medium lacking uracil. Log phase cells were concentrated to an A 600 of 100 in SC lacking leucine medium, incubated with 10% v/v of 590 μmnatural coelenterazine stock for 20 min, washed, and diluted 100-fold into fresh medium supplemented with 12.5 μm sphingosine, 10 mm BAPTA, and/or 5 μg/ml cycloheximide. The resulting luminescence was measured at intervals for 2–3 h. Cells were subsequently permeabilized with 250 μm digitonin, and the luminescence was recorded to standardize for aequorin loading between strains. Cells expressing pKC190 or pDM5 (26Cunningham K.W. Fink G.R. Mol. Cell. Biol. 1996; 16: 2226-2237Crossref PubMed Scopus (380) Google Scholar, 33Matheos D.P. Kingsbury T.J. Ahsan U.S. Cunningham K.W. Genes Dev. 1997; 11: 3445-3458Crossref PubMed Scopus (277) Google Scholar) were grown to log phase overnight at 30 °C in SC medium lacking uracil. Cultures were harvested and resuspended to a finalA 600 of 1 in 2 ml of fresh SC lacking uracil or YPD medium supplemented with sphingosine, NaCl, α-mating factor, and/or FK506 as noted in the text. Cells were incubated with shaking at 30 °C for 3–4 h before assaying for β-galactosidase activity as described previously (34Guarente L. Methods Enzymol. 1983; 101: 181-191Crossref PubMed Scopus (871) Google Scholar). Exogenous sphingosine added to culture medium can be taken up by yeast cells and phosphorylated to S1P by sphingosine kinases (13Saba J.D. Nara F. Bielawska A. Garrett S. Hannun Y.A. J. Biol. Chem. 1997; 272: 26087-26090Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar). S1P can be dephosphorylated by the phosphatase Lcb3p or degraded by the lyase Dpl1p (11Mandala S.M. Thornton R. Tu Z. Kurtz M.B. Nickels J. Broach J. Menzeleev R. Spiegel S. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 150-155Crossref PubMed Scopus (235) Google Scholar, 12Mao C. Wadleigh M. Jenkins G.M. Hannun Y.A. Obeid L.M. J. Biol. Chem. 1997; 272: 28690-28694Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 13Saba J.D. Nara F. Bielawska A. Garrett S. Hannun Y.A. J. Biol. Chem. 1997; 272: 26087-26090Abstract Full Text Full Text PDF PubMed Scopus (213) Google Scholar). To determine whether S1P accumulation can evoke calcium signaling in yeast, we first monitored the accumulation of 45Ca2+ from the medium into growing yeast cells treated with a wide range of exogenous sphingosine. The addition of sphingosine to the culture medium at concentrations >25 μm stimulated 45Ca2+ accumulation in wild-type yeast strains up to 2× the basal level (Fig.1 A). Mutants lacking the S1P phosphatase (lcb3 mutants) were indistinguishable from wild type in this assay. In contrast, dpl1 mutants lacking S1P lyase accumulated 5-fold higher amounts of45Ca2+ after treatment with only 10–15 μm sphingosine. These results show that exogenous sphingosine can stimulate Ca2+ accumulation and that this response can be inhibited by the S1P lyase Dpl1p. If phosphorylation of exogenous sphingosine was required to promote Ca2+ influx, mutants lacking the sphingosine kinases would exhibit less Ca2+ accumulation after treatment with sphingosine. Indeed, the 5-fold increase in45Ca2+ accumulation seen in a dpl1mutant at 12.5 μm sphingosine was abolished in dpl1 lcb4 lcb5 triple mutants (Fig. 1 B, left), indicating that this dramatic increase in Ca2+ accumulation was dependent on the S1P produced by Lcb4p and Lcb5p. However, thelcb4 lcb5 double mutants exhibited wild-type sensitivity to sphingosine (Fig. 1 B, right). Thus, exogenous sphingosine produced two separable calcium responses in yeast, one that was relatively small and independent of sphingosine kinases and another larger response that was dependent on sphingosine kinases and sensitive to S1P lyase and phosphatase. All future experiments will examine the properties of the latter S1P-specific response, which is prominent indpl1 mutants. Sphingosine kinases typically phosphorylate the naturald-isomer of sphingosine and are unable to act on thel-isomer (8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar). The addition of l-sphingosine to yeast cultures stimulated 45Ca2+ accumulation in dpl1 mutants only at very high concentrations similar to those effective in dpl1 lcb4 lcb5 triple mutants (Fig.1 C). Thus, l-sphingosine failed to stimulate Ca2+ accumulation through the pathway involving sphingosine kinases even in the supersensitive dpl1 strain. The results confirm the existence of two separable responses to added sphingosine, one that is not stereospecific or dependent on sphingosine kinases and one that is specific for the biologically active d-isomer of sphingosine, dependent on sphingosine kinases, and sensitive to S1P lyase and phosphatase. Yeast and other fungi synthesize phytosphingosine rather than sphingosine, as well as its precursor dihydrosphingosine. These molecules differ in the level of saturation and hydroxylation at C-4 (35Dickson R.C. Lester R.L. Biochim. Biophys. Acta. 1999; 1426: 347-357Crossref PubMed Scopus (168) Google Scholar). Exogenous phytosphingosine and dihydrosphingosine stimulated the nonspecific Ca2+ response much like sphingosine but only weakly stimulated the specific Ca2+ response indpl1 mutants (Fig.2 A). The weaker effects of exogenous phytosphingosine and dihydrosphingosine might be explained if they are poorer substrates than sphingosine for the kinases (7Nagiec M.M. Skrzypek M. Nagiec E.E. Lester R.L. Dickson R.C. J. Biol. Chem. 1998; 273: 19437-19442Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar, 8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar) or if their phosphorylated products are better substrates than S1P for Lcb3p. The lcb3 single mutants accumulate approximately 10-fold higher levels of phyto-S1P and dihydro-S1P than wild type during vegetative growth (9Kim S. Fyrst H. Saba J. Genetics. 2000; 156: 1519-1529Crossref PubMed Google Scholar, 14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar), but they do not accumulate more45Ca2+ than wild type with or without added sphingosine (Fig. 1 A). Therefore, a higher threshold level of the native S1Ps may be necessary to stimulate Ca2+influx. To evaluate the possible roles of phyto-S1P and dihydro-S1P more carefully, we measured 45Ca2+ accumulation into a dpl1 lcb3 double mutant. A dpl1 lcb3double mutant that is also auxotrophic for certain amino acids is inviable in standard medium unless sphingosine kinases are also inactivated (9Kim S. Fyrst H. Saba J. Genetics. 2000; 156: 1519-1529Crossref PubMed Google Scholar). However, a prototrophic dpl1 lcb3 double mutant is viable and accumulates ∼500 times higher levels of phyto-S1P and dihydro-S1P than wild type (14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar). We observed that the viable dpl1 lcb3 double mutant accumulated45Ca2+ from the medium at a constitutively high rate in the absence of added sphingolipids and also showed even greater sensitivity to added sphingosine than dpl1 mutants (Fig.1 A). The high rate of 45Ca2+accumulation in dpl1 lcb3 double mutants was not observed indpl1 lcb3 lcb4 lcb5 quadruple mutants (Fig. 2 B), which fail to accumulate detectable levels of dihydro-S1P and phyto-S1P (9Kim S. Fyrst H. Saba J. Genetics. 2000; 156: 1519-1529Crossref PubMed Google Scholar). These results confirm that accumulation of S1P-related molecules native to yeast can stimulate 45Ca2+ influx and accumulation and suggest that Dpl1p inhibits the response more potently than Lcb3p. Cch1p was identified previously as the probable pore-forming subunit of a plasma membrane Ca2+ influx channel that is activated by a variety of stimuli (22Paidhungat M. Garrett S. Mol. Cell. Biol. 1997; 17: 6339-6347Crossref PubMed Scopus (161) Google Scholar, 23Fischer M. Schnell N. Chattaway J. Davies P. Dixon G. Sanders D. FEBS Lett. 1997; 419: 259-262Crossref PubMed Scopus (149) Google Scholar). To determine whether S1P stimulates Cch1p activity, we monitored Ca2+ accumulation in adpl1 mutant and a cch1 dpl1 double mutant after the addition of sphingosine. The dpl1 cch1 double mutant displayed sensitivity to sphingosine, which was similar to that of thedpl1 single mutant, but the maximal level of45Ca2+ accumulation was lower in the cch1 dpl1 double mutant than in the dpl1 single mutant (Fig.3, A and B). The residual effect of sphingosine in the cch1 dpl1 double mutant was not observed in cch1 dpl1 lcb4 lcb5 quadruple mutants. Therefore, the Cch1p channel was required for the major component of the S1P-stimulated 45Ca2+accumulation under these conditions. The kinetics of the S1P-specific response were monitored using cells expressing the calcium-sensitive photoprotein aequorin in the cytoplasm. Treatment of a dpl1 mutant with 12.5 μm sphingosine produced a detectable increase in aequorin luminescence within 60–80 min of treatment that rose sharply and reached a plateau for at least 1 h (Fig. 4, A andB) (data not shown). Treatment of a dpl1 lcb4 lcb5 triple mutant in a parallel experiment resulted in little aequorin luminescence over this time frame, indicating that S1P accumulation elevates cytosolic-free Ca2+([Ca2+]c) primarily via the specific pathway. The protein synthesis inhibitor cycloheximide completely abolished the response of dpl1 mutants to sphingosine (Fig.4 B), suggesting that protein synthesis was necessary for Ca2+ influx and [Ca2+]c elevation in response to sphingosine. Aequorin luminescence in dpl1mutants was also abolished by the addition of the Ca2+ ion chelator BAPTA to the culture medium, suggesting that the [Ca2+]c elevation observed was the result of extracellular Ca2+ influx and not intracellular Ca2+ release (Fig. 4 A). In summary, the specific response to S1P stimulated Ca2+ influx through Cch1p and elevated [Ca2+]c. To determine whether [Ca2+]c elevation in response to S1P could activate downstream signaling pathways, the expression of a calcineurin-dependent reporter genePMC1-lacZ was quantitated in cells treated with sphingosine. Sphingosine treatment induced the expression of PMC1-lacZ in a dpl1 mutant but not in dpl1 lcb4 lcb5 triple mutants (Fig. 5 A). The induction of PMC1-lacZ in dpl1 mutants was completely blocked by treatment with either FK506, a cell-permeant inhibitor of calcineurin, or membrane-impermeant BAPTA (Fig.5 B). Therefore, the S1P-specific response activated the calcineurin-signaling pathway in yeast by a mechanism requiring influx of extracellular Ca2+. In dpl1 lcb3 double mutants, PMC-lacZ expression was constitutively high in the absence of added sphingosine. This expression still required the function of the sphingosine kinases Lcb4p and Lcb5p and was sensitive to calcineurin inhibitors (Fig. 5 C). Taken together, these findings demonstrate that accumulation of S1P and related endogenous molecules can stimulate calcium influx and signaling in yeast. Several external stimuli lead to the generation of calcium signals in yeast. Although none of these stimuli has been shown to affect metabolism of S1P-related molecules, it is conceivable that one or more of these stimuli causes increased accumulation of S1Ps, which then triggers calcium signaling. To test whether endogenous S1P-related molecules are required for generating calcium signals in response to known physiological stimuli, we compared the calcium responses of wild type or dpl1 mutant cells to those oflcb4 lcb5 or dpl1 lcb4 lcb5 mutant cells lacking sphingosine kinases over a wide range of conditions. The first stimulus tested, α-mating factor, triggers Ca2+ influx and signaling after a time lag similar to that of sphingosine (18Ohsumi Y. Anraku Y. J. Biol. Chem. 1985; 260: 10482-10486Abstract Full Text PDF PubMed Google Scholar). We found that α-mating factor stimulated Ca2+ influx and calcineurin-dependent induction of PMC1-lacZ inlcb4 lcb5 double mutants to the same degree as in wild type (Fig. 6 A) (data not shown). Thus, the sphingosine kinases (and presumably their products) were not required for Ca2+ signaling invoked in response to α-mating factor. The second stimulus tested, high salt, also induced the calcineurin-dependent expression ofFKS2-lacZ to an equal extent in wild type and in lcb4 lcb5 double mutants (Fig. 6 B). Next, the acute heat shock produced by shifting cells grown at 25–39 °C stimulated a transient elevation of [Ca2+]c in bothdpl1 and dpl1 lcb4 lcb5 mutants as detected by aequorin luminescence (Fig. 6 C). Hypotonic shock produced by diluting cells grown in standard medium with hypotonic medium (36Batiza A.F. Schulz T. Masson P.H. J. Biol. Chem. 1996; 271: 23357-23362Abstract Full Text Full Text PDF PubMed Scopus (136) Google Scholar) also stimulated aequorin luminescence in both wild type and lcb4 lcb5 double mutants (data not shown). Depletion of Ca2+ from secretory organelles using pmr1mutants was shown to stimulate Ca2+ accumulation in yeast (20Halachmi D. Eilam Y. FEBS Lett. 1996; 392: 194-200Crossref PubMed Scopus (60) Google Scholar, 21Strayle J. Pozzan T. Rudolph H.K. EMBO J. 1999; 18: 4733-4743Crossref PubMed Scopus (109) Google Scholar, 24Locke E.G. Bonilla M. Liang L. Takita Y. Cunningham K.W. Mol. Cell. Biol. 2000; 20: 6686-6694Crossref PubMed Scopus (180) Google Scholar). However, 45Ca2+ accumulation inpmr1 lcb4 lcb5 triple mutants was similar to that ofpmr1 mutants (Fig. 6 D). Chlorpromazine treatment stimulates Ca2+ influx and accumulation in wild-type cells (37Eilam Y. Biochim. Biophys. Acta. 1983; 733: 242-248Crossref PubMed Scopus (32) Google Scholar) and to an equal degree in lcb4 lcb5 double mutants (Fig. 6 E). Finally, Ca2+ accumulation inpgm2 mutants stimulated by growth in galactose medium (17Fu L. Miseta A. Hunton D. Marchase R.B. Bedwell D.M. J. Biol. Chem. 2000; 275: 5431-5440Abstract Full Text Full Text PDF PubMed Scopus (48) Google Scholar) also occurred in pgm2 lcb4 lcb5 triple mutants (Fig.6 F). In summary, the sphingosine kinases Lcb4p and Lcb5p were not required for calcium-signaling events triggered by any of the seven known stimuli. Therefore, it appears that the upstream stimulus for S1P signaling in yeast defines a novel event in yeast calcium signaling. The results reported here suggest that accumulation of S1P or related molecules in yeast can stimulate Ca2+ influx via Cch1p and other factors, resulting in the elevation of [Ca2+]c and activation of calcineurin signaling. Several conditions shown previously to accumulate S1P or the native derivatives phyto-S1P and dihydro-S1P were found to stimulate Ca2+ influx and signaling in a manner requiring the homologous sphingosine kinases Lcb4p and Lcb5p. Dpl1p, the S1P lyase, potently blocked the Ca2+ influx responses with or without Lcb3p, the major S1P phosphatase. Lcb3p was less effective in this regard and only significant in dpl1 mutants, possibly because of the activity of Ysr3p, a minor S1P phosphatase homologous to Lcb3p (11Mandala S.M. Thornton R. Tu Z. Kurtz M.B. Nickels J. Broach J. Menzeleev R. Spiegel S. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 150-155Crossref PubMed Scopus (235) Google Scholar). The direct targets or derivatives of S1P that lead to Ca2+ influx and signaling have not been determined. However, the 1-h lag time after the addition of sphingosine observed before the onset of Ca2+ influx and the sensitivity to cycloheximide suggests that the response might reflect the time necessary for expressing new proteins involved in Ca2+influx and sufficient buildup of S1P. In the presence of extracellular Ca2+ chelators such as BAPTA, sphingosine was completely unable to elevate [Ca2+]c (Fig. 4 A). Therefore, no evidence for S1P-triggered Ca2+ release was obtained in yeast thus far. Exogenous d-sphingosine also stimulated Ca2+accumulation to a smaller degree independent of the sphingosine kinases Lcb4p and Lcb5p. This kinase-independent response may be nonphysiological because a similar effect was detected using the biologically inactive isomer l-sphingosine and because of the very high doses necessary to achieve the response. We have noticed similar responses of yeast cells to low levels of other membrane active compounds such as lyso-phosphatidic acid and detergents. 3C. J. Birchwood, J. D. Saba, and K. W. Cunningham, unpublished observations. Unlike these nonspecific effects, the involvement of the Ca2+ channel protein Cch1p and the requirement for S1P biosynthesis and accumulation underscore the existence of specific mechanism for S1P-dependent calcium signaling in yeast. Cch1p activity was stimulated in pmr1 mutants (lacking the secretory pathway Ca2+ATPase) upon depletion of Ca2+ from secretory organelles (24Locke E.G. Bonilla M. Liang L. Takita Y. Cunningham K.W. Mol. Cell. Biol. 2000; 20: 6686-6694Crossref PubMed Scopus (180) Google Scholar). Treatment of mammalian cells with thapsigargin, a specific inhibitor of sarcoendoplasmic reticulum calcium ATPase that rapidly depletes Ca2+ from the endoplasmic reticulum and stimulates CCE mechanisms, was shown to stimulate sphingosine kinase activity (38Olivera A. Edsall L. Poulton S. Kazlauskas A. Spiegel S. FASEB J. 1999; 13: 1593-1600Crossref PubMed Scopus (74) Google Scholar). However, it seems improbable that accumulation of S1Ps serves as a messenger of CCE in yeast cells, because pmr1 lcb4 lcb5 triple mutants lacking the sphingosine kinases exhibited as much Ca2+ accumulation as pmr1 mutants (Fig. 6 D). Additionally, the injection of S1P in mammalian RBL cells failed to stimulate the CCE channel known as ICRAC, and inhibitors of sphingosine kinase failed to prevent ICRAC activation by thapsigargin (39Mathes C. Fleig A. Penner R. J. Biol. Chem. 1998; 273: 25020-25030Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). A more reasonable explanation is that accumulation of S1Ps stimulates CCE in animal and fungal cells by first triggering Ca2+ release from secretory organelles. S1P rapidly stimulates Ca2+ release from the endoplasmic reticulum of mammalian cells followed by stimulation of Ca2+ influx at the plasma membrane and signaling (3Mattie M. Brooker G. Spiegel S. J. Biol. Chem. 1994; 269: 3181-3188Abstract Full Text PDF PubMed Google Scholar, 5Ghosh T.K. Bian J. Gill D.L. Science. 1990; 248: 1653-1656Crossref PubMed Scopus (327) Google Scholar). The receptor for S1P has not been identified, but evidence suggests the Ca2+ release channel in the endoplasmic reticulum is distinct from the well characterized IP3 receptor and ryanodine receptor (3Mattie M. Brooker G. Spiegel S. J. Biol. Chem. 1994; 269: 3181-3188Abstract Full Text PDF PubMed Google Scholar). All these routes of Ca2+ release can rapidly deplete the endoplasmic reticulum of Ca2+ and activate CCE pathways. The yeast genome lacks sequences orthologous to the IP3 and ryanodine receptors, but nevertheless, yeast cells may utilize a mechanism resembling CCE to supply Ca2+to secretory organelles (24Locke E.G. Bonilla M. Liang L. Takita Y. Cunningham K.W. Mol. Cell. Biol. 2000; 20: 6686-6694Crossref PubMed Scopus (180) Google Scholar). Our data do not rule out the possibility that S1P activates a new class of Ca2+ release channels in yeast that is potentially related to the unidentified S1P-receptor in animal cells. Support for this hypothesis might be obtained through the identification of new factors required for S1P-stimulated Ca2+ signaling in yeast. What is the purpose of S1P-stimulated Ca2+ influx and signaling in yeast? That yeast would encounter either high sphingosine environments or conditions inactivating both S1P lyase and phosphatase seems improbable. Therefore, we examined a number of previously described stimuli that lead to Ca2+ signaling in wild type and in lcb4 lcb5 double mutants and found no evidence for the involvement of S1P or its derivatives in any of the processes. It is possible that a significant contribution of S1Ps to Ca2+signaling was masked by the action of functionally redundant pathways, similar to what has been previously proposed for the sphingosine kinases in the heat stress response (35Dickson R.C. Lester R.L. Biochim. Biophys. Acta. 1999; 1426: 347-357Crossref PubMed Scopus (168) Google Scholar). Alternatively, an untested stimulus may be found that causes S1P accumulation and stimulation of Ca2+ signaling. It seems probable that the stimulation of calcium influx and signaling by S1P is the result of a physiological phenomenon because of the similar responses to S1P in animal cells and the strong conservation of sphingosine kinases, S1P lyases, and S1P phosphatases among animals, fungi, and plants. Furthermore, sphingolipid perturbations have previously been shown to affect calcium homeostasis (32Tanida I. Takita Y. Hasegawa A. Ohya Y. Anraku Y. FEBS Lett. 1996; 379: 38-42Crossref PubMed Scopus (34) Google Scholar, 40Beeler T. Gable K. Zhao C. Dunn T. J. Biol. Chem. 1994; 269: 7279-7284Abstract Full Text PDF PubMed Google Scholar), perhaps indicating cross-regulation of the two systems. The analysis of yeast mutants lacking S1P synthesis and degradation factors has produced few insights into their physiological roles. Mutants lacking Dpl1p and Lcb3p exhibit enhanced tolerance of acute heat shock, whereas mutants lacking the sphingosine kinases exhibit slightly enhanced sensitivity (8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar, 35Dickson R.C. Lester R.L. Biochim. Biophys. Acta. 1999; 1426: 347-357Crossref PubMed Scopus (168) Google Scholar). The levels of dihydro-S1P and phyto-S1P also increase transiently (∼10 min) after heat shock (14Skrzypek M.S. Nagiec M.M. Lester R.L. Dickson R.C. J. Bacteriol. 1999; 181: 1134-1140Crossref PubMed Google Scholar). However, the degree of heat tolerance or sensitivity conferred by these mutants varies significantly with the conditions used (35Dickson R.C. Lester R.L. Biochim. Biophys. Acta. 1999; 1426: 347-357Crossref PubMed Scopus (168) Google Scholar). We observed a transient elevation of [Ca2+]c after heat stress, but the kinetics and magnitude were similar in dpl1and dpl1 lcb4 lcb5 double mutants (Fig. 6 C). Thus, the activity of the sphingosine kinases (and hence the ability to make S1Ps) had no effect on known calcium-signaling pathways under the conditions tested. Endogenous S1P may also participate in the normal diauxic shift of yeast cultures (8Lanterman M.M. Saba J.D. Biochem. J. 1998; 332: 525-531Crossref PubMed Scopus (60) Google Scholar, 16Gottlieb D. Heideman W. Saba J.D. Mol. Cell. Biol. Res. Commun. 1999; 1: 66-71Crossref PubMed Scopus (63) Google Scholar). We have not yet detected significant Ca2+ signaling events during the transition to diauxic growth. Therefore, S1P-mediated calcium signaling represents a novel calcium-signaling pathway in yeast that is, thus far, potentially activated by undiscovered stimuli. We thank David Bedwell for yeast strains and plasmids and Fujisawa USA, Inc., for the gift of FK506.