Báo cáo khoa học: Regulation of phospholipid biosynthesis by phosphatidylinositol transfer protein Sec14p and its homologues pot

Regulation of phospholipid biosynthesis by phosphatidylinositol transfer protein Sec14p and its homologues A critical role for phosphatidic acid Roman Holic ˇ , Milos ˇ Za ´ gors ˇ ek and Peter Griac ˇ Institute of Animal Biochemistry and Genetics, Slovak Academy of Sciences, Slovakia Transcription of yeast phospholipid biosynthesis structural genes, which contain a n inositol-sensitive upstream activa- ting sequence in their promoters, re sponds to the availability of the soluble precursors inositol and choline and to changes in phospholipid metabolism. The INO1 gene is deregulated (derepressed when inositol is pr esent) under the conditions of increased phosphatidylcholine (PtdCho) turnover, as occurs in the sec14D cki1D strain ( SEC14 encodes the majo r y east phosphatidylinositol transfer protein; CKI1 encodes choline kinase of the cytidine diphosphate choline pathway of PtdCho biosynthesis). Five proteins (Sfhp) share sequence homology with phosphatidylinositol transfer protein Sec14p. Two (Sfh2p and Sfh4p), when overexpressed largely complement the otherwise essential Sec14p requirement concerning growth and secretion. In this study, we analysed the ability of Sec14 homologues to c orrect the defect in regulation of phospholipid biosynthesis resulting from defective or missing Sec14p. We also analysed how PtdCho turnover relates t o the transcriptional regulation of phos- pholipid biosynthesis. The results show that (a) none of the Sec14 homologues w as able to substitute for Sec14p in its regulatory aspects of phospholipid biosynthesis, (b) removal of phospholipase D activity corrected the aberrant INO1 gene regulation in yeast strains with otherwise h igh PtdCho turnover, and (c) increased steady-state phosphatidic acid levels correlated with derepressed levels of t he INO1 gene. Overall, the results support t he model in w hich high phos- phatidic acid levels lead to derepression of the genes of phospholipid biosynthesis [Henry, S.A. & Patton-Vogt, J.L. (1998) Prog. Nucleic Acid Res. Mol. Biol. 61, 133–179]. Keywords: SEC14 homologues; INO1 regulation; phos- pholipid turnover; transcription; Saccharomyces cerevisiae. The majority of yeast phospholipid structural genes contain in their promoters inositol-sensitive upstream activating sequence ( UAS INO ) a nd ar e regulated in response to the availability of the soluble precursors inositol and choline. Yeast p hospholipid structural genes are derepressed in the absence o f inositol and repressed in t he presence of inositol [1]. UAS INO -containing genes are activated by two tran- scription factors – Ino2p and Ino4p [2] – and repressed by a negative regulator – Opi1p [3]. The INO1 gene (encoding inositol 1-phosphate synthase) [4] is the most regulated gene of the entire regulon and thus frequently serves as a reporter gene for the whole set of coordinately regulated genes of phospholipid biosynthesis [1,5]. In addition to the presence or absence of inositol, transcription of the INO1 gene responds to changes in phospholipid metabolism [6], correct INO1 regulation being dependent on ongoing phosphati- dylcholine (PtdCho) biosynthesis [7]. The INO1 gene is deregulated (derepressed when inositol is present) also under conditions of increased PtdCho degradation [8,9]. Critical analysis of conditions with defective regulation of the INO1 gene led to the development of a model in which the metabolic signal for INO1 derepression is generated by phosphatidic acid (PA) or a closely related metabolite [6]. This model proposes that the relative increase of P A, as a result of increased production, versus its utilization provides the signal for derepression of coordinately regulated UAS INO -containing gen es of phospholipid biosynthesis. A recent study by Loewen et al. [10] has demonstrated the mechanism by which Saccharomyces cerevisiae can regulate phospholipid biosynthesis. A fter the rapid consumption of PA following the addition of inositol (the negative regulator of phospholipid biosynthesis), Opi1 protein was released from the endoplasmic reticulum, e ntered the nucleus and repressed its target genes. Sec14p was originally identified as a phosphatidylinositol (PtdIns) transfer protein which catalyses the in vitro trans- port of PtdIns and PtdCho between artificial and b iological membranes [11]. In vivo, Sec14p perf orms an e ssentia l role in protein transport from the Golgi apparatus [12]. Connec- tion b etween the function of Sec14p and metabolism of the major membrane phospholipid, PtdCho, was established Correspondence to P. Griac ˇ , Institute of Animal Biochemistry and genetics, Slovak Academy of Sciences, Moyzesova 61, 900 28 Ivanka pri Dunaji, Slovakia. Fax: +421 245943932, Tel.: +421 245943151, E-mail: Peter.Griac@s avba.sk Abbreviations: CDP, cytidine diphosphate; INO1,geneencoding inositol 1-phosphate synthase; PA, phosphatidic acid; Pld1p, phos- pholipase D1; PtdCho, phosphatidylcholine 2 ; PtdIns, phosphatidyl- inositol 3,4 ; PtdSer, phosphatidylserine 3,4 ; Sec14p, phosphatidylinositol transfer protein; Sfh, Sec14 homologues; UAS INO , inositol-sensitive upstream activating sequence. (Received 17 August 2004, revised 23 September 2004, accepted 24 September 2004) Eur. J. Biochem. 271, 4401–4408 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04377.x by the following lines of evidence (Fig. 1), namely that (a) mutations in genes encoding structural enzymes of the cytidine diphosphate choline (CDP-choline) pathway for PtdCho biosynthesis bypass the essential requirement for Sec14p in secretion and cell viability [13], (b) PtdCho-bound Sec14p down-regulates a key enzyme of the CDP-choline pathway, namely choline phosphate cytidylyltransferase (Pct1p) [14], and (c) Sec14p regulates phospholipase D1 (Pld1p)-mediated PtdCho turnover, Sec14p being a negative regulator of this pathway in vegetatively growing cells [8,9,15] and having a positive effect on the production of PA originating from PtdCho in sporulation [16]. A connection between Sec14p and phospholipid biosynthesis exists also at the transcriptional le vel. The INO1 gene is not regulated properly i n t he absence of f unctional S ec14p, as demon- strated in the sec14 ts cki1D (the CKI1 gene encodes choline kinase, the structu ral gene of the CDP-choline pathway) strain when c ultured at a n on-permissive temperature of sec14 ts [8,9]. In yeast, five proteins ( Sec14 homologues, Sfh) share primary sequence homology with the major PtdIns transfer protein encoded by the SEC14 gene [17,18]. Sec14 homo- logues Sfh2p and Sfh4p, when overexpressed, complemen- ted very well the sec14 growth and secretion defects. Sfh 1p and Sfh5p were also able to complement these defects to some degree [17,18]. In this study, we analysed the ability of the Sec14 homologues to c orrect the r egulatory defect i n INO1 transcription resulting from defective or missing Sec14p. We show that none of the Sec14 homologues is able to substitute for Sec14p in reverting the derepressed levels of the INO1 transcript in the presence of inositol when the CDP-choline pathway is blocked. The reason why even the best growth phenotype-complementing homologue, Sfh2p, cannot substitute for Sec14p in its regulatory role toward phospholipid biosynthesis, lies in its inability to regulate phospholipase D-mediated PtdCho turnover. Using the sec14D ckiD pld1D triple mutant with overexpressed Sfh2p, we were able to perform a n in vivo test of the model that INO1 derepression can occur in response to metabolic signal generated via Pld1p- mediated PtdCho turnover. Moreover, we show that higher steady-state cellular l evels of PA correlate with deregulated levels of INO1 transcript in cells grown in the presence of inositol, providing support for the model that PA acts as a metabolic signal for INO1 derepression [6]. Experimental procedures Strains and culture conditions The yeast strains used in this study are listed i n Table 1. Recipient yeast strain PGY170, used to study INO1–lacZ expression, was a spore from a genetic cross of PGY145 [18] and BRS1069 [19]. Cells were grown aerobically at 30 °C with s haking in chemically defined synthetic media lacking inosito l and choline (I – C – ), supplemented with 75 l M inozitol (I + C – )or supplemented with 75 l M inositol and 1 m M choline (I + C + ) [20]. Plasmid and strain construction Standard genetic methods were used throughout this work [21]. Y east transformatio n was performed by using the lithium acetate method [22], with minor modifications. Plasmids. Episomal plasmids containing SEC14 and its homologues under their own promoters based on a yeast 2 lm plasmid, Y Eplac181 (yeast LEU2 marker) [23], were as described p reviously [18]. Episomal plasmid YEplac112- SFH2 (TRP1 marker) was constructed by subcloning SFH2 from YEplac181-SFH2 [18] into YEplac112, by u sing Sal I and SphI restriction enzymes [23]. Centromeric plasmid YCplac22- SEC14 (TRP1 marker) was constructed by subcloning SEC14 from YCp(SEC14)(URA3 marker) [24] (kindly provided b y V. Bankaitis, U niversity of North Carolina, Chapel Hill, NC, U SA) into YCplac22, by using EcoRI and HindIII restriction enzymess [23]. Gene disruption. Disruption of the Pld1p gene in strain PGY209 was p erformed by integrative transformation using the No tI/XhoI d isruption cassette f rom plasmid B913 (kindly provided by J . Engebrecht, Univ ersity of California San Diego, La Jolla, CA, USA). Assay for the Opi – (overproduction of inositol) phenotype To test for the excretion of inositol, as described in detail previously [25], yeast strain S3 was transformed with YEplac181 [23] based episomal plasmids containing SEC14 or SFH genes. Strains were patched on synthetic I – C – plates and allowed to grow at 30 °C for 2 days. The plates were then sprayed with a suspension of a diploid tester strain (AID), homozygous for ino1 and ade1 (Table 1), and incubated for another 2 days. b-Galactosidase assay b-Galacto sidase production was derived from the integrated version of the fully regulated INO1–lacZ construct [19]. Yeast strains were grown to mid-logarithmic phase in I + Fig. 1. Phospholipid metabolic pathways in ye ast. The pathways show n include the relevant steps discussed in the text. The dashed line r ep- resents degradation of phosphatidylcholine via a phospholipase D-mediated route. 4402 R. Holic ˇ et al.(Eur. J. Biochem. 271) Ó FEBS 2004 and I – media at 30 °C. The b-galactosidase assays were performed as described by Lopes et al.[19],exceptthat aliquots were removed from t he reaction mix at 5 , 10 and 15 or 20 min. b-Galactosidase units are defined as (A 420 Æ 5 min )1 Æmg )1 of protein) · 10 000. Protein was quanti- fied by using the method of Lowry et al.[26]. PtdCho turnover PtdCho turnover in strains with a cki1 genetic background was analysed as reported previou sly [8,27]. Yeast strains were grown overnight at 30 °CinI + C – media containing 1 lCiÆmL )1 of [methyl- 14 C]choline chloride. Cultures were harvested during mid-logarithmic growth phase, washed twice with sterile distilled water and resuspended in 5 mL of fresh unlabelled I + C – media. At the time-points indicated, 1.4 mL o f the culture was removed and the cells were collected by centrifugation 6 (6000 g,2°C, 3 min). The supernatant was saved as the medium fraction. The cell pellet was suspended in 0.5 mL of 5% (v/v) trichloroacetic acid and incubated on ice for 20 m in. After centrifugation 7 (6000 g,2°C, 3 min), the supernatant was saved as the intracellular water-soluble fraction. The pellet was resus- pended in 0.5 mL of 1 M Tris/HCl buffer (pH 8), centifuged (6000 g,2°C, 3 min), and t he resulting supernatant was combined with the intracellular water-soluble f raction, effectively neutralizing t he acidic extract. The final pellet was saved as the total membrane fraction. To solubilize t he cell pellet, it was suspended in 100 lL of 1% (v/v) Triton-X- 100, frozen at )70 °C and incubated in the presence of 10% (w/v) deoxycholate, overnight at 37 °C. Radioactivity of all fractions was determined by liquid scintillation counting. Radioactivity in glycerophosphocholine and choline in the medium fraction was d etermined after se paration on a cation-exchange column, Dowex 50 WX 8 (200–400 mesh; Serva), as described by Cook & Wakelam [28]. Phospholipid composition Steady-state labelling w ith 32 P-labelled orthophosphate (2 mC iÆmL )1 ; I CN) was performed by u sing the method of Atkinson et al. [29]. Cells were labelled o vernight, for at least six generations, with 5 lCiÆmL )1 of 32 P-labelled orthophosphate in 5 mL of vitamin-defined synthetic I + C – media, and harvested during the late logarithmic phase of growth. Labelled lipids were extracted as described by Atkinson et al. [29]. 2D pape r chromatography on silica- impregnated paper was performed according to the method of Steiner & Lester [30]. Labelled spots, corresponding to specific lipids, were cut out after autoradiography and quantified by liquid scintillation. Results Two of the Sec14 homologues, namely Sfh2p and Sfh4p, when overexpressed under their own p romoters from multicopy episomal plasmids, efficiently complemented the growth and secretion defects of the sec14 ts strain. To a low (but significant) degree, Sfh1 protein also complemented the sec14 ts -associated growth d efect [18]. T he Opi – (over- production of inositol) phenotype is indicative of over- expression of inositol-1-phosphate synthase owing to misregulation of the INO1 gene [1,25]. The S3 strain (sec14D pct1D), containing an empty cloning vector, dis- plays a strong Opi – phenotype (Fig. 2), in agreement with a previous report [8] showing that sec14 mutants containing a block in the CDP-choline pathway exhibit overexpession of the INO1 gene. The Opi – phenotype disappeared when the S 3 s train was transformed with an episomal plasmid containing the SEC14 gene (Fig. 2). However, none of the Sec14 homologues was able to fully suppress the Opi – phenotype in the sec14D pct1D strain. To measure in a quantitative manner the levels of INO1 gene expression, we used the fully regulated INO1–lacZ reporter construct integrated into the g enome of the sec14D cki1D strain. As expected, SEC14 was able to restore wild-type regulation of t he INO1 gene, derepression of the gene in medium lacking inositol, and repression in medium containing inositol. N one of the Sec14 homo- logues, even t hose that complemented the sec14-associated growth defect fairly well, restored the c orrect regulation of the INO1 gene in response t o inositol (Fig. 3). Next, we i nvestigated the inability of Sfh2p, w hich complemented ve ry well the growth defect associated w ith the sec14 mutation, to complement the INO1 gene regula- tory defect. PtdCho turnover, consistent with previously published data [8], was enhanced in the sec14D cki1D strain under the conditions in which the regulatory defect occurred (inositol-containing media) (Fig. 4A). The sec14D cki1D strain, containing SEC14 on a centr omeric plasmid Table 1. Yeast strains. Name Genotype Source PGY145 Mat a leu2, his3, trp1, ura3, cki1::HIS3, sec14::kanMX P. Griac ˇ [18] BRS1069 Mat a ade2–1, his3–11,15, leu2–3,112, can1–100, trp1–1, ura3–1, ura3–1::INO1-lacZ(URA3) (pJH334) S. Henry [19] S3 Mat a leu2,his3, ura3, pct1::URA3, sec14::kanMX P. Griac ˇ [18] AID Mat a/a ade1/ade1 ino1/ino1 S. Henry [25] PGY170 Mat a ade2, his3, ura3, trp1, leu2, cki1::HIS3, sec14::kanMX, ura3::INO1-lacZ (URA3) This work PGY209 Mat a ade2, his3, ura3, trp1, leu2, cki1::HIS3, sec14::kanMX, ura3::INO1-lacZ (URA3), YEplac112 – SFH2 (TRP1) This work PGY210 Mat a ade2, his3, ura3, trp1, leu2, cki1::HIS3, sec14::kanMX, pld1::LEU2, ura3::INO1-lacZ (URA3), YEplac112 – SFH2 (TRP1) This work PGY216 Mat a ade2, his3, ura3, trp1, leu2, cki1::HIS3, sec14::kanMX, ura3::INO1-lacZ (URA3), YCplac22 (TRP1) This work PGY218 Mat a ade2, his3, ura3, trp1, leu2, cki1::HIS3, sec14::kanMX, ura3::INO1-lacZ (URA3), YCplac22 – SEC14 (TRP1) This work Ó FEBS 2004 Phosphatidylcholine turnover and INO1 1 regulation (Eur. J. Biochem. 271) 4403 (Fig. 4 B), displayed greatly reduced PtdCho turnover. I n contrast to Sec14p, Sfh2p was unable to attenuate high PtdCho turnover (Fig. 4C). Deletion of the Pld1p gene, pld1D,inthesec14D cki1D strain containing Sfh2p expressed from t he episomal plasmid, diminished the majority of PtdCho turnover (Fig. 4D). To test the hypothesis that the inability of Sfh2p to decrease PtdCho turnover to wild-type levels i s responsible for the inab ility of S fh2p to complement the INO1 regulatory defect caused by the absence of Sec14p, we measured INO1 expression in the sec14D cki1D pld1D strain containing SFH2 on an episomal plasmid. It was shown previously that the phospholipase D 1 (PLD1) gene product plays an essential role in the suppression of the otherwise lethal sec14D defect by CDP-choline p athway muta- tions [9,17]. As a consequence, the triple mutant sec14D cki1D pld1D is not viable. Introduction of the SFH2 gene on a multicopy episomal plasmid renders the triple mutant sec14D cki1D pld1D viable [18]. Note that introduction of Sfh2p f rom the episo mal plasmid does not decrease the high PtdCho turnover of the sec14D cki1D strain (Fig. 4A,C). Next, we compared INO1 gene expres- sion in the triple mutant sec14D cki1D pld1D (SFH2)tothe same strain with an intact PLD1 gene, sec14D cki1D (SFH2) (Fig. 5 ). The results show clearly that INO1 expression in thetriplemutant,sec14D cki1D pld1D (SF H2), corresponds to the ÔnormalÕ INO1 expression, as represented by the sec14D cki1D strain containing the SEC14 gene on a centromeric plasmid, and not to the deregulated situation found in the sec14 D cki1D or se c14D cki1D (SFH2)strains. Two products – PA and choline – are generated from every molecule of PtdCho catabolized via a phospholipase D-mediated route. In previous experiments [20], no corre- lation was found between the r elative content of PtdCho Fig. 2. Overproduction of the inositol (Opi – ) phenotype. The sec14 pct1 mutant strain (S3, Table 1) was transformed with plasmids con- taining SEC14, SEC14 homologues SFH1– SFH5, and an empty cloning vector, YEplac181. Excretion o f inosito l (an i ndicator of aberrant INO1 gene regulation) results in halo-type growth of the tester strain around the strains being tested. The sec14 pct1 strain, containing SEC14 plasmid, represents wild- type INO1 regulation, and the empty cloning vector serves as a negative control. Fig. 3. Sec14 homologues are not able to correct aberrant regulation of the INO1 gene. The sec14 cki1 strain (PGY170, Table 1) containing SEC14, SEC14 homologues SFH1–SFH5, and an empty cloning vector (YEplac181) was grown at 30 °C in d erepressing (I – C – )or repressing (I + C + ) conditions to the mid-logarithmic phase of growth. The b-galactosidase assay was performed as described in the Experi- mental proc edure s. Data are expressed as the mean ± SD of three independent experiments. The sec14 cki1 strain containing the SEC14 plasmid represents wild-type INO1 regulation, and the empty cloning vector serves as a negative control. 4404 R. Holic ˇ et al.(Eur. J. Biochem. 271) Ó FEBS 2004 and/or free choline availability, and INO1 gene regulation. Therefore, w e focused on the correlation between the other PtdCho degradation product, PA, and INO1 gene regula- tion. Steady-state phospholipid compositions of the corres- ponding strains were directly co mpared. Figure 6A represents comparison of the phospholipid compos ition i n the sec14D cki1D (SEC14)strain(ÔnormalÕ PtdCho tu rnover and ÔnormalÕ INO1 expression) and the sec14D cki1D strain (enhanced PtdCho turnover and d eregulated INO1 expres- sion). Figure 6B represents comparison of the phospholipid composition i n the sec14D cki1D pld1D (SFH2)strain (genetically blocked Pld1p-mediated PtdCho turnover and ÔnormalÕ INO1 expression) and the sec14D cki1D (SFH2) strain (enhance d PtdCho t urnover and d eregulated INO1 exp ression). In both pairs of strains, enhanced ABCD Fig. 4. The Sec14 homologue, Sfh2, is not able to control phospholipase D-mediated phosphatidylcholine turnover. The sec14 cki1 strain (PGY170, Table 1), containing empty cloning vector YCplac22 (A), SEC14 (B), the SEC14 homologue, SFH2 (C), and the sec14cki1pld1strain containing the SEC14 homologue, SFH2 (PGY210) (D), were cultured at 30 °C to the mid-logarithmic phase of growth in I + C – medium containing 1 lCiÆlL )1 of 14 C-labelled choline chloride. At time zero, the cells were centrifuged, washed and reinoculated into nonradioactive I + C – medium. Data are expressed as a percentage of total label at each time-point: (m), radioactivity i n the culture medium; (j), radioactivity in the cellular, water- soluble fraction; (r), radioactivity in the membrane fraction. The data represent the average of at least t wo independent experiments. Fig. 5. Phospholipase D-mediated high phosphatidylcholine turnover is responsible for deregulation of t he INO1 gene. The sec14 cki1 strain (PGY170, Table 1), containing empty cloning vector YCplac22, SEC14,theSEC14 homologue (SFH2)andthesec14cki1pld1strai n containing the SEC14 homolo gue, SFH2 (PGY210), were cultured at 30 °C to th e mid-logarithmic p hase of growth at 30 °C in derepressing (I – C – )orrepressing(I + C – ) conditions. The sec14 cki1 strain, con- taining SEC14 plasm id, represents wild-type INO1 regulation; the sec14 c ki1 strain, contain ing the empty cloning vector, YCplac22, serves as a dere gulat ed INO1 control. The b-galactosidase assay was performed as described in the Materials and methods. Data represent the mean ± SD of three to five independent experiments. Fig. 6. Steady-state phospholipid composition. The sec14 cki1 strain (PGY170, Table 1) containing empty cloning vector, YCplac22, SEC14 (A), th e SEC14 homologue, SFH2,andthesec14cki1pld1 strain containing th e SEC14 hom ologue, SFH2 (PGY210), (B) were cultured for five to six generations in I + C – medium at 30 °Ccon- taining 5 lCiÆmL )1 of 32 P-labelled orthophosphate. The cells were harvested a nd the phospholipid s e xtracted and resolved as described in the Experimental procedures. The term ÔothersÕ includes lipids remaining near the origin, and cytidine diphosphate diacylglycerol, cardiolipin 8,98,9 and other minor lipids. Values represent the percentage of total lipid-associated 32 P incorporated into each phospholip id species. Data are expressed a s the mean value of two independent experiments. Ó FEBS 2004 Phosphatidylcholine turnover and INO1 1 regulation (Eur. J. Biochem. 271) 4405 PtdCho turnover resulted in a lower steady-state level of PtdCho and a corresponding compensatory increase in phosphatidylethanolamine ( PtdEtn) l evels c ompared with their counterparts that have a ÔnormalÕ PtdCho turnover. Both strains sec14D cki1D and sec14D cki1D (SFH2), with a derepressed expression of the INO1 gene in the medium containing inositol, display significantly higher steady-state levels of PA than their counterparts that have a ÔnormalÕ (repressed) expression of the INO1 gene (and low levels of PtdCho turnover). Thus, high steady-state levels of P A correlate with th e inability of the INO1 gene to repress in response to inositol availability, and low steady-state levels of PA correlate with the normal regulation of the INO1 gene – its ability to repress in response to i nositol availability. Discussion Products of five yeast genes, named SFH1–5, exhibit significant sequence homology to Sec14p [17,18]. Sfh1p, which shows greatest similarity to Sec14p, c onserves all recognized critical structural motifs of Sec14p [31,32]. Sfh2, Sfh3 and Sfh4 proteins share modest homology with Sec14p throughout their primary sequences. The recognized signa- ture motif s LLRFLRARKF, DGRPVY, YYPERMGK- FY and INAP of fungal Sec14 proteins [32] are clearly recognizable in Sfh2p, Sfh3p and Sfh4p primary sequences. These motifs are conserved only t o a limited degree in Sfh5p. Primary sequence homology shared by Sec14p and Sfh proteins assumes functional s ignificance because some of the Sec14 homologues (namely Sfh2 and Sfh4), when overexpressed, complemented the sec14-related growth and secretory defects [17,18]. Sfh3p a nd Sfh5p, under the transcriptional control of their own promoters, failed to do so. Sfh1p, which displays th e highest degree of similarity to Sec14p, c omplemented the sec14 -related growth defect only to a limited degree [18]. We hypothesize that the reason for this weak growth complementation co uld be a result of the different subcellular localization of Sec 14p from Sf h1p [18]. All Sec14 homologues, except (interestingly) for Sfh1p, display PtdIns transfer activity in vitro. Sec14 homologues do not, however, show PtdCho transfer activity typical for Sec14p [17]. Our goal was to determine whether any of the Sec14 homologues could substitute for Sec14p in its role as a regulator of phospholipid biosynthesis. Under conditions where Sec14p is nonfunctional and the CDP-choline pathway for PtdCho biosynthesis is blocked, the INO1 gene, the most highly regulated of a set of genes encoding enzymes of phospholipid biosynthesis, is deregulated (dere- pressed in the presence of inositol) [8]. Using two criteria – Opi – phenotype and the b-galactosidase activity assay – we tested the ability of all five Sec14 homologues to revert the INO1 regulatory d efect caused by the absence of functional Sec14p. Our results (Fig. 2) demonstrate that none of the S ec14 homologues, when expressed from multicopy plasmids in the sec14D pct1D strain, was able to fully suppress the Opi – phenotype. The overexpression of two genes (SFH2 and SFH4) that rescue growth of sec14 ce lls seems, to some extent, to decrease the inositol secretion. The Opi – phenotype is an excellent indicator of INO1 regulation, reflecting INO1 expression in media w ithout inositol. Nevertheless, it is a multifactorial phenotype (depending also on the metabolic status o f the cells, their growth rate and the ability of the cells to transport surplus inositol to the media) and therefore the O pi – phenotype is being considered a qualitative, rather than a quantitative, indicator of INO1 gene regulation. Therefore, we also assessed INO1 expres- sion using the INO–lacZ construct. b-Galactosidase activity derived from the INO1 promoter in a yeast strain without Sec14p ( sec14D), and containing a genetic block in the CDP-choline pathway (this time cki1D to dem onstrate the independence on the nature of the CDP-choline pathway block) (Fig. 3), confirmed the Opi – phenotype results, showing that none of the Sec14 homologues is able to suppress the INO1 gene regulatory defect imposed by the absence of Sec14p when the CDP-choline pathway is blocked. What is the special function of Sec14p that neither the SFH gene product most similar to Sec14p (Sfh1p, showing 6 2.5% identity to Sec14p), nor the SFH gene product that almost completely complements the sec14 - associated growth and secretion defects (Sfh2p) [17,18], can substitute for? Sec14p acts as a negative r egulator of Pld1p- mediated PtdCho turnover pathway in vegetatively growing cells [8,9,15]. P aradoxically, however, the simultaneous deletion of several SFH genes significantly reduced phos- pholipase D activity in vivo [17], a nd overexpression of Sfh2p a nd Sfh4p rather increased Pld1p-mediated PtdCho turnover [18]. Many of these results were obtained by using yeast strains with a temperature-sensitive sec14 ts allele grown at t he restrictive temperature of 37 °C, and i t has been demonstrated that PtdCho turnover in S . cerevisiae varies as a function of temperature elevation, from 30 to 37 °C [27]. Moreover, the thermosensitive sec14 ts allele may contain some residual activity influencing Pld1p-mediated lipid turnover. Thus, we compared PtdCho turnover in the sec14D cki1D strain (Fig. 4A) and in the same strain transformed with a multicopy plasmid containing the Sec14 homologue gene, SFH2 (Fig. 4C). PtdCho turnover wasmeasuredinI + C – medium at 30 °C, cond itions under which the INO1 regulatory defect occurred. The same sec14D cki1D strain, transformed with a centromeric, low- copy-number plasmid containing the SEC14 gene (Fig. 4B), served as a control. The fate of the label after shifting the cultures to unlabelled medium was compared. There was a much higher PtdCho turnover rate in strains containing either empty vector or SFH2, as c ompared t o t he strain containing wild-type SEC14. Our results correspond very well with results obtained previously using the sec14 ts allele and following PtdCho turnover at 37 °CinI – C – media [8,18]. T his e xperiment clearly demonstrates the inability of the Sec14 homologue, Sfh2, to substitute for Sec14p in regulating the turnover of PtdCho. Next, we wanted to provide direct evidence that this elevated PtdCho turnove r is responsible for deregulation of the INO1 gene. To achieve this we disrupted the m ajor route for PtdCho degrad- ation b y Pld1p by using a triple mutant strain, sec14D cki1D pld1D, exp ressing S fh2p . The PLD1 gene product plays an essential role in suppression of the sec14 defect by CDP-choline pathway mutations [9], and, as a consequence, the triple mutant sec14D cki1D pld1D is not viable. However, introduction of th e SFH2 gene product expressed from the high-copy-number plasmid renders the triple mutant viable. Introduction of the Sfh2p into the sec14D cki1D strain does not change the transcriptional 4406 R. Holic ˇ et al.(Eur. J. Biochem. 271) Ó FEBS 2004 regulation of the INO1 gene (Figs 2 and 3) and also does not suppress the high Pld1p-mediated PtdCho turnover in the sec14D cki1D strain (Fig. 4). Direct comparison of INO1 regulation in the sec14 D cki1D (SFH2)strain,andin the same s train in which the Pld1p-mediated PtdCho turnover was abolished (Fig. 4), shows the major role that Pld1p-mediated turnover plays i n t he transcriptional regu- lation of the INO1 gene under our experimental conditions. Henry & Patton-Vogt [6] proposed a model in which the relatively high levels of PA, or another closely related metabolite, leads to derepression of the UAS INO -containing genes. Here, excess PA produced by phospholipase D- mediated turnover in the se c14 cki1 strain leads to d ere- pression of the INO1 gene. A still-controversial question is whether the molecule to which t he transcriptional regulation responds is PA or some other closely related metabolite. For example, diacylglycerol and PA are related to each other by a single phosphorylation/dephosphorylation s tep [1]. PA is a key intermediate in the formation of glycerophospholipids and triacylglycerols. There is always a fine balance b etween the production and the utilization of P A. In yeast, PA can be formed de novo from glycerol 3-phosphate and dihydro- acetone phosphate [33]. Phospholipase D-mediated PtdCho turnover produces one molecule of PA for every molecule of PtdCho hydrolyzed. On the demand site, PA is rapidly used in the synthesis of pho spholipids and triacylglycerols. The experiments in which we measured the steady-state levels o f phospholipids using 32 P-labelling, showed that deregulation of the INO1 gene correlates with higher s teady-state levels of PA (Fig. 6). The steady-state level of PA in the sec14D cki1D strain (deregulated level o f the INO1 transcript) is twice as high as the PA level in the same strain with introduced wild- type SEC14 (normal, regulated level of the INO1 tran- script). Similarly, the PA steady-state level is almost three times higher in the sec14D cki1D strain with a high-copy- number SFH2 gene (deregulated level of the INO1 transcript) as in the same strain in which Pld1p-mediated turnover was a bolished with pld1 disruption (normal, regulated level of the INO1 transcript). Using steady-state [ 32 P]phospholipid labelling, we measured the overall cellular levels of PA. It i s possible that in some m embrane com- partments local PA changes may far exceed those overall cellular changes of PA, thus executing the change that is critical for transmitting the signal(s) from alterations of phospholipid metabolism to the transcriptional machinery. Recently, a mechanism was described, explaining how changes in PA levels can be communicated to the tran- scriptional apparatus [ 10]. In a n elegant series of experi- ments it was shown t hat Opi1p, a negative transcriptional regulator of pho spholipid biosynthesis, i s inhibited by binding PA on the endoplasmic reticulum. Following the addition of inositol, the PA pool was rapidly consumed for the biosynthesis of PtdIns, releasing Opi1p from the endoplasmic reticulum and allowing its nuclear transloca- tion and repression of target genes, including INO1.Itis possible that because of high phospholipase D-mediated phospholipid turnover (e.g. in the sec14D cki1 D mutant) formation of P A ( de novo formation combined with PA from lipid turnover) exceeds its utilization, thus maintaining the PA pool above the threshold level for effective release of Opi1p to the nucleus. Addition of inositol to the g rowth medium under the condition of increased phospholipase D-mediated turnover, nevertheless partially repressed INO1 transcription (Figs 3 and 5) when compared to media lacking inositol. It is possible, that even under the condition of high PtdCho turnover, the addition of inositol can consume part of the PA pool, permitting that partial repression of INO1 transc ription. Alternatively, the a ddi- tion of inositol can regulate INO1 transcription to a limited degree via another mechanism that is independent of the PA levels. In summary, our results show that (a) none of the Sec14 homologues is able to substitute for Sec14p in its regulatory aspects t oward phospholipid biosynthesis, (b) removal of Pld1p activity in the strains with a high PtdCho turnover rate and resulting deregulation of the INO1 gene reverts this deregulation and s imultaneously suppresses high PtdCho turnover, a nd (c) increased steady-state PA levels corres- pond with the deregulation of the INO1 gene, supporting a previous model [6] that excess PA (in this case produced by phospholipase D -mediated PtdCho t urnover) leads to derepression of the INO1 gene. Acknowledgements We thank Vytas B ankaitis (University of North Carolina at Chapel Hill, Cha pel Hill, NC, USA), JoAnn E ngebrecht (University of California, San Diego, La Jolla, CA, USA), and Susan Henry (Cornell University, I thaca, NY, USA) for providing the strains and plasmids used in this study. The expert technical assistance of Mariana Vitekova ´ is acknowledged. This study was supported by V EGA 2/4130/4 a nd Science and Technology Assistance Agency (Slovak Republic) APVT- 51-016502 grants. References 1. Carman, G.M. & Henry, S.A. (1989) Phospholipid biosynthesis in yeast. Annu.Rev.Biochem.58, 635–669. 2. Ambroziak, J. & Henry, S.A. (1994) INO2 and INO4 gene pro- ducts, po sitive regulators of phospholipid biosynthesis in Sac- charomyces cerevisiae, form a complex that binds to the INO1 promoter. J. Biol. Chem. 269, 15344–15349. 3. White, M.J., Hirsch, J.P. & Henry, S.A. (1991) The OPI1 gene of Saccharomyce s cerevisiae, a negative regulator of phospholipid biosynthesis, encodes a protein containing polyglutamine tracts and a leucine zipper. J. Biol. Chem. 266, 863–872. 4. Donahue, T.F. & Henry, S.A. (1981) myo-Inositol-1-phosphate synthase. Characteristics of the enzyme and identification of its structural gene in yeast. J. Biol. Chem. 256, 7077–7085. 5. Greenberg, M.L. & L opes, J.M. (1996) Genetic r egulation o f phospholipid biosynthesis in Saccharomyces cerevisiae. Microbiol. Rev. 60, 1–20. 6. Henry, S.A. & Patton-Vogt, J.L. (1998) Genetic regulation of phospholipid metabolism: yeast as a model eukaryote. Prog. Nucleic Acid Res. Mol. Biol. 61, 133–179. 7. McGraw, P. & Henry, S.A. (1989) Mutations in the Saccharo- myces cerevisiae opi3 gene: effects on phospholipid methylation, growth and cross-pathway regulation of inositol synthesis. Genetics 122, 317–330. 8. Patton-Vogt, J.L., Griac, P., Sreenivas, A., Bruno, V., Dowd, S., Swede, M.J. & Henry, S.A. ( 1997) Role of the yeast phosphati- dylinositol/phosphatidylcholine transfer protein (Se c14p) in phosphatidylcholine turnover a nd INO1 regulation. J. Biol. Chem. 272, 20873–20883. 9.Sreenivas,A.,Patton-Vogt,J.L.,Bruno,V.,Griac,P.& Henry, S.A. (1998) A role for phospholipase D (Pld1p) in growth, Ó FEBS 2004 Phosphatidylcholine turnover and INO1 1 regulation (Eur. J. Biochem. 271) 4407 secretion, and regulation of membrane lipid synthesis in yeast. J. Biol. C hem. 273, 16635–16638. 10. Loewen, C.J., Gaspar, M.L., Jesch, S.A., Delon, C., Ktistakis, N.T.,Henry,S.A.&Levine,T.P.(2004) Phospholipid metabolism regulated by a transcription factor sensing phosphatidic acid. Science 304, 1644–1647. 11.Szolderits,G.,Hermetter,A.,Paltauf,F.&Daum,G.(1989) Membrane pro perties modulate the a ctivity of a pho sphatidyli- nositol transfer protein fro m the yeast, Saccharomyces cerevisiae. Biochim. Biophys. Acta 986, 301–309. 12. Bankaitis, V.A., Aitken, J.R., Cleves, A.E. & Dowhan, W. (1990) An essential role for a phospholipid tran sfer protein in yeast Golgi function. Nature 347, 561–562. 13. Cleves, A.E., McGee, T.P., Wh itters, E.A., Champion, K.M., Aitken, J.R., Dowhan, W., Goebl, M. & Bankaitis, V.A. ( 1991) Mutations in the CDP-choline pathway fo r phospholipid bio- synthesis bypass the requirement for an essential phospholipid transfer protein. Cell 64, 789–800. 14. Skinner, H.B., McGee, T.P., McMaster, C.R., Fry, M.R., Bell, R.M. & Bankaitis, V .A. (1995) The Sac charomyces cerevisiae phosphatidylinositol-transfer protein effects a ligand-dependent inhibition of choline-phosphate cytidylyltransferase activity. Proc. NatlAcad.Sci.USA92, 112–116. 15. Xie, Z., Fang, M., Rivas, M.P., Faulkner, A.J., Sternweis, P.C., Engebrecht, J.A. & Bankaitis, V.A. (1998) Phospholipase D activity is required for suppre ssion of yeast ph osphatid ylinositol transfer pro tein d efect s. Proc. Natl Acad. Sci. USA 95, 12346– 12351. 16. Rudge, S .A., Sciorra, V.A., Iwamo to, M., Zhou, C., Strahl, T., Morris, A.J., T horner, J. & Engebrecht, J. (2004) Roles of phosphoinositides and of Sp o14p (phospholipase D)-generated phosphatidic acid during yeast sporulation. Mol. Biol. Cell. 15, 207–218. 17. Li, X ., Routt, S.M., X ie, Z., Cui, X., Fang, M., Kearns, M.A ., Bard, M., Kirsch, D.R. & Bankaitis, V.A. (2000) Identification of a novel family of nonclassic yeast pho sphatidylinosito l transfer proteins whose function modulates phospholipase D activity and Sec14p-independent cell growth. Mol. Biol. Cell. 11 , 1989–2005. 18. Schnabl, M., Oskolkova, O.V., Holic, R., Brezna, B., Pichler, H., Zagorsek,M.,Kohlwein,S.D.,Paltauf,F.,Daum,G.&Griac,P. (2003) Subcellular localization of yeast Sec14 homologues and their involvement in regulation of phospholipid turnover. Eur. J. Biochem. 270, 3133–3145. 19. Lopes,J.M.,Hirsch,J.P.,Chorgo,P.A.,Schulze,K.L.&Henry, S.A. (1991) Analysis of sequences in the INO1 promoter that are involved in its r egulation by phospholipid precursors. Nucleic Acids Res. 19, 1687–1693. 20. Griac, P., Swede, M.J. & Henry, S.A. (1996) The role of phos- phatidylcholine biosynthesis in the regulation of the INO1 gene of yeast. J. Biol. Chem. 271, 25692–25698. 21. Burke, D., Dawson, D. & Stearns, T. (2000) Methods in Yeast Genetics, a Cold Spring Harbor Laboratory Course Manual.Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 22.Gietz,D.,StJean,A.,Woods,R.A.&Schiestl,R.H.(1992) Improved method for high efficiency transformation of intact yeast cells. Nucleic Acids Res. 20, 1425. 23. Gietz, R.D. & Sugino, A. (1988) New yeast–Escherichia coli shuttle v ectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74, 527–534. 24. Phillips, S.E., Sha, B ., Topalof, L., Xie, Z., Alb, J.G., Klenchin, V.A.,Swigart,P.,Cockcroft,S.,Martin,T.F.,Luo,M.&Bank- aitis, V.A. (1999) Yeast Sec14p deficient in phosphatidylinositol transfer activity is functional in vivo. Mol. Cell. 4, 187–197. 25. Greenberg, M.L., Reiner, B. & Henry, S.A. (1982) Regulatory mutations of i nositol biosynthesis in yeast: isolation of inositol- excreting mutants. Genetics 100, 1 9–33. 26. Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J. (1951) P rotein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275. 27. Dowd, S.R., Bier, M.E. & Patton-Vogt, J .L. (2001) Turnover of phosphatid ylcho lin e in Saccharomyces cerevisiae. The role of the CDP-choline pathway. J. Biol. Chem. 276, 3756–3763. 28. Cook, S.J. & Wakelam, M.J. (1989)Analysisofthewater-soluble products of pho sphatidylcholine breakdown by ion-exchange chromatography. Bombesin an d TPA (12-O-tetradecanoyl- phorbol 13-ace tate) stimu late choline generation in Swiss 3T3 cells by a common mechan ism. Biochem. J. 263, 581–587. 29. Atkinson, K.D., Jensen, B., Kolat, A.I., Storm, E.M., Henry, S.A. & Fogel, S . (1980) Yeast mutants aux otrophic for choline or ethanolamine. J. Bacteriol. 141, 558–564. 30. Steiner, M.R. & Lester, R.L. (1972) In vitro studies of phospho- lipid biosynthesis in Saccharom yces cerev isiae. Biochim. Biophys. Acta 260, 222–243. 31. Sha, B., Phillips, S.E., Bankaitis, V.A. & Luo, M. (1998) Crystal structure of the Saccharomyces cerevisiae phosphatidylinositol- transfer protein. Nature 391, 506–510. 32. Kearns, M.A., Monks, D.E., Fang,M.,Rivas,M.P.,Courtney, P.D., Chen, J., Prestwich, G .D., Theibert, A.B., Dewey, R.E. & Bankaitis, V.A. (1998) Novel developmentally regulated phos- phoinositide bind ing proteins from soybean whose expression bypasses t he requirement for an essential phosphatidylinositol transfer protein in yeast. EMBO J. 17, 4004–4017. 33.Athenstaedt,K.,Weys,S.,Paltauf,F.&Daum,G.(1999) Redundant systems o f phosphatidic acid b iosynthesis via acylation of glycerol-3-phosphate or dihydroxyacetone phos- phate in the yeast Saccharomyces cerevisiae. J. Bacteriol. 181, 1458–1463. 4408 R. Holic ˇ et al.(Eur. J. Biochem. 271) Ó FEBS 2004 . Regulation of phospholipid biosynthesis by phosphatidylinositol transfer protein Sec14p and its homologues A critical role for phosphatidic acid Roman Holic ˇ , Milos ˇ Za ´ gors ˇ ek and. transcriptional regulation of phos- pholipid biosynthesis. The results show that (a) none of the Sec14 homologues w as able to substitute for Sec14p in its regulatory aspects of phospholipid biosynthesis, . for Sec14p [17]. Our goal was to determine whether any of the Sec14 homologues could substitute for Sec14p in its role as a regulator of phospholipid biosynthesis. Under conditions where Sec14p