Genetics, Vol. 149, 1707-1715, August 1998, Copyright © 1998

GIT1, a Gene Encoding a Novel Transporter for Glycerophosphoinositol in Saccharomyces cerevisiae

J. L. Patton-Vogta and S. A. Henrya
a Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213

Corresponding author: J. L. Patton-Vogt, 4400 Fifth Ave., Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, PA 15213., jp5s{at}andrew.cmu.edu (E-mail).

Communicating editor: A. G. HINNEBUSCH


*  ABSTRACT
*TOP
 *ABSTRACT
*MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

Phosphatidylinositol catabolism in Saccharomyces cerevisiae cells cultured in media containing inositol results in the release of glycerophosphoinositol (GroPIns) into the medium. As the extracellular concentration of inositol decreases with growth, the released GroPIns is transported back into the cell. Exploiting the ability of the inositol auxotroph, ino1, to use exogenous GroPIns as an inositol source, we have isolated mutants (Git-) defective in the uptake and metabolism of GroPIns. One mutant was found to be affected in the gene encoding the transcription factor, SPT7. Mutants of the positive regulatory gene INO2, but not of its partner, INO4, also have the Git- phenotype. Another mutant was complemented by a single open reading frame (ORF) termed GIT1 (glycerophosphoinositol). This ORF consists of 1556 bp predicted to encode a polypeptide of 518 amino acids and 57.3 kD. The predicted Git1p has similarity to a variety of S. cerevisiae transporters, including a phosphate transporter (Pho84p), and both inositol transporters (Itr1p and Itr2p). Furthermore, Git1p contains a sugar transport motif and 12 potential membrane-spanning domains. Transport assays performed on a git1 mutant together with the above evidence indicate that the GIT1 gene encodes a permease involved in the uptake of GroPIns.

PHOSPHATIDYLINOSITOL (PI) is an essential membrane component of the yeast Saccharomyces cerevisiae. While PI acts as a precursor to several other phospholipids, including sphingolipids [IPC, MIPC, M(IP)2C] and polyphosphoinositides (PIP, PI 3-P, PIP2), it can also be deacylated to form extracellular glycerophosphoinositol (GroPIns). The production of extracellular GroPIns is a major catabolic pathway in S. cerevisiae, accounting for ~50% of the phosphorus and inositol lost from PI during growth in rich medium (ANGUS and LESTER 1972 Down). The production of extracellular GroPIns is regulated by both glucose and inositol (ANGUS and LESTER 1975 Down; PATTON et al. 1995 Down) and is postulated to be the result of the activity of cell surface phospholipases localized in the plasma membrane and/or periplasmic space (ANGUS and LESTER 1975 Down). S. cerevisiae not only releases GroPIns into the medium but is also capable of transporting GroPIns back into the cell in an energy-requiring process regulated by the amount of inositol in the medium (PATTON et al. 1995 Down).

Phosphoinositide deacylation, resulting in the production of lysophosphoinositols and GroPIns, has important biological consequences in mammalian cells. For example, lysophosphatidylinositol has been shown to have mitogenic activity in k-ras-transformed epithelial cells (FALASCA and CORDA 1994 Down). Furthermore, the level of GroPIns increases during hemopoietic cell differentiation (MOUNTFORD et al. 1994 Down). Similarly, during transformation by ras and other oncogenes, the amounts of GroPIns and polyphosphoinositide deacylation products increase (VALITUTTI et al. 1991 Down; CORDA and FALASCA 1996 Down). GroPIns-4-P has been reported to be an inhibitor of adenylate cyclase in thyroid cells (IACOVELLI et al. 1993 Down), and its formation has been shown to be hormone-induced (FALASCA et al. 1997 Down). In S. cerevisiae, the addition of glucose to glucose-starved cells stimulates the production of extracellular GroPIns, GroPIns-4-P, and GroPIns-4,5-P2 (HAWKINS et al. 1993 Down). This suggests that the nutrient signal of glucose refeeding activates a phospholipase(s) and/or lysophospholipase that act to deacylate the respective phosphoinositides. Thus, S. cerevisiae would appear to be a good system in which to study phosphoinositide deacylation and subsequent metabolic events.

S. cerevisiae requires a source of inositol to synthesize the inositol-containing phospholipids. The required inositol can be obtained from endogenous biosynthesis or by transporting exogenous inositol into the cell. Inositol-1-phosphate synthase carries out the rate-limiting step in inositol biosynthesis and is encoded by the INO1 gene (CARMAN and HENRY 1989 Down). By exploiting the ability of the inositol auxotroph, ino1, to use exogenous GroPIns as an inositol source, we have isolated mutants defective in the uptake and subsequent metabolism of GroPIns. One such mutant is defective in a single gene of previously unassigned function termed GIT1 (glycerophosphoinositol), which appears to encode a transporter for GroPIns. To our knowledge, this is the first report of such a transporter in a eukaryotic cell. Furthermore, our analysis of the ability of yeast cells to take up GroPIns as a source of inositol provides a novel phenotype (Git-) that is displayed in some mutants with an Ino- phenotype (i.e., spt7 and ino2 mutants) and not in others (i.e., ino1 and ino4 mutants).


*  MATERIALS AND METHODS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
*RESULTS
 *DISCUSSION
 *LITERATURE CITED

Materials:
Materials and sources are as follows: crude soybean PI, GroPIns (Sigma Chemical Co., St. Louis); silica gel plates, GF/C filters (Whatman); Bio-Rex 70 (Bio-Rad, Richmond, CA); yeast media components (Difco, Detroit); oligonucleotides (GIBCO, Grand Island, NY); and glycerophospho-myo-[2-3H]inositol (American Radiolabeled Chemicals Inc., St. Louis).

Isolation of GroPIns from crude soybean PI:
Soybean PI (10 mg/plate) suspended in chloroform:methanol (1:1) was spotted onto 1 mM silica gel preparative plates (Whatman). The chromatograms were run in chloroform:methanol:acetic acid:water (32:4:5:1) and the lipids visualized with I2 vapor. The band running at the same Rf as standard PI (Sigma Chemical Co.) was marked and scraped into a test tube following evaporation of the iodine. The PI was eluted from the silica gel with chloroform:methanol:water (16:16:5), the tube was centrifuged to sediment the silica gel, and the solvent containing PI was dried under N2. The PI was resuspended in methanol:toluene (1:1) and subjected to mild alkaline methanolysis as described previously (LESTER and STEINER 1968 Down). The resulting GroPIns was desalted over a column of Bio-Rex 70 (50–100 mesh, sodium form), eluting with H2O.

Strains and culture conditions:
Strains (Table 1) were grown aerobically at 30° with shaking. Turbidity was monitored by measurement at A650 nM on a Beckman (Foster City, CA) DU 64 spectrophotometer. Synthetic complete media were prepared as described previously (PATTON et al. 1995 Down). Some media were supplemented with the indicated amounts of myo-inositol or GroPIns. YEPD media consisted of 20 g glucose, 10 g yeast extract, and 20 g bacto peptone per liter. The wild-type strain (trp1 ura3 leu2 his3 MAT{alpha}) and the ino1 (JP3) strain (trp1 ura3 leu2, his3, ino1::HIS3 MAT{alpha}) were provided by P. MCGRAW (University of Maryland, College Park, MD). Strain T303 (his3 leu2 ura3 ade2 MATa) was provided by G. CARMEN (Rutgers University, New Brunswick, NJ). Strain spt7 (spt7::LEU2 his4 leu2 ura3 MAT{alpha}) and the YCp50-based plasmid containing the SPT7 gene were provided by K. ARNDT (University of Pittsburgh, PA).


 
View this table:
In this window
In a new window

  		Table 1. Strains

Mutagenesis and genetic manipulations:
The ino1 (JP3) strain was subjected to ethylmethane sulfonate (EMS) mutagenesis as described by LINDEGREN et al. 1965 Down with a survival rate of 40%. Cells were spread onto YEPD plates and incubated at 30° for 3 days. Colonies were replica printed onto synthetic media (i) lacking inositol (I-), (ii) containing 75 µM inositol (I+), and (iii) lacking inositol but containing 25 µM GroPIns (GroPIns+). After 3–4 days incubation at 30°, colonies able to grow on I+ plates but unable to grow on I- plates and GroPIns+ plates were selected. Two mutants isolated in this manner were backcrossed to an inositol auxotroph of the opposite mating type (JP16), and strains git1 ino1 (JPM1) and git2 ino1 (JPM2) were isolated from the resulting diploids by tetrad dissection.

Mating, sporulation, and tetrad dissection were performed using standard methodologies (ROSE et al. 1990 Down). Sporulation medium was supplemented with 100 µM inositol, since diploid strains that are inositol auxotrophs require exogenous inositol for sporulation (SCHROEDER and BREITENBACH 1981 Down).

Bacterial and yeast transformations:
Bacterial strains were transformed with plasmid DNA using calcium chloride (ROSE et al. 1990 Down), and yeast strains were transformed using lithium acetate (ROSE et al. 1990 Down).

Isolation of the GIT1 gene:
The GIT1 gene was cloned by complementation of the GroPIns auxotrophy of the git1 ino1 (JPM1) mutant. DNA from the yeast genomic library YCp50-LEU2 (obtained from P. HIETER) was transformed into the git1 ino1 strain. Approximately 40,000 transformants were screened for leucine and GroPIns prototrophy and for inositol auxotrophy. Two colonies were isolated that contained a single complementing plasmid. The plasmid was recovered from yeast and amplified in Escherichia coli. Since the YCp50 library was constructed by inserting yeast DNA in the BamHI site of the tetracycline resistance gene (tetr) of the vector, an 18-base primer derived from a section of the tetr gene ~40 bases from the BamHI site was used for sequencing (courtesy of MARK HILLER, Carnegie Mellon University, Pittsburgh, PA). Using the Saccharomyces Genome Database (SGD), the open reading frames (ORFs) contained on the complementing plasmid were analyzed. A 2-kb BamHI to ClaI fragment, containing a single ORF and 406 bp upstream of ATG, was subcloned into the vector pRS314 (SIKORSKI and HIETER 1989 Down). The resulting plasmid, pJP100, was shown to contain the complementing activity. This ORF is YCR098C in the SGD, and the GenBank accession number is X59720. The identity of the gene contained in pJP100 was verified by sequencing into the gene on both ends using the universal priming sites (T7, T3) contained in the pRS314 plasmid.

Isolation of the GIT2/SPT7 containing plasmid:
The GIT2 gene was cloned by complementation of the GroPIns auxotrophy of the git2 ino1 (JPM2) mutant. DNA from the yeast genomic library YCp50-LEU2 was transformed into the git2 ino1 (JPM2) strain. The transformation efficiency of this strain was quite low and, out of those transformants, a relatively high percentage contained the complementing plasmid. Approximately 200 transformants were screened for leucine and GroPIns prototrophy and for inositol auxotrophy. Two colonies were isolated that contained a single complementing plasmid. The plasmid was recovered from yeast and amplified in E. coli. Sequencing was performed using a primer derived from the tetr gene, as described for the sequencing of the GIT1 gene. With the aid of SGD, the ORFs contained on the complementing plasmid were determined.

DNA sequencing:
DNA was sequenced with an ABI PRISM 377 automatic DNA sequencer (University of Pittsburgh Research Support Facilities).

Construction of GIT1 disruption alleles:
A git1::HIS3 disruption allele in which 1 kb of the GIT1 ORF was replaced by HIS3 was constructed using a PCR-targeting approach (LORENZ et al. 1995 Down). Two bifunctional oligos were designed (GIBCO) that consisted of 45 nt at their 5' ends homologous to the target sequences. The 3' ends contained 20 nt homologous to the flanking regions of the marker gene. The marker gene was amplified from vector pRS303 using these primers. The resulting PCR product was used for transformation using a one-step gene disruption procedure. Integration at the correct locus would result in a disrupted git1 locus in which nucleotides 288–1226 were removed and replaced with a 1.3-kb fragment of pRS303 bearing the HIS3 gene. Histidine prototrophs were screened by PCR to verify integration at the correct locus.

Transport assays:
Cell cultures were grown in I- synthetic media to log-phase (A650 of 0.3 to 0.8). The cells were harvested, washed with fresh media, and resuspended to an A650 of 5 in fresh media. Assays were started by adding 50 µl of glycerophospho[2-3H]inositol at the indicated concentration to 200 µl of cell suspension. Assays were carried out for 5 min at 30°. Following termination of the assays by the addition of 10 ml ice-cold water, cells were collected by filtration through glass fiber filters. The filters were washed with 20 ml water and subjected to liquid scintillation counting. For the experiments of Table 2, assays were performed with 10 µM glycerophospho[2-3H]inositol and a 40-fold excess (400 µM) of the indicated nonradiolabeled compounds. The experiments involving GroPIns transport (Figure 2), and GroPIns production and reutilization (Figure 3) were performed three times with essentially identical results. Because these processes displayed some variation with slight changes in growth rate and culture density, statistical analysis was not performed on the independent experiments; only a single representative experiment is presented.



View larger version (15K):
In this window
In a new window
Download PPT slide
  		Figure 1. Growth using (A) GroPIns or (B) inositol as an inositol source. Strains were inoculated at A650 = 0.005 into synthetic media containing either 50 µM GroPIns or 50 µM inositol. At the indicated times, A650 was measured. Wild type, {square}; ino1 (JP3), {bullet}; git1ino1 (JPM1), {blacksquare}; git 2 ino1 (JPM2), {circ}.



View larger version (54K):
In this window
In a new window
Download PPT slide
  		Figure 2. Complementation of git1 mutant with pJP100. Wt, ino1 (JP3), git1ino1 (JPM1), and git1ino1 (JPM1) containing pJP100 (git1ino1 + pJP100) were spotted onto plates lacking an inositol source (I-G-), containing 75 µM inositol but lacking GroPIns (I+G-), and lacking inositol but containing 75 µM GroPIns (I-G+). Plates were allowed to incubate at 30° for 4 days.



View larger version (18K):
In this window
In a new window
Download PPT slide
  		Figure 3. GroPIns transport is greatly decreased in a git1::HIS3 (JP89) deletion mutant. Transport assays were performed as described in MATERIALS AND METHODS with the indicated concentrations of GroPIns. Wild type, {square}; git1::HIS3 (JP89), {bullet}.


 
View this table:
In this window
In a new window

  		Table 2. Effect of related compounds on GroPIns transport

Determination of GroPIns and inositol in the media:
Cells were inoculated in synthetic medium containing 10 µM inositol and 3 µCi/ml [3H]inositol. At various time points, 1-ml aliquots of culture were removed and centrifuged to pellet the cells. The resulting supernatants were analyzed for inositol and GroPIns using anion exchange chromatography, as described previously (PATTON et al. 1995 Down). In short, the supernatants were passed through Dowex-1 resins, the resins were washed with water to remove all free [3H]inositol, and [3H]GroPIns was eluted from the resin with 5 mM sodium tetraborate, 60 mM ammonium formate. Liquid scintillation counting was performed in a Beckman (Fullerton, CA) LS5801 liquid scintillation counter using Ecolume liquid scintillation cocktail (ICN Biomedicals).


*  RESULTS
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
*DISCUSSION
 *LITERATURE CITED

Identification of mutants defective in GroPIns utilization:
The inositol auxotroph, ino1 (JP3), is capable of using either inositol or GroPIns as an inositol source (Git+ phenotype). Following EMS mutagenesis, approximately 8,000 colonies were screened for growth on media containing inositol and on media containing GroPIns. Previously, we reported that GroPIns can enter the cell via both inositol transporter-dependent and -independent pathways (PATTON et al. 1995 Down). Strains bearing a deletion in the minor inositol transporter gene, ITR2, do not display aberrant growth on inositol or GroPIns in an ino1 background, so we did not expect to reisolate itr2 mutants in this screen. However, during the screening, three mutants exhibiting poor growth in the presence of inositol were eliminated to avoid the reanalysis of itr1 mutants defective in the major inositol transporter (NIKAWA et al. 1991 Down). Focusing only on mutants able to grow in the presence of inositol, but unable to grow when GroPIns was supplied as the sole source of inositol (Git- phenotype), seven mutants were identified for further study. The growth characteristics of two such mutants, git1 ino1 (JPM1) and git2 ino1 (JPM2), wild type, and the parental ino1 (JP3) strain are shown in Figure 1.

Genetic analysis:
To determine whether the mutations conferring the Git- growth phenotype were dominant or recessive with respect to wild type, the mutants were crossed to an ino1 (Git+) strain of the opposite mating type (JP16) and were examined for growth on GroPIns. All such diploids grew as well on GroPIns as the parent strain, ino1 (JP3), demonstrating that the Git- growth phenotype was recessive; one mutant was dropped from further genetic analysis because of its inability to mate. To determine whether the mutant phenotype was conferred by a single gene, the diploids heterozygous for the mutation conferring the Git- phenotype but homozygous for ino1 were sporulated, and the resulting tetrads analyzed for growth on GroPIns. One candidate was eliminated because of failure of its progeny to show 2:2 segregation of the Git- phenotype. Three candidates were eliminated from further genetic analysis because of the inability of the corresponding diploids to produce tetrads containing four viable spores. In progeny derived from two diploids simultaneously homozygous for ino1 and heterozygous for git1 or git2, the Git- mutant growth phenotype segregated 2:2 as expected for a phenotype conferred by a mutation in a single gene, although the sporulation efficiency was low. It is known that diploid strains homozygous for the ino1 mutation cannot sporulate unless inositol is supplemented (SCHROEDER and BREITENBACH 1981 Down). However, the addition of 100 µM exogenous inositol did not completely cure the sporulation defect in our Git+ Ino- (i.e., ino1 homozygous) strains. Nevertheless, out of the five full tetrads that we were able to obtain from the git1 ino1 x GIT1 ino1 (JP16) diploid and eight tetrads derived from the git2 ino1 x GIT2 ino1 (JP16) diploid, all segregated 2:2 with regard to growth on GroPIns. The git1 ino1 (JPM1) and git2 ino1 (JPM2) segregants obtained from those crosses were crossed to each other. In each case, the resulting diploids were able to grow on GroPIns, indicating that the mutations harbored in the git1 ino1 (JPM1) and git2 ino1 (JPM2) strains complement each other and are, thus, most likely to reside at different genetic loci.

GroPIns transport activity is down-regulated by the availability of inositol in the media and up-regulated by the absence of inositol (PATTON et al. 1995 Down). The phospholipid biosynthetic genes are known to be regulated in a similar fashion by the negative regulatory protein encoded by the OPI1 gene and by the positive transcription factors encoded by the INO2 and INO4 genes (CARMAN and HENRY 1989 Down). Thus, we tested the positive regulatory mutants ino2 (SH303) and ino4 (SH307) for their ability to grow on GroPIns. The ino4 (SH307) mutant was able to grow on GroPIns while the ino2 (SH303) mutant was not (data not shown), suggesting that growth on GroPIns requires the INO2 gene product but not the INO4 gene product. To determine if an ino2 mutant was reisolated in our screen, an ino2 (SH303) inositol auxotroph was crossed to the Git- mutants and the resulting diploids were tested for their ability to grow on GroPIns. In each case, the diploids were able to grow on GroPIns, indicating that the mutations harbored in these Git- strains were not likely to be ino2 alleles.

GIT2 is allelic to SPT7:
The git2 mutation was complemented using a YCp50-based genomic library. Partial sequencing of the complementing plasmid followed by data analysis utilizing the SGD localized the complementing activity to a 7-kb fragment on chromosome II. A single ORF (previously identified as SPT7) included within that fragment was focused upon, as it appeared most likely to contain the complementing activity. Based upon the following facts (data not shown), GIT2 was determined to be identical to SPT7: (1) A YCp50 plasmid containing a 7-kb EcoRI-ClaI fragment encompassing the SPT7 gene (GANSHEROFF et al. 1995 Down) was able to complement the git2 mutation. (2) A characterized spt7 deletion mutant, which is an inositol auxotroph (GANSHEROFF et al. 1995 Down), was crossed with an ino1 strain (JP16) and sporulated, and random spore analysis was performed since the sporulation efficiency was quite poor. Both spt7 INO1 and spt7 ino1 (JP61) segregants derived from this cross were unable to grow on GroPIns. (3) Diploids produced by crossing an spt7 ino1 strain (JP61) with two different git2 ino1 segregants [JP2-9(5C) and JP2-9(9A)] from a cross of the original git2 ino1 mutant to a GIT1 ino1 (JP16) strain, were unable to grow on GroPIns and failed to sporulate. Since diploids homozygous for an spt7 mutation have also been reported to be sporulation defective (GANSHEROFF et al. 1995 Down), further genetic analysis on these diploids was not attempted. The plasmid bearing the SPT7 gene was not able to complement the Git- phenotype of any of the six other mutants isolated in this study.

Cloning and disruption of GIT1:
The git1 mutation was complemented using a YCp50-based genomic library. Partial sequencing of the complementing plasmid allowed the GIT1 gene to be localized to the right arm of chromosome III. A single ORF (YCR098C), labeled as "probable metabolite transport protein" in the SGD, was subcloned into vector pRS314 to produce pJP100 and was transformed into the git1 ino1 mutant. This subclone contained the complementing activity (Figure 2). However, pJP100 was not able to complement the Git- phenotype of any of the six other mutants isolated in this study. The GCG BestFit program (University of Wisconsin, Madison, WI), revealed significant sequence homology between the predicted amino acid sequence of the GIT1 ORF and the amino acid sequences of the S. cerevisiae transport proteins Pho84p, Itr1p, and Itr2p. Also, an analysis of the GIT1 promoter region revealed the presence of a core basic helix-loop-helix binding motif, CACGTG (CROWTHER et al. 1996 Down), and three Stress Response Element consensus sequences, AGGG or CCCT (RUIS and SCHULLER 1995 Down), which could potentially be involved in transcriptional regulation.

A disruption allele of the cloned gene was created by replacing a 938-bp region in the middle of the ORF by a 1.3-kb DNA fragment bearing the HIS3 gene; this fragment was used to replace the chromosomal copy of GIT1 in a wild-type strain. The gene disrupted strain, git1::HIS3 (JP89), no longer containing an ino1 allele and having neither an Ino- nor a Git- phenotype, was crossed to a wild-type strain of the opposite mating type (T303) and the resulting diploid was sporulated. Since the git1 mutation confers no growth phenotype in a wild-type genetic background (i.e., in the absence of an ino1 mutation), His+ prototrophy was used to follow the gene disruption, and the git1 phenotype was confirmed by assay of GroPIns transport activity. The 15 tetrads derived from this cross segregated 2:2 with regard to His+ prototrophy. All four spores in each of three complete tetrads were assayed for GroPIns transport activity. The lack of GroPIns transport activity cosegregated with His+ prototrophy, and all His- spores showed wild-type transport activity (data not shown). Furthermore, when the git1 ino1 (JPM1) mutant was crossed to a strain of the git1::HIS3 ino1 genotype [derived from sporulating a git1::HIS3 (JP89) x ino1 (JP16) diploid], the diploid was unable to grow on GroPIns, confirming that the constructed deletion mutation was most likely allelic to the original git1 mutation.

Phenotypic analysis of git1::HIS3 mutant:
The doubling times and ultimate culture densities of the git1::HIS3 INO1 (JP89) strain were nearly identical to wild type when it was grown in rich medium (YEPD) as well as in defined I+, I-, and GroPIns+ media. The portion of chromosome III containing the ORF YCR098C had been analyzed as part of the genome sequencing project, and, consistent with the above observations, it was reported that disruption of this ORF results in no obvious phenotype as compared to wild type when tested for growth on lactose, lactate, or glycerol media. It also reportedly showed no osmotic or detergent sensitivity. Sporulation of the heterozygous and homozygous git1 diploid strains was also reportedly normal (SOR et al. 1992 Down), and we likewise observed no defect in the sporulation of git1/git1 diploids that are Ino+ (i.e., not homozygous for ino1).

GroPIns transport activity:
At concentrations of 10 µM and below, GroPIns transport ability was virtually abolished in the git1::HIS3 (JP89) strain (Figure 3). At concentrations above 10 µM, the git1::HIS3 (JP89) strain displays some transport capability but much less than the wild-type strain. This observed GroPIns transport activity at high GroPIns concentrations is most likely the result of transport occurring through another, lower affinity permease, such as an inositol transporter (PATTON et al. 1995 Down).

The experiments shown in Figure 3 were performed on cultures grown in the absence of inositol, since inositol was shown to repress GroPIns transport (PATTON et al. 1995 Down). In fact, wild-type cells grown in the presence of 75 µM inositol and assayed with 10 µM [3H]GroPIns exhibit only 6% of the GroPIns transport activity observed for cells grown in the absence of inositol.

GIT1 substrate specificity:
The substrate specificity of the Git1p was assessed by examining the ability of various compounds to inhibit GroPIns transport (Table 2). Transport assays were performed with 10 µM [3H]GroPIns and a 40-fold excess (400 µM) of nonradiolabeled compounds. Glycerol and inositol had virtually no effect upon GroPIns transport. GroPC and GroPE inhibited transport to some extent, and glycerol-3-phosphate and GroPS caused a marked inhibition of GroPIns transport (Table 2).

Production of extracellular GroPIns and its reutilization through GIT1:
A dynamic relationship exists between extracellular inositol concentration and the production and subsequent reutilization of extracellular GroPIns. Wild-type and git1::HIS3 (JP89) cells were inoculated in synthetic media containing 10 µM [3H]inositol, and the concentrations of extracellular inositol and GroPIns were analyzed as the cultures progressed (Figure 4). Both strains displayed an initial production of GroPIns. However, when the level of inositol reached approximately 1 µM (A650 {approx} 1), the wild-type strain began to transport GroPIns back into the cell, as evidenced by the decrease in extracellular GroPIns levels. In contrast, the GroPIns concentration in the git1::HIS3 (JP89) culture began to level off at 1 µM inositol, but there was no subsequent decrease corresponding to GroPIns reuptake.



View larger version (13K):
In this window
In a new window
Download PPT slide
  		Figure 4. A git1 deletion mutant (JP89) cannot reutilize GroPIns when the medium becomes depleted of inositol. Cells were inoculated in synthetic medium containing 10 µM inositol and 3 µCi/ml [3H]inositol. At the indicated times, the culture medium was analyzed for inositol (A) and GroPIns (B). Wild type, {blacksquare}; git1::HIS3 (JP89), {square}.


*  DISCUSSION
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
*LITERATURE CITED

The GIT1 gene (YCR098C) resides on the right arm of chromosome III. It consists of 1556 bp encoding a 518-amino acid protein with a predicted molecular mass of 57.3 kD. A KYTE and DOOLITTLE 1982 Down hydropathy profile of the deduced amino acid sequence (SOR et al. 1992 Down) indicates the presence of 12 potential membrane-spanning regions and hydrophilic N and C termini. A sugar transport motif [SDRIGR(K/R)(4-5)G] at amino acid 329 was also identified (SOR et al. 1992 Down). The GIT1 gene was originally sequenced as part of the S. cerevisiae genome sequencing project (this ORF was first labeled as YCR137 but is now YCR098C), but no function was assigned to it other than homology to carbohydrate transport proteins (SOR et al. 1992 Down). Within the S. cerevisiae genome, GIT1 bears similarity to a number of other metabolite transport proteins including the inorganic phosphate transporter, PHO84, and the inositol transporters, ITR1 and ITR2. Using the algorithm of SMITH and WATERMAN 1981 Down with gap weight 3.0 and gap weight length 0.1, GIT1 is predicted to have 49.6% similarity and 26.2% identity with PHO84, 51.1% similarity and 23.7% identity with ITR1, and 50.8% similarity and 26.2% identity with ITR2.

The git1::HIS3 mutation eliminates measurable GroPIns transport activity as defined as transport below 10 µM GroPIns (Figure 3). Thus, we infer that this specific GroPIns transport activity is carried out by the GIT1 gene product, Git1p. We have shown previously that GroPIns can be transported into the cell intact, without first being hydrolyzed to inositol extracellularly (PATTON et al. 1995 Down). Consistent with this finding is the fact that at a concentration 40 times that of radiolabeled GroPIns, unlabeled inositol and glycerol had virtually no effect upon GroPIns transport (Table 2). Thus, inositol and glycerol appear to have little or no affinity for Git1p while GroPC and GroPE may have limited affinity because they do cause some inhibition of GroPIns transport. However, both glycerol-3-phosphate and GroPS caused marked inhibition, suggesting that they have significant affinity for Git1p. It is interesting to note that GroPS and glycerol-3-phosphate, like GroPIns, are negatively charged glycerophosphate esters. Whether GroPS and glycerol-3-phosphate are actually transported by Git1p remains to be established.

At GroPIns concentrations below 10 µM, the git1::HIS3 (JP89) mutant cannot transport GroPIns, but above this concentration transport is observed (Figure 3). We attribute the residual transport seen in the git1::HIS3 (JP89) deletion mutant and in the wild-type strain at higher concentrations to the action of other nonspecific transporter(s), most likely the major inositol transporter (PATTON et al. 1995 Down). In synthetic medium containing nonlimiting amounts of inositol, the extracellular GroPIns concentration reaches 1–2 µM per A650 of culture during the course of growth (Figure 4; ANGUS and LESTER 1972 Down). Thus, under normal laboratory conditions the concentration of extracellular GroPIns in yeast cultures does not go above 10 µM, making Git1p the primary GroPIns transporter under normal physiological conditions. This pathway, by which extracellular GroPIns produced via PI degradation is subsequently taken up and reutilized by the cell, is dramatically affected by the concentration of inositol in the media (Figure 4). While inositol availability is required for the production of extracellular GroPIns, when extracellular inositol becomes limiting (approximately 1 µM), GroPIns is transported back into the cell through the action of the Git1p permease. The starting inositol concentration for the experiment of Figure 4 (10 µM) is approximately the inositol concentration found in standard synthetic media when using yeast nitrogen base (Difco) with or without amino acids. Thus, under normal laboratory growth conditions, there exists a dynamic interplay between extracellular inositol and extracellular GroPIns levels. In the git1::HIS3 (JP89) strain, this recycling pathway does not occur.

The genetic screen described here also resulted in the identification of SPT7, a global transcription factor (GANSHEROFF et al. 1995 Down). We have shown GIT2 to be allelic to SPT7. Mutations in SPT7 were originally isolated as suppressors of Ty and {partial} insertion mutations in the 5' region of the HIS4 and LYS2 genes (GANSHEROFF et al. 1995 Down). Although it is tempting to speculate that SPT7 is involved in the transcriptional regulation of GIT1, that has yet to be demonstrated. The INO2 gene, which encodes a positive transcriptional regulator of INO1 and other phospholipid biosynthetic genes, is also required for the utilization of GroPIns as an inositol source. Surprisingly, mutations at the INO2 locus were not identified in our screen, suggesting that the screening may not have identified all loci capable of conferring the Git- phenotype when mutated. While growth on GroPIns requires the INO2 gene product, it does not require the INO4 gene product. Other phenotypic differences between ino2 and ino4 mutants have been observed. For example, the ino4 mutant expresses a small amount of INO1 transcript under limiting inositol conditions, while the ino2 mutant does not (GRAVES 1996 Down). In contrast, the expression of a fusion gene driven by the INO4 promoter (ASHBURNER and LOPES 1995 Down) and derepression of IPC synthase activity when inositol is supplemented in the growth media require a functional INO4 but not INO2 gene product (KO et al. 1994 Down).

Although E. coli is known to have transporters for glycerophosphodiesters, this is the first report of such a transporter being identified in a eukaryotic cell. In E. coli, glycerophosphodiesters and glycerol-3-phosphate can be transported either via the pho regulon-dependent Ugp system (SCHWEIZER et al. 1982 Down; XAVIER et al. 1995 Down) or the glp regulon-dependent GlpT system (LARSON et al. 1982 Down; XAVIER et al. 1995 Down). The Ugp system is induced by phosphate starvation, while the GlpT system is induced in response to the presence of glycerol and glycerol-3-phosphate in the medium. Similarly, the transport of GroPIns via Git1p in S. cerevisiae is dependent upon the nutritional environment of the cell.

As shown previously (PATTON et al. 1995 Down), inositol causes repression of the ascribed Git1p transport activity (i.e., specific transport of GroPIns at concentrations below 10 µM). Data reported in a recent publication by WODICKA et al. 1997 Down, who performed genome-wide expression monitoring in S. cerevisiae, lend credence to the idea that GIT1 transcription is regulated by nutritional factors. These authors identified GIT1 as one of ~250 genes showing significant regulation in response to nutrient availability. GIT1 mRNA was reported to be 31-fold more abundant in minimal medium as compared to rich medium; only four other genes displayed a greater fold difference under those conditions (WODICKA et al. 1997 Down). Consistent with previously published data, other phospholipid biosynthetic genes were also reported by WODICKA et al. to be regulated by nutrient availability. The mRNA of OPI3, which is the structural gene for phospholipid-N-methyltransferase, which is coregulated with INO1, was found to be 18-fold more abundant in minimal as compared to rich medium, and INO1 mRNA was found to be highly abundant in minimal medium and undetectable in rich medium (WODICKA et al. 1997 Down). Our inspection of the promoter region (defined as 1000-bp upstream of the start codon) of GIT1 failed to detect a copy of the repeated element UASINO (CATGTG), which is found in the INO1 and OPI3 promoters, but we did find a bHLH consensus sequence (CACGTG) located at nucleotides 518–523 upstream of ATG to which INO2 and INO4 (bHLH proteins) might bind to activate transcription (BACHHAWAT et al. 1995 Down).

S. cerevisiae adjusts its metabolism to use the available sources of carbon, phosphate, sulfate, and nitrogen, among other things (JONES et al. 1992 Down). We now report that yeast cells can use the GroPIns that they have released into the medium as a result of PI turnover during normal growth, as an alternate source of inositol during times of inositol depletion. Phospholipid metabolites recycle during normal growth in yeast and in other organisms. For example, we have recently shown that when PC is acted upon by a phospholipase D in wild-type S. cerevisiae, the released choline is immediately reincorporated into PC (PATTON-VOGT et al. 1997 Down). Similarly, in Chinese hamster ovary cells it appears that ethanolamine is continually released from CDP-ethanolamine derived PE and recycled back into PE (SHIAO and VANCE 1995 Down). In yeast, the production and reutilization of GroPIns following the deacylation of PI, most likely mediated by a phospholipase B or a phospholipase A and a lysophospholipase, is a major metabolic route (ANGUS and LESTER 1972 Down, ANGUS and LESTER 1975 Down; PATTON et al. 1995 Down). Furthermore, we have shown that GroPIns production and reutilization is regulated by the nutrient status of the cell. The exact mechanism of this regulation and its significance for cellular function will be the subject of future studies. Clearly, this pathway of PI turnover must be taken into account in future studies of PI metabolism and signal transduction in yeast. In mammalian cells, the deacylation of phosphoinositides and the production of glycerophosphoinositols has already been shown to have important biological consequences (VALITUTTI et al. 1991 Down; IACOVELLI et al. 1993 Down; FALASCA and CORDA 1994 Down; MOUNTFORD et al. 1994 Down; CORDA and FALASCA 1996 Down; FALASCA et al. 1997 Down).


*  ACKNOWLEDGMENTS

We thank GEORGE CARMAN, KAREN ARNDT and PAT MCGRAW for strains used in this study. This work was supported in part by Grant GM-19629 to S.A.H. from the National Institutes of Health and by a fellowship from the American Heart Association to J.L.P.

Manuscript received January 8, 1998; Accepted for publication April 30, 1998.


*  LITERATURE CITED
*TOP
 *ABSTRACT
 *MATERIALS AND METHODS
 *RESULTS
 *DISCUSSION
 *LITERATURE CITED

ANGUS, W. W. and R. L. LESTER, 1972  Turnover of inositol and phosphorus containing lipids in Saccharomyces cerevisiae; extracellular accumulation of glycerophosphorylinositol derived from phosphatidylinositol. Arch. Biochem. Biophys. 151:483-495[Medline].

ANGUS, W. W. and R. L. LESTER, 1975  The regulated catabolism of endogenous and exogenous phosphatidylinositol by Saccharomyces cerevisiae leading to extracellular glycerophosphorylinositol and inositol. J. Biol. Chem. 250:22-30[Abstract/Free Full Text].

ASHBURNER, B. P. and J. M. LOPES, 1995  Autoregulated expression of the yeast INO2 and INO4 helix-loop-helix activator genes effects cooperative regulation on their target genes. Mol. Cell. Biol. 15:1709-1715[Abstract].

BACHHAWAT, N., Q. OUYANG, and S. A. HENRY, 1995  Functional characterization of an inositol-sensitive upstream activation sequence in yeast: A cis-regulatory element responsible for inositol-choline-mediated regulation of phospholipid biosynthesis. J. Biol. Chem. 270:25087-25095[Abstract/Free Full Text].

CARMAN, G. M. and S. A. HENRY, 1989  Phospholipid biosynthesis in yeast. Annu. Rev. Biochem. 58:635-669[Medline].

CORDA, D. and M. FALASCA, 1996  Glycerophosphoinositols as potential markers of Ras-induced transformation and novel second messengers. Anticancer Res. 16:1341-1350[Medline].

CROWTHER, D. J., N. A. KENT, A. F. SPIT, and J. MELLOR, 1996  Basic helix-loop-helix transcription factors and the cross-regulation of sulphate and phosphate metabolism in yeast. Biochem. Soc. Trans. 24:1995-1996.

FALASCA, M., A. CARVELLI, C. IURISCI, R.-G. QIU, and M. H. SYMONS et al., 1997  Fast receptor-induced formation of glycerophosphoinositol-4-phosphate, a putative novel intracellular messenger in the Ras pathway. Mol. Biol. Cell 8:443-453[Abstract].

FALASCA, M. and D. CORDA, 1994  Elevated levels and mitogenic activity of lysophosphatidylinositol in k-ras-transformed epithelial cells. Eur. J. Biochem. 221:383-389[Medline].

GANSHEROFF, L. J., C. DOLLARD, P. TAN, and F. WINSTON, 1995  The Saccharomyces cerevisiae SPT7 gene encodes a very acidic protein important for transcription in vivo.. Genetics 139:523-536[Abstract].

GRAVES, J. A., 1996 Analysis of the role of the OPI1 gene product in the negative regulation of the phospholipid biosynthetic pathway of S. cerevisiae, pp. 59–63 in Biological Sciences. Carnegie Mellon University, Pittsburgh.

HAWKINS, P. T., L. R. STEPHENS, and J. R. PIGGOTT, 1993  Analysis of inositol metabolites produced by Saccharomyces cerevisiae in response to glucose stimulation. J. Biol. Chem. 268:3374-3383[Abstract/Free Full Text].

IACOVELLI, L., M. FALASCA, S. VALITUTTI, D. D'ARCANGELO, and D. CORDA, 1993  Glycerophosphoinositol 4-phosphate, a putative endogenous inhibitor of adenylylcyclase. J. Biol. Chem. 268:20402-20407[Abstract/Free Full Text].

JONES, E. W., J. R. PRINGLE and J. R. BROACH (Editors), 1992 The Molecular and Cellular Biology of the Yeast Saccharomyces: Gene Expression. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

KO, J., S. CHEAH, and A. S. FISCHL, 1994  Regulation of phosphatidylinositol:ceramide phosphoinositol transferase in Saccharomyces cerevisiae.. J. Bacteriol. 176:5181-5183[Abstract/Free Full Text].

KYTE, J. and R. F. DOOLITTLE, 1982  A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132[Medline].

LARSON, T. J., G. SCHUMACHER, and W. BOOS, 1982  Identification of the glpT-encoded sn-glycerol-3-phosphate permease of Escherichia coli, an oligomeric integral membrane protein. J. Bacteriol. 152:1008-1021[Abstract/Free Full Text].

LESTER, R. L. and M. R. STEINER, 1968  The occurrence of diphosphoinositide and triphosphoinositide in Saccharomyces cerevisiae.. J. Biol. Chem. 243:4889-4893[Abstract/Free Full Text].

LINDEGREN, G., Y. L. HWANG, Y. OSHIMA, and C. C. LINDEGREN, 1965  Genetical mutants induced by ethylmethanesulfonate in Saccharomyces.. Can. J. Genet. Cytol. 7:491-499[Medline].

LORENZ, M. C., R. S. MUIR, E. LIM, J. MCELVER, and S. C. WEBER et al., 1995  Gene disruption with PCR products in Saccharomyces cerevisiae.. Gene 158:113-117[Medline].

MOUNTFORD, J. C., C. M. BUNCE, P. J. FRENCH, R. H. MICHELL, and G. BROWN, 1994  Intracellular concentrations of inositol, glycerophosphoinositol and inositol pentakisphosphate increase during haemopoietic cell differentiation. Biochim. Biophys. Acta 1222:101-108[Medline].

NIKAWA, J., Y. TSUKAGOSHI, and S. YAMASHITA, 1991  Isolation and characterization of two distinct myo-inositol transporter genes of Saccharomyces cerevisiae.. J. Biol. Chem. 266:11184-11191[Abstract/Free Full Text].

PATTON, J. L., L. PESSOA-BRANDAO, and S. A. HENRY, 1995  Production and reutilization of an extracellular phosphatidylinositol catabolite, glycerophosphoinositol, by Saccharomyces cerevisiae.. J. Bacteriol. 177:3379-3385[Abstract/Free Full Text].

PATTON-VOGT, J. L., P. GRIAC, A. SREENIVAS, V. BRUNO, and S. DOWD et al., 1997  Role of the yeast phosphatidylinositol/phosphatidylcholine transfer protein (Sec14p) in phosphatidylcholine turnover and INO1 regulation. J. Biol. Chem. 272:20873-20883[Abstract/Free Full Text].

ROSE, M. D., F. WINSTON and P. HIETER, 1990 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

RUIS, H. and C. SCHULLER, 1995  Stress signaling in yeast. BioEssays 17:959-965[Medline].

SCHROEDER, R. and M. BREITENBACH, 1981  Metabolism of myo-inositol during sporulation of myo-inositol-requiring Saccharomyces cerevisiae.. J. Bacteriol. 146:775-783[Abstract/Free Full Text].

SCHWEIZER, H., M. ARGAST, and W. BOOS, 1982  Characteristics of a binding protein-dependent transport system for sn-glycerol-3-phosphate in Escherichia coli that is part of the pho regulon. J. Bacteriol. 150:1154-1163[Abstract/Free Full Text].

SHIAO, Y.-J. and J. E. VANCE, 1995  Evidence for an ethanolamine cycle: differential recycling of the ethanolamine moiety of phosphatidylethanolamine derived from phosphatidylserine and ethanolamine. Biochem. J. 310:673-679.

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

SMITH, T. F. and M. S. WATERMAN, 1981  Identification of common molecular subsequences. J. Mol. Biol. 147:195-197[Medline].

SOR, F., G. CHÉRET, F. FABRE, G. FAYE, and H. FUKUHARA, 1992  Sequence of the HMR region on chromosome III of Saccharomyces cerevisiae.. Yeast 8:215-222[Medline].

VALITUTTI, S., P. CUCCHI, G. COLLETTA, C. DI FILIPPO, and D. CORDA, 1991  Transformation by the K-RAS oncogene correlates with increases in phospholipase A2 activity, glycerophosphoinositol production and phosphoinositide synthesis in thyroid cells. Cell. Signalling 3:321-332[Medline].

WODICKA, L., H. DONG, M. MITTMANN, M.-H. HO, and D. J. LOCKHART, 1997  Genome-wide expression monitoring in Saccharomyces cerevisiae.. Nature Biotechnol. 15:1359-1367[Medline].

XAVIER, K. B., M. KOSSMAN, H. SANTOS, and W. BOOS, 1995  Kinetic analysis by in vivo 31P nuclear magnetic resonance of internal Pi during the uptake of sn-glycerol-3-phosphate by the pho regulon-dependent Ugp system and the glp regulon-dependent GlpT system. J. Bacteriol. 177:699-704[Abstract/Free Full Text].