(Received for publication, January 10, 1997, and in revised form, February 19, 1997)
From the Department of Biochemistry and Molecular Biology, University of Texas-Houston Medical School, Houston, Texas 77225
Amine-containing phospholipid synthesis in Saccharomyces cerevisiae starts with the conversion of CDP-diacylglycerol (CDP-DAG) and serine to phosphatidylserine (PS), whereas phosphatidylinositol (PI) is formed from CDP-DAG and inositol (derived from inositol 1-phosphate). In this study the regulation of PS synthase (encoded by CHO1/PSS), PI synthase (encoded by PIS1), and inositol 1-phosphate synthase (encoded by INO1) activities by the in vivo level of CDP-DAG synthase activity (encoded by CDS1) is described. Reduction in the level of CDP-DAG synthase activity from 10-fold over wild type levels to 10% of wild type levels results in a 7-fold increase in PS synthase activity, which follows a similar change in the CHO1/PSS mRNA level. INO1 mRNA also increases but only after CDP-DAG synthase activity falls below the wild type level. PI synthase activity follows the decrease of the CDP-DAG synthase activity, but there is no parallel change in the level of PIS1 mRNA. These changes in CHO1/PSS and INO1 mRNA levels are mediated by a mechanism not dependent on changes in the expression of the INO2-OPI1 regulatory genes. CDS1 expression is repressed in concert with INO2 expression in response to inositol.
The major phospholipids in yeast, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine (PS),1 and phosphatidylinositol (PI), are synthesized by two branches of the phospholipid biosynthetic pathway that diverge from the common liponucleotide precursor CDP-diacylglycerol (CDP-DAG) (1-3). Synthesis of amine-containing lipids starts with the conversion of CDP-DAG and serine to PS, which is sequentially converted to phosphatidylethanolamine and phosphatidylcholine. PI is synthesized from CDP-DAG and inositol, which can be synthesized de novo from glucose 6-phosphate catalyzed by inositol 1-phosphate synthase, the product of the INO1 gene (4). PS synthase encoded by the CHO1/PSS gene (5, 6) and PI synthase encoded by the PIS1 gene (7), respectively, catalyze the above two CDP-DAG-dependent reactions. The cellular activities of PS synthase and PI synthase, together with the availability of inositol, have been considered to be the major determinants in the partitioning of CDP-DAG between these two branches of the pathway and consequently determine the final membrane phospholipid composition (8). Therefore, it is not surprising that these enzymes are regulated at multiple levels to coordinate phospholipid biosynthesis.
Studies of the regulation of phospholipid biosynthesis by inositol
demonstrate that the INO1 and CHO1/PSS genes are
derepressed in wild type cells grown in medium lacking inositol and
choline (or ethanolamine), and repressed when these precursors of
phospholipid biosynthesis are present (9, 10); choline and ethanolamine are precursors in the DAG-dependent pathway leading to
phosphatidylcholine and phosphatidylethanolamine biosynthesis (1,
3). The regulation of INO1 and CHO1/PSS gene
expression by inositol and choline has been shown to occur at the level
of mRNA abundance (11, 12) and is mediated by the 5
cis-acting UASINO (upstream activation sequence)
and the trans-acting INO2-INO4-OPI1 regulatory
gene products (10, 12). The INO2 and INO4 genes
encode positive regulators of transcription that are required for
derepression in response to inositol and choline deprivation. Ino2p is
the transcriptional activator that is dependent on dimerization with Ino4p for its binding to the UASINO sequence. Strains
carrying ino2 or ino4 mutant alleles fail to
express the INO1 gene product and thus need inositol
supplement for growth (13). INO2 expression is also
coordinately regulated by inositol (14). The OPI1 gene product is required for repression of gene expression in response to
inositol and choline supplementation. The opi1 mutant
strains constitutively overexpress the INO1 and
CHO1 genes (15). Studies using the CAT reporter
gene fused to the endogenous promoter of the OPI1 gene or
the INO2 gene suggest that OPI1 is required for the initial response to inositol but that regulation of INO2
expression establishes the magnitude and cooperativity of the response
(14, 16).
Expression of the PIS1 gene, as with the INO4 or
OPI1 genes, does not respond to inositol regulation,
although a UASINO sequence is also found 5 to the
PIS1 open reading frame (17). Instead, PIS1
expression is regulated at the transcriptional level by carbon sources
in growth medium (18). Relative to cells grown in a glucose medium,
expression of the PIS1 gene is reduced when cells were grown
in a medium containing glycerol and elevated when cells were grown in a
medium containing galactose. Carbon source regulation involves the
binding of Mcm1 protein to the cis-MCEs sequences and the
involvement of the two-component regulatory gene SLN1 (18).
We previously reported the isolation of the CDS1 gene encoding CDP-DAG synthase activity in yeast (19). By gene interruption, we showed that this gene is essential for spore germination and vegetative cell growth. Overexpression of CDP-DAG synthase activity results in an elevation in the apparent initial rate of synthesis of PI relative to PS. A reduction of CDP-DAG synthase activity to 10% of the wild type level has an opposite effect on the above phospholipid biosynthetic steps. Reduced synthase levels also resulted in inositol excretion. By characterizing the cdg1 mutant (20), we found a point mutation within the CDS1 gene that results in the cdg1 phenotype of pleiotropic deficiencies in phospholipid biosynthesis, including a reduced level of CDP-DAG synthase activity, an elevated phosphatidic acid content, a constitutive elevated level of inositol 1-phosphate synthase activity, and an elevated PS synthase activity (21). All of these results are suggestive of alteration of expression of the INO1 and CHO1/PSS genes at reduced cellular CDP-DAG synthase levels.
Cloning of the CDS1 gene affords the opportunity to study the role of CDP-DAG synthase and its product in the regulation of phospholipid metabolism. In this study, we examined the regulation of phospholipid biosynthetic enzymes by CDP-DAG synthase activity. We found that the increased PS synthase activity and the inositol excretion in response to reduced CDP-DAG synthase activity was due to an increased steady-state level of CHO1/PSS and INO1 mRNA transcripts, respectively, whereas the parallel decrease in PI synthase activity was not a result of a change in the PIS1 mRNA transcript level. However, regulation of expression of the CHO1/PSS and INO1 genes in response to CDP-DAG synthase activity levels did not correlate with factors that regulate the expression of these genes in response to inositol and choline.
Simple salts, buffers, and amino acids were purchased from Sigma and Fisher. YEP broth and synthetic media for yeast growth and selection were from Bio 101, Inc. Yeast nitrogen base without amino acids was from Difco Laboratories. Trizol reagent was a product of Life Technologies, Inc. Positively charged nylon membranes, Genius 1 kit, and reagents for nonradioactive detection of nucleic acid were purchased from Boehringer Mannheim. Radiochemicals were obtained from Amersham Corp. Oligonucleotides were prepared commercially by Genosys Biotechnologies. The Gene Amp polymerase chain reaction reagent kit was from Perkin-Elmer Cetus. Qiaex II gel extraction kit was from Qiagen Inc. CDP-DAG was synthesized by reaction of phosphatidic acid and CMP-morpholidate and purified by phase separation and silica gel chromatography (22).
Strains, Plasmids, and Growth ConditionsThe yeast strains used in this study were YPH102 (MATa ura3-52 lys2-801 ade2-101 leu2-D1 his-D200) (23), YSD90A (MATa ura3-52 lys2-801 ade2-101 leu2-D1 his-D200 cds1::TRP1) (19), and MC13 (MATa ino1-13 lys2 can1) (24). YSD90A depends on a plasmid-borne copy of the CDS1 gene for viability. MC13 is an inositol auxotroph. Plasmid pSDG2 (19) was derived from YCpGAL (25), a single copy number plasmid provided by J. Bonner. This plasmid contained the CDS1 gene under GAL1 promoter control. Plasmid phCDS (also known as pYES.CDS (26)) carries a human CDS gene on vector pYES2 (Invitrogen, San Diego, CA); expression of this gene is under GAL1 promoter control. Plasmid pBluescript KS was purchased from Stratagene (La Jolla, CA) and used for subcloning and DNA sequencing purposes. All yeast cultures were grown at 30 °C in complete synthetic medium (CSM) with the indicated nutrient drop-outs for selection purposes as described previously (19); all complete synthetic media were supplemented with the indicated carbon source and yeast nitrogen base containing 2 mg/ml (11 µM) inositol based on the Difco manual (27) except where indicated. For the study of inositol regulation, cells were maintained in a synthetic medium containing 60 µM inositol and 1 mM choline before they were shifted to an inositol- and choline-free medium (28).
Enzyme AssaysActivities of CDP-DAG synthase, PI synthase, and PS synthase were determined as described previously (21). Briefly, the total cell membrane fraction was resuspended in 20 mM Tris-maleate, pH 7.0. CDP-DAG synthase activity was measured by the incorporation of radiolabel into CDP-DAG in 50 mM Tris-maleate buffer, pH 6.5, 20 mM MgCl2, 1% Triton X-100, 1.0 mM [5-3H]CTP (9,000 cpm/nmol), 0.5 mM phosphatidic acid, and 50 µg of membrane protein in a total volume of 0.1 ml (29). PS synthase activity was measured by the incorporation of radiolabel into PS in 50 mM Tris-HCl buffer, pH 8.0, 0.6 mM MnCl2, 0.2 mM CDP-DAG, 0.25% Triton X-100, 0.5 mM L-[3-3H]serine (10,000 cpm/nmol), and 75 µg of membrane protein in a total volume of 0.1 ml (30). PI synthase activity was measured by the incorporation of 0.5 mM myo-[2-3H]inositol (10,000 cpm/nmol) into PI in 50 mM Tris-HCl buffer, pH 8.0, 2 mM MnCl2, 0.2 mM CDP-DAG, 0.15% Triton X-100, and 75 µg of membrane protein in a total volume of 0.1 ml (22). All assays were carried out at 30 °C for either 5 min (CDP-DAG synthase) or 20 min (PS and PI synthases).
Probes for RNA AnalysisDouble-stranded DNA probes were used for Northern blot analysis. The probe for CDS1 detection was a BstEII-BglII fragment from plasmid pSDG1 (19). The remaining probes used in this study were polymerase chain reaction products generated using YPH102 genomic DNA as a template and the polymerase chain reaction primers listed in Table I designed according to the DNA sequences of CHO1/PSS (5), INO1 (4), INO2 (31), OPI1 (32), PIS1 (7), and TCM1 (33). Each probe was subcloned into plasmid pBluescript KS and verified to be correct by DNA sequencing before being labeled by random priming and incorporation of digoxigenin-dUTP using the Genius 1 kit according to the manufacturer's instructions. Concentrations of labeled probes for individual detections were 5 ng/ml TCM1, 20 ng/ml CHO1, INO1, PIS1, and 50 ng/ml INO2, OPI1.
|
Yeast cells were harvested at different time
points and were frozen at 80 °C before RNA preparation. Total RNA
was isolated from these samples by a glass-bead disruption/Trizol
extraction procedure. Yeast pellets from 10 ml of mid-log phase cell
culture were mixed with 0.2 ml of prechilled silicon beads (diameter, 0.3 mm) and 1 ml of Trizol and disrupted in a Mini-Beadbeater (Biospec
Products) by a 2-min burst at 2800 rpm. The aqueous phase was then
separated, and RNA was precipitated according to the instructions for
RNA isolation (Life Technologies, Inc.). Typically, 30 mg of RNA was
run on a 1.5% agarose gel overnight, transferred to a nylon membrane,
and UV cross-linked to the nylon membrane using a Stratalinker device
(Stratagene). Denatured probe was hybridized to the membrane-bound RNA,
detected with anti-digoxigenin antibody, visualized by autoradiography
according to the Genius System User's Guide (Boehringer Mannheim), and
quantitated by densitometry.
As previously reported (19), the
cds1::TRP1 strain (cds1 mutant) depends
on the expression of a plasmid-borne copy of CDS1 for
viability. Introduction of a single copy plasmid, pSDG2
(PGAL1-CDS1), into a cds1 mutant
supported cell growth in galactose-containing induction medium, but not
in a glucose-containing noninduction medium where CDP-DAG synthase
activity was undetectable in cell extracts at the point of cell arrest.
To study the cellular response at low CDP-DAG synthase levels, we
further characterized the growth phenotype of this mutant. Mutant
cds1 cells grew normally in galactose medium compared with
the wild type strain YPH102 because the growth of the former strain was
fully supported by overexpression of the plasmid-borne CDS1
gene. Cell growth rate gradually decreased as indicated by the increase
in the doubling time over the first 16 h after the cells were
shifted to a glucose medium followed by arrest of growth by 24 h
(Fig. 1A). There was a significant difference
in cell viability between 16 and 24 h of growth. About 80% of the
cells were viable after 16 h versus less than 20%
after 24 h as evidenced by colony formation on
galactose-containing plates (Fig. 1B), indicating that
arrest of cell growth was irreversible after prolonged periods below a
threshold level of CDP-DAG synthase activity.
PS and PI Synthase Activities and INO1 Expression Are Regulated by CDP-DAG Synthase Activity
CDP-DAG synthase activity in strain
YSD90A/pSDG2 grown in galactose induction medium was 10-fold as high as
the wild type level (strain YPH102). This activity was reduced after
cells were shifted to the glucose noninduction medium, reaching the
wild type level within 8 h (approximately 3 generations), and only about 10-15% of that level after 16 h of cell growth (Fig.
2A). PI synthase activity also followed the
CDP-DAG synthase activity trend starting at 40% above wild type level
and dropping first to 100% and then to 30% of the wild type level
(Fig. 2C). PI synthase activity in wild type yeast YPH102
was higher when cells were grown in galactose medium than in glucose
medium, but there was only a 1.4-fold change in the wild type strain as
compared with about a 4-5-fold change in the mutant strain. PS
synthase activity, however, showed the opposite trend, beginning at
40% of the wild type level, increasing first to the wild type level
and then to almost 3-fold as high as the wild type level in the
cds1 mutant (Fig. 2B). Carbon source did not
affect PS synthase or CDP-DAG synthase activities in the wild type
strain.
We next asked whether these biochemical changes were a result of
regulation of existing enzyme activity or an alteration in steady-state
mRNA levels. Expression of the CHO1/PSS, INO1, and PIS1 genes was quantitated by Northern blot analysis of
total cellular RNA from both the cds1 mutant (carrying
pSDG2) and the wild type strain YPH102. The results were normalized
using a TCM1 probe specific for ribosomal protein L3 (33)
previously shown to be unaffected by factors affecting the expression
of phospholipid biosynthetic enzymes (12). Transcripts of both the
target genes and the TCM1 gene in the cds1 mutant
were significantly lower at the 16-h time point than at the earlier
time points, which indicated that the cells were very sick by this
time; therefore, we chose time points from 0 to 12 h, where there
was no change in the TCM1 mRNA levels. As shown in Fig.
3, the amount of CHO1/PSS transcript
increased gradually from below the wild type level in the induction
medium to almost 3 times as high as the wild type level, paralleling
the reduction of cellular CDP-DAG synthase activity. This increase was
in agreement with the change of PS synthase activity over the course of
the experiment. The INO1 mRNA level was higher in cells
grown in galactose medium than in glucose medium in both strains,
YPH102 and YSD90A (0 h versus 4 h). However, there was
no increase in the mutant strain relative to the wild type strain until
CDP-DAG synthase activity was well below the wild type level (Fig. 3),
correlating with our previous observation that the cds1
mutant excreted inositol into the growth medium only when cellular
CDP-DAG synthase activity was low (19, 21). The PIS1
mRNA level remained the same (Fig. 3) despite the significant
reduction in PI synthase activity over the course of the experiment. We
were unable to measure CDS1 transcript levels in the
cds1 mutant extracts due to interference by pSDG2 plasmid DNA.
Effect of Human CDP-DAG Synthase Activity on CHO1/PSS and INO1 Expression
Because both CDP-DAG synthase protein and activity
were overexpressed in the induction medium, the above regulation could be due to either the amount of activity, which determines the cellular
CDP-DAG level, or the amount of CDS1 gene product. To address this point, we tested the effect of a human CDP-DAG synthase, isolated from a leukocyte cDNA library, on the transcription of the
above genes. GAL1 promoter-controlled overexpression of this gene complements the growth phenotype of the yeast cds1
mutant (26). However, CDP-DAG synthase activity is only 20-30% of the wild type yeast level, resulting in inositol excretion even in the
induction medium. This transformant stopped growing in noninduction medium after 1-2 generations of growth (data not shown), consistent with synthase activity being near the threshold level for growth. Cells
were grown in galactose medium to mid-log phase, shifted to a glucose
medium, and grown for an additional 3 h. Total RNA was isolated
from these cells and was used for quantitation of CHO1/PSS
and INO1 transcripts by Northern blot analysis and
normalized to the TCM1 transcript. Both transcripts were
higher in the cds1 mutant than in the wild type control
(Fig. 4). A shift from galactose to glucose medium did
not significantly affect levels of these transcripts, suggesting that
both CHO1/PSS and INO1 were fully overexpressed
at 20-30% of the wild type CDP-DAG synthase level. This result argues
that it is CDP-DAG synthase activity, and not synthase protein level,
that regulates CHO1/PSS and INO1 expression.
Transcription of INO2 and OPI1 as a Function of CDP-DAG Synthase Activity
CDP-DAG synthase regulation of CHO1/PSS and
INO1 expression showed surprising similarity to the response
of these two genes to inositol and choline, which is mediated by the
INO2-INO4-OPI1 gene products (9, 10). To determine the
potential involvement of these regulatory factors in the alteration in
gene expression brought about by changes in CDP-DAG synthase activity,
we examined the levels of INO2 and OPI1
transcripts in the cds1 mutant by quantitative Northern blot
analysis. The data in Fig. 5 show that transcripts of
INO2 and OPI1 were the same in both wild type
yeast and the cds1 mutant regardless of the change of
CDP-DAG synthase activity in the latter strain. Therefore, regulation
of gene expression by CDP-DAG synthase activity is not correlated with
changes in expression of regulatory genes that respond to inositol.
Regulation of CDS1 Expression by Inositol
Expression of
CDP-DAG synthase activity, like that of the other enzymes in the
phosphatidylcholine biosynthetic pathway, is coordinately regulated by
inositol and choline in the growth medium (34). However, the molecular
mechanism for the regulation of this enzymatic activity has not been
established. Because a cis-acting UASINO
sequence has also been found 5 to the CDS1 gene (19), we
examined whether CDS1 expression also changes in concert
with the expression of the INO2 regulatory gene.
Wild type strain YPH102 and the ino1 mutant MC13 were first
grown in a synthetic medium containing 60 µM inositol and
1 mM choline (+ino). Cells were shifted to inositol- and
choline-free media (ino), and growth was continued for 4 h
before harvesting. Transcripts of CDS1 and INO2
were assayed by quantitative Northern blot analysis and normalized to
the TCM1 transcript. In both wild type yeast and
ino1 mutant strains, INO2 transcript levels were low in cells growing in +ino medium as compared with
ino medium, and
the levels of the CDS1 transcript followed those of
INO2 (Fig. 6). These results indicate that
regulation of CDS1 expression by inositol and choline is
correlated with the transcript level of the cognate activator gene
INO2, which is also regulated by precursors of phospholipid
biosynthesis.
The results reported here demonstrate that phospholipid biosynthetic activity can be regulated by variations in CDP-DAG synthase activity. A shift of the cds1 mutant from induction to noninduction medium abolished the expression of the plasmid-borne CDS1 gene and caused changes over a 100-fold range in CDP-DAG synthase specific activity from 10-fold above to about 10% that of the wild type level. Although CDP-DAG synthase specific activity was reduced, total activity in the culture remained the same, which argues that the decreasing specific activity was due to dilution, not inactivation, of this enzyme. Such a change in enzymatic activity would dramatically affect the potential synthetic capability for CDP-DAG, a substrate for both PS and PI biosynthesis. In our previous studies (19), we found that down-regulation of plasmid-borne CDS1 expression in a cds1 mutant resulted in a reduction in both the initial rate of biosynthesis and the steady-state level of PI and an opposite increase with respect to PS; we also observed significant excretion of inositol. In this study we showed that the molecular basis for such alterations in phospholipid composition is a result of changes in the effective enzymatic activities of the two branch enzymes, PS synthase and PI synthase, and an apparent increase in inositol 1-phosphate synthesis.
The most novel finding in this study was that PS synthase and inositol 1-phosphate synthase activities are regulated at the transcriptional level, whereas PI synthase is regulated posttranscriptionally in response to changes in CDP-DAG synthetic capability. This is the first indication that these important phospholipid biosynthetic enzymes are regulated in response to an upstream enzymatic activity in the phospholipid biosynthetic cascade. Although regulation of expression of both CHO1/PSS and INO1 by CDP-DAG synthase activity is qualitatively similar, there are significant differences in the response of these two genes. The increase in the steady-state level of the CHO1/PSS gene transcript followed the dilution of CDP-DAG synthase activity during cell growth and division (Fig. 3). On the other hand, INO1 expression was not further repressed over wild type levels by overproduction of CDP-DAG synthase activity and was only derepressed when synthase levels were reduced below wild type levels.
Previous results suggested that the PI synthase level is not subject to coordinate regulation by inositol and choline (17) but is primarily regulated by the available concentration of its substrates, inositol and CDP-DAG (8). The results reported here demonstrate that PI synthase activity also responds to the cellular CDP-DAG synthase activity. Overall, PI synthase activity was higher in cells grown in galactose medium than in glucose medium, consistent with the recent report (18) on the effects of carbon source on the PIS1 transcript levels. However, Northern blot analysis did not explain the decrease in PI synthase activity that followed the trend in CDP-DAG synthase activity. Therefore, regulation of this activity could be at the translational level or could be due to protein modification or degradation. There are examples of modification of phospholipid biosynthetic enzymes by posttranslational events. PS synthase has been shown to be phosphorylated in vitro and in vivo by the cyclic AMP-dependent protein kinase (35). The inositol transporter is degraded in the vacuole by the vacuole proteases following incubation of yeast cells in an inositol-containing medium, and degradation is blocked in the vacuolar protease mutant pep4 strain (36). Our preliminary studies showed that interruption of the PEP4 gene in the cds1 mutant did not affect the decrease in PI synthase activity (data not shown). We are currently pursuing further studies aimed at understanding the molecular basis for the regulation of PI synthase activity in response to CDP-DAG synthase activity.
Despite the similar effect of CDP-DAG synthase activity and inositol upon coordinate regulation of gene expression, the mechanisms appear to be different. Our results clearly show that there was no detectable change in the amount of INO2 and OPI1 transcripts as a function of CDP-DAG synthase activity levels (Fig. 5), whereas the response of INO2 and CDS1 expression to inositol was dramatic (Fig. 6). The latter result is in agreement with studies from other groups demonstrating that inositol regulates gene expression via the INO2-INO4-OPI1 regulatory genes (1-3). Inositol regulation of CDS1 expression parallels the amount of INO2 transcript, as does the expression of CHO1/PSS and INO1 (14). Our results are the first report of coordinate and unique regulation by CDP-DAG synthase activity levels of several genes related to phospholipid metabolism. Clearly, INO2-INO4-OPI1 regulatory genes are not the only ones that control expression of phospholipid biosynthetic genes. INO1 transcription is also sensitive to mutations that affect the more general transcriptional apparatus, such as SWI1, SWI2, and SWI3, and these mutant strains are inositol auxotrophs (37). Mutation of the SIN3 gene (also known as CPE1/UME4/RPD1/GAM2/SDI1) also affects a variety of genes and causes constitutive derepression of INO1 transcription (38-41). Expression of INO1 is also regulated by SIN1 gene product (37).
It is not clear what senses a change in CDP-DAG synthase activity and how this signal is transmitted downstream to regulate the phospholipid biosynthetic pathway. However, CDP-DAG itself has been shown to inhibit a 45-kDa PI 4-kinase activity in a dose-dependent manner (42). The latter catalyzes the committed step in the synthesis of the phosphoinositides PIP and PIP2 in Saccharomyces cerevisiae (43). PIP2 or its hydrolysis products play an essential role in yeast proliferation (44). It is thus possible that CDP-DAG levels could play an important role in the regulation of phospholipid metabolism and cell growth via effects on the phosphoinositide pools (42). The observation that CDP-DAG synthase activity perturbs phospholipid metabolism by affecting the expression and the level of phospholipid biosynthetic enzymes is nevertheless significant. These results demonstrate the complexity in the control of phospholipid metabolism necessary to ensure the efficient response of the cell to changes in the environment.