ABSTRACT
In
Saccharomyces cerevisiae
, regulation of the phospholipid biosynthetic genes,
INO1
,
CHO1
,
CHO2
and
OPI3
, is known to occur at the level of transcript abundance. Derepression in
response to inositol deprivation requires the
INO2
and
INO4
regulatory genes. Repression in response to inositol supplementation requires
the
OPI1
regulatory gene. Here, we examined the role of the
UME6
global negative regulatory gene in expression of the phospholipid biosynthetic
genes. These studies were stimulated by the finding that the
INO1
promoter included a
UME6
cognate
cis
-acting regulatory sequence (URS1). We found that the
UME6
negative regulatory gene was involved in regulation of phospholipid biosynthetic gene expression through two distinct regulatory pathways. One pathway was the direct repression of
INO1
expression through the URS1 element. Surprisingly, the
UME6
gene was also required for derepression of
CHO1
,
CHO2
and
OPI3
gene expression. Consistent with this observation, the
UME6
gene was required for wild-type levels of expression of the
INO2
positive regulatory gene. Therefore, the
UME6
gene has both a negative and a positive role in regulating phospholipid
biosynthesis.
In yeast, transcription is carried out by RNA polymerase II in concert with a
set of general transcription factors including TFIID, TFIIA and TFIIB (
1
). Transcription is regulated through the action of gene-specific transcription factors which bind in a sequence-specific manner to regulatory regions found in promoters (
1
). While considerable progress has been made in understanding the mechanisms
controlling the activation of gene expression, transcriptional repression is
also important in the regulation of many genes (
2
). The yeast
UME6
gene product represses transcription of a diverse set of genes involved in
meiosis (
3
-
5
), heat shock response (
6
) and arginine catabolism (
7
). The
UME6
gene is also a positive regulator of some early meiotic genes in sporulating
cells (
8
,
9
). The molecular mechanism of how the
UME6
gene product functions as an activator is currently unknown. However, the
transcriptional activation function is known to be dependent on the
IME1
gene and has only been observed in sporulating cells (
8
,
9
). Here, we examined the role of the
UME6
gene in expression of the yeast phospholipid biosynthetic genes under
vegetative growth conditions.
In
Saccharomyces cerevisiae
, regulation of the genes in the phospholipid biosynthetic pathway in response
to the soluble lipid precursors inositol and choline has been shown to occur at
the level of transcription of the
INO1
(
10
),
CHO1
(
11
),
CHO2
and
OPI3
structural genes (
12
,
13
). Expression of these structural genes requires a common set of regulatory
genes and a common
cis
-acting DNA element. The positive regulatory genes include
INO2
and
INO4
, which encode basic helix-loop-helix (bHLH) proteins (
14
,
15
). The
INO2
and
INO4
gene products form a heterodimer that interacts with the UAS
INO
element and is essential for
INO1
expression (
14
,
16
,
17
,
18
). Strains bearing mutant alleles of the
INO2
or
INO4
genes are inositol auxotrophs (
19
) because they are unable to derepress
INO1
transcription (
10
). In contrast, the products of the
OPI1
and
SIN3
regulatory genes act to repress the activities of the Ino2 and Ino4 proteins (
20
-
22
). Strains bearing mutant alleles of these negative regulators display an
inositol excretion phenotype (Opi
+
phenotype) which correlates with overexpression of the
INO1
gene (
23
,
24
). In addition to the Opi
+
phenotype, these mutant strains constitutively overexpress the structural genes
in the phospholipid biosynthetic pathway, and further experimental evidence
indicates that the Opi1 and Sin3 repressors function through the only common
cis
-acting element found in these promoters, the UAS
INO
element (
22
,
25
).
The
INO1
promoter includes a URS1 sequence which represses UAS
INO
-driven expression of a
CYC1-lacZ
reporter gene (
26
). These observations prompted the present investigation of the role of the
UME6
gene in regulating transcription of the
INO1
gene. We report that the
UME6
gene was required for URS1-mediated repression of the
INO1
gene. Unexpectedly, the
UME6
gene was also required for induction of
CHO1
,
CHO2
and
OPI3
gene expression. The inability to induce expression of these genes in a
ume6
[Delta] mutant strain coincided with decreased activity of the
INO2
promoter. Therefore, these results provide the first evidence for a regulatory
role for the
UME6
gene in phospholipid biosynthesis and in controlling expression of the
regulatory gene
INO2.
The yeast strains used in this work are listed in Table
1
. Strain BRS2005 was constructed by transformation of BRS1001 with a restriction fragment
containing an
opi1
[Delta]
::LEU2
allele (
25
). Similarly, BRS2009 was constructed by transformation of BRS1001 with a
restriction fragment containing a
ume6
[Delta]
::LEU2
allele (
27
). The diploid strain BRS1005 was homozygous for an
ino1-13
allele which conferred inositol auxotrophy. All yeast cultures used in this
study were grown at 30oC in complete synthetic media containing 2% glucose (vol/vol) and either
containing 75 [mu]M inositol and 1 mM choline (I
+
C
+
) or lacking inositol and choline (I
-
C
-
) (
10
).
Excretion of inositol was determined using a plate assay in a manner described
previously (
28
). Briefly, yeast strains to be tested were patched onto plates lacking inositol
and choline that had a reduced agar concentration (1.2%). The strains to be
tested were allowed to grow for 72 h, and a suspension of the inositol
auxotroph reporter strain BRS1005 was streaked away from the patch. Growth of
the reporter strain was scored after an additional 72 h incubation at 30oC.
Yeast transformants were assayed for CAT activity as previously described (
20
). Units of CAT activity were defined as counts per minute measured in the
organic phase and expressed as a percentage of total counts per minute (percent
conversion) divided by the amount of protein assayed (in micrograms) and the
time of incubation (in hours). Total protein concentration was determined using
a Bio-Rad Protein Assay Kit (Bio-Rad, Rockville Center, NY).
RNA was isolated from yeast using a glass-bead disruption and hot phenol extraction procedure (
29
). Northern and slot blot hybridizations were performed as described previously (
10
). Results were visualized by autoradiography and gene-specific c.p.m. quantitated using the Betascope 603 Blot Analyzer (Beta-gen, Waltham, MA). RNA probes (cRNA) for the Northern and slot blot hybridizations were
synthesized using the Gemini II Core System (Promega, Madison, WI) from
plasmids linearized with a restriction enzyme and transcribed with an RNA polymerase as follows (plasmid/restriction enzyme/RNA polymerase): pAB309[Delta]/
Eco
RI/SP6 (
TCM1
), pMH203/
Eco
RI/SP6 (
OPI3
), pAS103/
Hin
dIII/T7 (
CHO1
), pTG109/
Bam
HI/T7 (
CHO2
) and pJH310/
Hin
dIII/T7 (
INO1
).
Table 1
Yeast strains were transformed using lithium acetate by a method previously
described (
30
).
Plasmids pBM-INO2, pBM-INO4 and pBM-INO1 (used in the CAT assays) contained PCR-generated promoter sequences fused upstream of the
cat
reporter gene, and have been described in detail elsewhere (
20
). Plasmid pBM-MURS contained the portion of the
INO1
promoter found in pBM-INO1 (-453 to +1) with a PCR-generated mutant URS1 element replacing the native URS1
element. The mutant URS1 element was constructed using a previously described
strategy (
31
) (Fig.
1
A). Complimentary oligonucleotides, MURS1 (5'-CT
TC
One class of regulatory mutants that affect phospholipid biosynthesis share the
The
SIN3
and
UME6
genes have been linked to URS1-mediated repression (
21
,
26
,
7
). Since a
sin3
[Delta] mutation has a pleiotropic effect on phospholipid biosynthetic gene
expression and the products of the
UME6
and
SIN3
genes often function collectively, we examined if the
UME6
gene also had a role in controlling expression of the phospholipid biosynthetic
genes. For this, total RNA was isolated from wild-type strain (BRS1001) and an isogenic
ume6
[Delta]
mutant strain (BRS2009) grown in media lacking (derepressing) or containing
inositol and choline (repressing). For comparison, we also isolated RNA from an
opi1
[Delta] mutant strain (BRS2005). Expression of the phospholipid biosynthetic
genes was quantitated by slot blot hybridization with appropriate cRNA probes,
and normalized for loading variations to expression of the constitutive
TCM1
gene (
32
).
Since a strain harbouring a
ume6
[Delta] allele had the Opi
+
phenotype (Fig.
2
), we first examined expression of the
INO1
gene because its overexpression typically correlates with the Opi
+
phenotype (
25
). Quantitation of
INO1
mRNA levels (Fig.
3
) in these strain backgrounds demonstrated the different effects the negative
regulators
OPI1
and
UME6
had on
INO1
gene expression. As has been shown previously, in the
opi1
[Delta] mutant background
INO1
was overexpressed in the presence or absence of inositol and choline (
10
,
25
) (Fig.
3
). Contrastingly, in the
ume6
[Delta] background, the
INO1
gene was modestly overexpressed in derepressing conditions but its expression
was still subject to regulation in the presence of inositol and choline (Fig.
3
). Although, the degree of inositol-mediated repression was only ~30% in the
ume6
[Delta] strain (BRS2009).
As has been reported for the
sin3
mutant strain (
21
), we observed that expression of other phospholipid biosynthetic genes (
CHO1
,
CHO2
and
OPI3
) was also altered by the
ume6
[Delta] mutation (Fig.
3
). In marked contrast to its effect on
INO1
gene expression, the
ume6
[Delta] mutation led to a significant decrease in the expression of the other
phospholipid genes to wild-type repressed levels (Fig.
3
). The
opi1
[Delta] mutation led to constitutive expression of these same genes at levels
greater than or equal to those seen in the wild-type background under derepressing conditions (Fig.
3
) which was similar to its effect on
INO1
expression (Fig.
3
).
The
ume6
[Delta] mutation eliminated derepression of
CHO1
,
CHO2
and
OPI3
gene expression (Fig.
3
). This raised the possibility that the
ume6
[Delta] mutation had altered transcription of
INO2
and
INO4
activator
genes. Previous work demonstrates that expression of the
INO2
transcriptional activator is regulated in the presence of inositol and choline
in a manner similar to that of the other phospholipid biosynthetic genes (
20
). Based on these findings, we examined expression of
INO2
in the wild-type (BRS1001) and
ume6
[Delta] mutant (BRS2009) strains under repressing and derepressing conditions. For this, we used a plasmid that contains 500 basepairs of
the sequence upstream of the AUG translation start codon of the
INO2
gene fused to a
GAL4
-
cat
fusion reporter (
20
). A single copy of this fusion was integrated into the yeast genome by
homologous recombination at the
GAL4
locus. We found that expression of the
INO2
-
cat
reporter was dramatically reduced in the
ume6
[Delta] strain (BRS2009) as compared to the isogenic wild-type strain (BRS1001) (Fig.
4
A). In the
ume6
[Delta] strain, CAT activity was reduced ~2-fold under repressing conditions, and reduced ~3-fold under derepressing conditions (Fig.
4
A). Northern blot analysis confirmed that the
UME6
gene does regulate transcription of the
INO2
gene (data not shown).
Using the same strategy, we also tested whether expression of the
INO4
positive regulatory gene was altered in the
ume6
[Delta] strain. Previous work demonstrates that
INO4
-
cat
is constitutively expressed under both repressing and derepressing conditions (
20
). We observed that
INO4
-
cat
expression was unaffected by the
ume6
[Delta] mutation (Fig.
4
B). Thus,
UME6
is required for proper regulation of
INO2
gene expression. This correlates with the observation that, of the two
transcriptional activator genes, only
INO2
expression is regulated in response to inositol and choline (
20
).
The Ume6-dependent induction of early meiotic genes requires the
IME1
gene (
8
), suggesting that the
IME1
gene might also be required for induction of
CHO1
gene expression.
CHO1
transcription was quantitated in a wild-type strain (SFY59) and an isogenic strain carrying an
ime1
[Delta] allele (REE2276). The data showed that
CHO1
transcription was unaffected by the
ime1
[Delta] mutant allele (data not shown). Similarly, transcription of the
INO2
gene was also not affected by the
ime1
[Delta] mutant allele (data not shown).
Repression of phospholipid biosynthesis in response to exogenous inositol and choline is a complex process involving at least two
cis
-acting sequences, the UAS
INO
and the URS1 element and two
trans
-acting factors encoded by the
OPI1
and
SIN3
genes (
21
,
26
,
34
,
25
). Here, we report that the
UME6
negative regulatory gene was also required for proper regulation of the genes
involved in phospholipid biosynthesis. Surprisingly, a
ume6
[Delta] mutation had disparate effects on expression of the genes involved in phospholipid biosynthesis. Among the structural genes in the phospholipid
biosynthetic pathway,
INO1
expression was most dramatically affected. As is the case with mutations in the
SIN3
(
21
) or
OPI1
(
25
) genes, a
ume6
[Delta] mutant strain overproduced inositol and excreted it into the growth
medium, indicating overexpression of the
INO1
gene (Fig.
2
). Consistent with the Opi
+
phenotype, the
INO1
gene was overexpressed in a
ume6
[Delta] mutant strain, grown under repressing conditions (presence of inositol
and choline), to levels seen in the wild-type strain under derepressing conditions (Fig.
3
).
Mutations in either of the negative regulators,
OPI1
or
SIN3
, lead to constitutive expression of the other co-regulated structural genes in the phospholipid biosynthetic pathway,
including the
CHO1
,
CHO2
and
OPI3
genes (Fig.
3
) (
21
). By contrast, a
ume6
[Delta] mutation renders these genes constitutive, but at levels identical to
those observed for a wild-type strain under repressed conditions (Fig.
3
). Thus, our results identified a novel positive regulatory role for the
UME6
gene in controllling expression of the
CHO1
,
CHO2
and
OPI3
genes.
The positive regulatory role for the
UME6
gene on
CHO1
,
CHO2
and
OPI3
expression suggested that
UME6
may have been required for proper expression of the
INO2
and
INO4
transcriptional activator genes. It is known that expression of an
INO2-cat
fusion gene is regulated in response to inositol and choline, while expression
of an
INO4-cat
fusion construct is known to be constitutive (
20
). Moreover, in an
opi1
[Delta] mutant, the
INO2-cat
fusion gene is constitutively overexpressed whereas expression of the
INO4-cat
gene is unaffected (
20
). In contrast to the
opi1
[Delta] effect,
INO2
-
cat
expression in a
ume6
[Delta] mutant strain was markedly decreased under both repressing and
derepressing conditions when compared to a wild-type strain (Fig.
4
). Thus, the
UME6
gene had a positive regulatory role in transcription from the
INO2
promoter. This decreased expression of the
INO2
activator gene in the
ume6
[Delta] strain can explain the effect of the
ume6
[Delta] mutation on expression of the
CHO1
,
CHO2
and
OPI3
genes. We have previously shown that
INO2
gene expression is required for increased expression of the
CHO1
gene when cells are grown under derepressing conditions (
35
). Thus, at low levels of
INO2
gene expression,
CHO1
gene expression becomes nearly unresponsive to inositol and choline supplementation. Therefore, we suggest that the severe
reduction in
INO2
expression in the
ume6
[Delta] mutant strain was responsible for eliminating repression of
CHO1
,
CHO2
and
OPI3
gene expression by inositol and choline.
One parsimonious explanation for the reduction in
INO2-cat
expression lies in the ability of a
ume6
[Delta] mutant strain to overproduce inositol and excrete it into the growth
media (Fig.
2
). The
ume6
[Delta] mutation may cause a significant rise in the internal pools of inositol,
resulting in repression of
INO2-cat
expression; however, we do not favour this explanation based on the following
observation. A strain harbouring a mutation in the general transcriptional
repressor
sin3
also overexpresses the
INO1
gene and excretes inositol while expressing the
CHO1
,
CHO2
and
OPI3
genes at derepressed levels (
21
), indicating that
INO2
expression is not repressed by excess inositol.
Many yeast genes in unrelated systems are known to contain a URS1 element in
their promoters, and to require this element for repression of gene expression.
In this report, we directly examined the role of the URS1 element in repression
of
INO1
gene expression. We created two fusions of the
INO1
promoter to the
cat
reporter gene, which were identical except for a mutation of the URS1 element
of one reporter construct. The reporter constructs containing the mutation in
the URS1 element gave constitutive CAT activity (Fig.
5
) regardless of strain genotype, indicating that the URS1 element is crucial for
repression of
INO1
. Curiously, the wild-type strain gave levels of CAT activity that were higher than in the
ume6
[Delta] strain. This effect may be due to the lower expression of the
INO2
activator gene in the
ume6
[Delta] strain (Fig.
4
). Since there was no synergy between the
ume6
[Delta] mutation and the mutant URS1 element, we concluded that
UME6
regulates
INO1
gene expression primarily through the URS1 element. Regulation involving the
URS1 element is quite complex and can involve several different system-specific players. In the case of the
CAR1
gene, which is involved in nitrogen metabolism, the
UME6
gene is absolutely required for URS1-mediated repression (
7
), although it is the products of the
BUF1
and
BUF2
(
RPA1
and
RPA2
) genes that actually bind to the URS1
CAR1
element (
33
). However, in the case of the meiosis-specific gene
SPO13
, experiments using an MBP-Ume6 fusion protein have demonstrated direct binding of the Ume6 fusion
protein to the URS1
SPO13
element (
27
). In addition, full repression of the
SPO13
gene also requires the product of the
SIN3
gene (
36
), which is not required for repression of
CAR1
(
7
).
Six different systems of repression using the
SIN3
and
UME6
regulatory genes and the URS1 element can now be defined. There are systems
(e.g.
SPO13
and
INO1
) that use both
SIN3
and
UME6
as repressors through a URS1-dependent pathway (
22
,
27
; and results presented here). There are systems that use either
SIN3
(e.g.
HO
) (
37
) or
UME6
(e.g.
CAR1
) (
7
) as repressors through URS1-dependent pathways. There are also systems that use
SIN3
, as a repressor, but are URS1-independent (e.g.
TRK2
and
INO1
) (
22
,
38
). In addition, the
UME6
gene product has been shown to be required as a URS1-dependent activator of meiotic genes during meiotic development (
8
,
9
). Finally, our results demonstrated that the
UME6
gene controls phospholipid biosynthetic gene expression through a mechanism
that involved stimulation of expression of the
INO2
activator gene but is URS1-independent.
We propose a model to explain the role of the
UME6
gene product on expression of the phospholipid biosynthetic genes (Fig.
6
). The model predicts that the Ume6 protein functions to directly inhibit transcription
of the
INO1
gene and that it may stimulate transcription of the
INO2
gene either directly or indirectly. The direct mechanism would require that
Ume6 function as a transcriptional activator of the
INO2
gene. This mechanism is difficult to envision since the
INO2
promoter lacks any URS1-like sequences and since Ume6 was not capable of activating transcription
in a diploid cell during vegetative growth (
8
). Therefore, this mechanism would require that Ume6 function as a URS1-independent,
IME1
-independent, haploid-specific transcriptional activator. We currently favour the indirect
mechanism which predicts that Ume6 would function to repress a negative
regulator of
INO2
transcription. This indirect mechanism would not require the presence of a URS1
element in the
INO2
promoter and would not be dependent on the
IME1
gene.
Strain
Genotype
Source
BRS1001
MATa,
ade2-1
,
his3-11,15
,
leu2-3,112
,
can1-100,
trp1-1
,
ura3-1
This lab
BRS2005
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100
,
trp1-1, ura3-1, opi1::LEU2
This study
BRS2009
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1, ume6::LEU2
This study
BRS1005
MATa,
ade1, ino1-13
This lab
MAT [alpha],
ade1, ino1-13
SFY59
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1, ade6
C. Steber and R.E. Esposito
REE2276
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1, ade6,
C. Steber and
ime1::URA3
R.E. Esposito
BPA101
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This lab
gal4::pBM-INO2::URA3
BPA102
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This lab
gal4::pBM-INO4::URA3
JCJ101
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This lab
gal4::pBM-INO1::URA3
JCJ102
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100, trp1-1, ura3-1,
This study
gal4::pBM-INO1::URA3,ume6::LEU2
JCJ103
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This study
gal4::pBM-MURS::URA3
JCJ104
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This study
gal4::pBM-MURS::URA3, ume6::LEU2
JCJ105
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This study
gal4::pBM-INO2::URA3, ume6::LEU2
JCJ106
MATa,
ade2-1, his3-11,15, leu2-3,112, can1-100,
trp1-1, ura3-1,
This study
gal4::pBM-INO4::URA3, ume6::LEU2
REFERENCES
Return