|
|
|
|||||||
|
|||||||||
(Received for publication, August 1, 1994; and in revised form, December 20, 1994) From the
The LEU4 gene of Saccharomyces cerevisiae and
the enzyme encoded by LEU4,
In the yeast Saccharomyces cerevisiae, the genes for a
particular metabolic pathway are generally unlinked and located on
different chromosomes. Each gene has its own promoter and regulatory
sequences. The promoters of these RNA polymerase II-transcribed genes
typically contain at least one TATA box that serves as focal point for
the assembly of the preinitiation complex. A minimal promoter without
upstream regulatory sequences is capable of directing basal level
transcription only. Upstream activating or repressing sequences (UASs ( The LEU4 gene and the enzyme encoded by LEU4 ( The
pleiotropic mode of action of the Leu3p Here we report that the major
regulatory elements of the LEU4 promoter are a Leu3p-binding
element (UAS
Figure 1:
General structure of plasmids
pYH1-pYH20. The vertical arrow indicates the position of wild
type and mutant LEU4 promoters. The curved arrows indicate the direction of transcription. See ``Materials and
Methods'' and Fig. 2for details of construction and
location of mutations.
Figure 2:
Deletions and point mutations of the LEU4 promoter and their effects on LEU4-lacZ expression. Large deletions (A) and point mutations or
small deletions (B) were generated and sequenced as described
under ``Materials and Methods'' (see also Table 1). The
5` end points of the large deletions are shown relative to the
beginning of the open reading frame of LEU4 (designated +1, see (7) ). L, UAS
Fig. 2A shows the
results of a serial deletion experiment. Plasmids containing either the
wild type LEU4 promoter (to position -679) or truncated
promoters were used to transform four different types of cells: those
that were wild type with respect to LEU3 and GCN4,
those that lacked LEU3, those that were deficient in GCN4, and those that were deficient in both LEU3 and GCN4. When cells were starved for leucine, a condition that
stimulates both LEU3- and GCN4-dependent
transcription, the full-length promoter (plasmid pYH1) supported a high
level of expression of the LEU4-lacZ fusion (specific
To find out
whether the Leu3p-mediated control and the general amino acid control
are the only activation mechanisms of LEU4, and to define the
relative role of the Leu3p- and Gcn4p-mediated activation more clearly,
presumptive control elements were destroyed individually and in various
combinations by creating base pair substitutions or small deletions
within the consensus sequences (Fig. 2B). The elements
chosen for mutation were the UAS
Figure 3:
Additive activation of LEU4-lacZ expression from UAS
Figure 4:
DNase I footprint analysis of the proximal
region of the LEU4 promoter. A, different
concentrations of yTBP, as indicated across the top, were incubated
with end-labeled DNA fragments of the LEU4 promoter (positions
-343 to -43). DNase I footprinting was performed as
described under ``Materials and Methods.'' The numbers on the left of each panel refer to positions of the LEU4 promoter relative to the start of the open reading frame
(+1). They were assigned with the aid of a sequencing ladder. B, the protected regions on either strand (coding strand on
top) are underlined. Potential additional protection is shown
by small capital letters.
In this study, we have identified five functional cis elements that govern the expression of the LEU4 gene. The
three distal elements are are located between positions -455 and
-353 (relative to the start of the LEU4 open reading
frame) and encompass one Leu3p-binding element (UAS The additive and independent activation of LEU4 through
UAS As is the case for many yeast
genes under general control, LEU4 has multiple sequences in
its promoter that are homologous to the 5`-TGACTC-3` consensus
sequence. However, only two out of at least four putative general
control response elements are functional in vivo. This
phenomenon seems to be a recurring theme in the general control system
of yeast. Only two out of seven TGACTC-like sequences in the noncoding
region of HIS3 are apparently functional(40) .
Similarly, while most of the five general control repeats in the
noncoding region of HIS4 are required for full derepression
under starvation conditions, one repeat stands out in its
significance(41, 42) . The one general control repeat
present in the noncoding region of ARO7 does not support
Gcn4p-mediated regulation even though it specifically binds Gcn4p in vitro(43) . It is not known why some Gcn4p-binding
elements function in vivo while others do not; it has been
suggested that additional features, e.g. chromatin structure,
may be responsible for the differences in efficiency(43) . The
two functional general control elements of LEU4 are not
entirely equivalent. Mutating the GCE-B element not only reduces the
starvation/surfeit ratio but also yields lower absolute values of LEU4 expression, suggesting that the GCE-B element is involved
in basal level regulation of LEU4 also. Again, there are
several prior examples of such dual functionality of Gcn4p-binding
sites(40, 42, 44) . The regulation of LEU4 expression by Leu3p-
Volume 270,
Number 10,
Issue of March 10, 1995 pp. 5270-5275
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-isopropylmalate synthase,
occupy a special position in amino acid metabolism.
-Isopropylmalate synthase catalyzes the first committed step in
leucine biosynthesis. However, the reaction product
-isopropylmalate is not only an intermediate in the leucine
biosynthetic pathway, but also functions as co-activator of at least
six genes, both within and outside of the leucine pathway. The
metabolic importance of -isopropylmalate appears to be reflected
in the surprisingly multifaceted regulation of LEU4 expression. This report describes an analysis of functional cis elements in the LEU4 promoter. Five such elements
were identified. Three distal elements, designated UAS
,
GCE-A, and GCE-B, are responsible for regulation by the regulatory
proteins Leu3p and Gcn4p, respectively. The incremental activation of LEU4 by these elements is additive and independent. In
addition, two proximal elements were localized. One of these conforms
to the TATA consensus sequence and exhibits high affinity for TATA
binding protein. The other element shows strong sequence identity with
the Bas2p binding site and appears to be involved in basal and
phosphate-mediated regulation of LEU4.
)or URSs), by interacting with their cognate
transcriptional regulators, can cause intricate and complex responses
to environmental
signals(1, 2, 3, 4) .-isopropylmalate synthase EC 4.1.3.12,
[-IPMS]) provide an interesting example of multiple
controls. -IPMS catalyzes the first, committed step in leucine
biosynthesis. Physiological and genetic studies have shown that LEU4 expression is regulated by the availability of leucine
and by amino acid starvation through the general control of amino acid
biosynthesis(5, 6, 7, 8, 9) .
The regulation of LEU4 by leucine is indirect and operates
through -IPM, a product of the -IPMS catalyzed reaction. When
leucine is in short supply, diminished feedback inhibition of
-IPMS causes the -IPM level to rise. -IPM subsequently
interacts with the regulatory protein Leu3p, which in turn activates LEU4 expression. Leu3p is a well-studied DNA binding protein
of the 2-Zn-6-cysteine cluster type(10, 11) . It binds
to a consensus sequence (5`-GCCGGNNCCGGC-3`, designated
UAS) that is found in the promoters of LEU1, LEU2,
LEU4, ILV2, ILV5, and GDH1(12, 13, 14, 15, 16) .
A current model postulates that Leu3p binds to the UAS elements regardless of whether -IPM is present or absent.
When -IPM is absent, Leu3p is inert or acts as a repressor of
transcription; incoming -IPM then changes Leu3p from an inactive
(repressive) to an active
configuration(16, 17, 18) .
-IPM complex and the
function of -IPM as a more general metabolic signal (13) place LEU4 and its gene product at the hub of a
regulatory network. It was therefore important to understand in greater
detail how the synthesis of -IPM is regulated. Specifically, we
wanted to identify functional cis elements of the LEU4 promoter and learn in what way and to what extent they contribute
to the expression of LEU4.) and two general control response (Gcn4p
binding) elements (GCEs). The Gcn4 protein has long been known to bind
to regulatory sequences of the 5`-TGACTC-3` type and to activate
transcription of at least 30 separate genes in response to starvation
for any one of several amino acids(19) . Activation through the
UAS and GC elements is additive and independent. In
addition, LEU4 is subject to basal level regulation that
includes a response to the phosphate concentration in the medium and is
probably mediated by the Bas2 protein. Bas2p, together with Bas1p, has
been shown to be required for basal level transcription of the HIS4 gene and to be involved in the regulation of purine and phosphate
metabolism in
yeast(20, 21, 22, 23) . Finally, the LEU4 promoter is shown to contain one functional TATA element.
Strains and Media
The S. cerevisiae strains used here were XK25-1B (MATa ura3-52),
XK53-31 (MATa ura3-52 leu1), XK147-2C (MAT ura3-52 leu1 LEU4
), XK154-2D (MATa ura3-52 leu1 LEU4gcn4-101), XK154-8A (MAT ura3-52 leu1 LEU4 leu3-
2::HIS3), and
XK154-5B (MAT ura3-52 leu1 LEU4gcn4-101 leu3-2::HIS3). The Escherichia coli strains were TG1 (K12 (lac-pro) supE hsdS/F` tra36 proA
BlacI
lacZM15) and CJ236 (dut1
ung1 thi-1 relA1/pCJ105(CM
)). YPD (1% yeast
extract, 2% Bacto-peptone, and 2% glucose) and SD medium (0.67% yeast
nitrogen base without amino acids, 2% glucose) with various additions
were used as indicated. YPD medium that has depleted inorganic
phosphate was prepared as described by Rubin(24) . The E.
coli strain TG1 was routinely used for DNA manipulations. Strain
CJ236 was used for isolation of uracil-containing single-stranded (ss)
DNA. E. coli cells were grown in L broth or 2 
YT media (25) with the addition of 100 µg/ml ampicillin where
needed.Site-directed Mutagenesis
The oligonucleotides
used in this study are shown in Table 1. M13mp11/401-6 is a
M13mp11 derivative containing a 786-bp XbaI-BamHI LEU4 promoter fragment. The BamHI site was newly
created two codons downstream from the translational start of LEU4. For ss DNA production, CJ236 cells were infected with
M13mp11/401-6 phage stock and shaken vigorously at 37 °C in
the presence of 0.25 µg/ml uridine and 30 µg/ml
chloramphenicol. Uracil-containing ss DNA was isolated and
site-directed mutagenesis of the LEU4 promoter was performed
according to the procedure supplied by Bio-Rad. Mutations were
confirmed by sequencing following a United States Biochemical Corp.
(Cleveland, OH) protocol.
LEU4`-`lacZ Fusion Plasmid Construction
Plasmid
pSEYC102 was a gift from S. Emr, University of California, San Diego.
For deletion constructs the M13mp11/401-6 phage DNA containing
the LEU4 promoter (see above) was cut with restriction enzymes PvuII, StuI, RsaI, TaqI, or BsaAI, blunt ended with the Klenow fragment of DNA polymerase
when necessary, then cut with BamHI. The isolated fragments
were ligated to SmaI-BamHI-digested pSEYC102, giving
rise to pYH1 (contains a 690-bp PvuII-BamHI promoter
fragment), pYH2 (393-bp StuI-BamHI fragment), pYH3
(342-bp RsaI-BamHI fragment), pYH4 (326-bp PvuII-BamHI fragment), pYH5 (220-bp TaqI-BamHI fragment), and pYH6 (119-bp BsaAI-BamHI fragment). An additional series of fusion
plasmids was constructed by performing site-directed mutagenesis on
M13mp11/401-6, followed by ligation of the isolated mutant PvuII-BamHI fragments to SmaI-BamHI-digested pSEYC102 (plasmids pYH7-pYH15 and
pYH17-pYH20). Plasmid pYH16 was constructed by excising a StuI-BamHI fragment containing a mutated GCE-B box
from M13mp11/401-6 and ligating it to SmaI-BamHI-digested pSEYC102. In all of the fusion
plasmids, the first two amino acid codons of LEU4 (Met-Val)
were attached to the 10th codon of lacZ (Val) via a bridge of
three codons (Arg-Asp-Pro) introduced by the newly constructed BamHI site.Yeast Transformation, Cell Growth, and
The LEU4-lacZ fusion
plasmids were used to transform strains XK147-2C, XK154-2D,
XK154-8A, XK154-5B, XK53-31, or XK25-1B using a
modified lithium acetate method(26) . Transformants were
purified once, and single colonies were precultured in branched-chain
amino acid surfeit medium (SD plus 2 mM leucine, 1 mM isoleucine, and 1 mM valine). Precultures were divided
and used to inoculate both a surfeit and an amino acid starvation
medium, the latter consisting of SD plus 0.2 mM leucine. The
calculated starting OD
-Galactosidase Assays
was 0.05. Cells were incubated at
30 °C and harvested at an OD of 0.9 (surfeit medium
mid-log phase) or 0.4-0.5 (starvation medium late log phase).
-Galactosidase activity was measured as described previously with
minor modifications(10) . Briefly, harvested cells were washed
once with double-distilled H
O, once with Z buffer (100
mM sodium phosphate buffer, pH 7.0, 10 mM KCl, 1
mM MgSO
, 25 mM -mercaptoethanol) and
resuspended in Z buffer. Twenty µl of 0.1% SDS and 50 µl of
chloroform were added to 1-ml aliquots of cell suspensions, and the
mixture was vortexed for 30 s. After a 15-min incubation at 30 °C,
0.2 ml of an o-nitrophenyl--D-galactopyranoside
solution (4 mg/ml in Z buffer) were added, and the incubation was
continued until visible color developed. The reaction was stopped by
adding 0.5 ml of 1 M NaCO
. Cell debris
was removed and the OD
measured immediately. Specific
activities were determined according to Miller(27) . To assay
cells grown in low phosphate medium, plasmid-bearing strains were grown
overnight in 10 ml of SD medium. Cells were washed with sterile
distilled HO, transferred to low phosphate YPD, and grown
at 30 °C until an OD of about 1.8. To minimize the
difference in residual phosphate concentration in the low phosphate YPD
medium, the same batch of medium was used for all experiments. The
harvested cells were assayed for -galactosidase activity as
described above.DNase I Footprinting
A partially purified
preparation of yeast TATA-binding protein (yTBP) was a gift from Karen
Arndt, Harvard University. DNase I was obtained from Boehringer
Mannheim. The probe used for footprinting was a LEU4 promoter
fragment, prepared by polymerase chain reaction, extending from
position -343 to position -43. It was end-labeled with
[
P]ATP and purified using a Microcon spin
column (Amicon, Inc.). The probe was cut with restriction enzymes HaeIII and DdeI, respectively, to obtain two
individual fragments labeled at one end only. Fifteen ng of DNA were
incubated with or without TBP in 25 mM Tris-HCl buffer, pH
8.0, containing 6.25 mM MgCl, 0.5 mM EDTA, 10% glycerol, 0.5 mM dithiothreitol, 200 ng of
poly(dGdC), and 0.01% Nonidet P-40 in a total volume of 100
µl for 30 min at 23 °C. Then 0.25 units of DNase I were added.
After 1 min at 23 °C, the reaction was stopped by adding 600 µl
of precooled stopping solution (0.05 M potassium acetate plus
5 µg of tRNA dissolved in 588 µl of ethanol). The DNA that
precipitated during 2 h at -70 °C was pelleted by
centrifugation at 4 °C for 20 min. The pellet was washed once with
ice-cold 90% ethanol and vacuum-dried. Three µl of loading solution
(95% formamide, 20 mM EDTA, 0.05% bromphenol blue, 0.05%
xylene cyanol (FF)) were added, and the sample was heated at 90 °C
for 5 min and chilled on ice. Samples were loaded on a 6% sequencing
gel to resolve the DNA fragments. The gel was fixed, dried, and
autoradiographed.
Mutational Analysis of the Leu4 Promoter Identifies
Three Upstream Regulatory Elements
The promoter analysis was
performed using plasmids in which normal and mutated LEU4 promoter sequences were fused to the E. coli lacZ gene (Fig. 1). The plasmid was of the ARS1-CEN4 type,
assuring a consistently low copy number of approximately one copy/cell.
The LEU4-lacZ fusion was constructed such that it retained no
more than two 5` codons of the LEU4 open reading frame. This
strategy eliminated the mitochondrial import signal of the native Leu4
protein(8) , thus avoiding complications that might arise from
sequestering the Leu4--galactosidase fusion protein in
mitochondrial compartments.
(dyad symmetrical center at -445); A, GCE-A
(centered on position -419); B, GCE-B (centered on
position -356); C, GCE-C (centered on position
-319; D, GCE-D (centered on position -99). The
-galactosidase activities were measured in plasmid-bearing strains
XK147-2C, XK154-8A, XK154-2D, and XK154-5B (the
pertinent genotypes are shown in parentheses) grown either in
branched-chain amino acid surfeit medium (SD plus 2 mM leucine
and 1 mM each of isoleucine and valine, SURF) or in starvation medium (SD plus 0.2 mM leucine, STARV). The numbers represent averages
of at least two independent trials with errors <20%. Filled and open circles, wild type and mutant UAS; filled and open rectangles, wild type and mutant GCE
sequences.
-galactosidase activity of 199) when LEU3 and GCN4 were both intact. When either LEU3 or GCN4 were
dysfunctional, the level of expression dropped to about 40-50% of
the high level. The level of expression dropped to very low values when
both genes were dysfunctional. These results suggested that
Leu3p-mediated control and general control of amino acid biosynthesis
are the major mechanisms by which LEU4 is regulated and that
under starvation conditions each control contributes about equally to
the final level of expression. The first deletion, extending to
position -382 and represented by plasmid pYH2, caused the LEU4 promoter to lose its response to Leu3p; the general
control response was retained, with a starvation/surfeit ratio of
greater than 3. The next two deletions, extending to positions
-332 and -315, respectively, (plasmids pYH3 and pYH4)
caused the loss of both Leu3p-mediated and general control and resulted
in a low level of expression (``basal level I,'' 17-20
units of activity) with a starvation/surfeit ratio of close to 1.
Either deletion apparently removed elements responsible for both major
controls. A deletion extending to -208 (plasmid pYH5) lowered the LEU4-lacZ expression to ``basal level II''
(8-9 units of activity), again with a starvation/surfeit ratio of
about 1. Finally, a deletion extending to -109 (plasmid pYH6)
resulted in near zero expression of the LEU4-lacZ fusion gene.
These latter results suggested the presence of two separate elements
controlling the basal expression of LEU4. element and four
segments that conform to the conserved 6 bp core sequence
(5`-TGACTC-3`) of the Gcn4p recognition
element(19, 28) . These four segments were designated
GCE-A, B, C, and D. Incapacitating the presumed UAS sequence by creating a 9 bp deletion (-450 to -442;
plasmid pYH7) had the same effect as deleting the LEU3 gene
(compare pYH7 with pYH1 in a leu3-2 GCN4 background), indicating that the
sequence around position -445 is indeed the only functional
UAS of the LEU4 promoter. To avoid interference
from Leu3p regulation, the analysis of the relative importance of the
presumptive GCE boxes was performed in a UAS-negative
background. Disabling GCE-A (plasmid pYH8) resulted in a reduction of
the starvation/surfeit ratio from 4.7, seen with plasmid pYH7, to 2.7.
A similar starvation/surfeit ratio (2.9) was obtained upon inactivation
of GCE-B (plasmid pYH9), although in this case lower absolute levels of
-galactosidase were observed. Mutating GCE boxes C and D, either
separately or together, did not significantly affect the expression of LEU4 (compare plasmids pYH10, 11, and 15 with pYH7). Mutating
both GCE-A and -B, on the other hand, led to low expression (pYH12)
that was indistinguishable from the level of expression seen with
plasmid pYH7 in a gcn4
background.
Additional permutations confirmed the above results. Thus, simultaneous
mutation of GCE-A and -C or of GCE-A and -D had the same effect as
mutating GCE-A alone (compare plasmids pYH13 and pYH14 with pYH8), and
mutation of GCE-B superimposed upon the -382 deletion (plasmid
pYH16) yielded values similar to those of plasmid pYH12. We conclude
that, of the four Gcn4p recognition elements, only GCE-A and -B are
functional. Destruction of these cis elements by site-directed
mutagenesis has the same effect as genetic elimination of the trans-acting factor Gcn4.The UAS
To pursue the question of whether the LEU3- and GCN4-dependent controls operate additively
or in some other fashion, we constructed an additional plasmid (pYH17)
that was deficient in GCE-A and GCE-B. It was used, together with
plasmids pYH1 (wild type LEU4 promoter), pYH7 (promoter
lacking the UAS and GC Elements Activate LEU4
Expression Additively element), and pYH12 (promoter deficient
in UAS, GCE-A, and GCE-B), to transform a strain
(XK53-31) that was wild type with respect to LEU3 and GCN4 and also produced a normal, feedback-sensitive -IPM
synthase, thus allowing a normal response to leucine starvation and
surfeit conditions. A comparison of the specific activities of
-galactosidase in the transformed cells grown under conditions of
leucine starvation and surfeit is shown in Fig. 3. It is evident
that general control and Leu3p-mediated control of LEU4-lacZ are additive. Under the conditions of the experiment, there was a
3.1-fold activation from the UAS alone (plasmid pYH17).
In the absence of UAS, but the presence of the two
functional GCE elements, a 5.3-fold activation was observed (plasmid
pYH7). When all elements were present, there was a 6.7-fold activation
(plasmid pYH1). The two GCE elements also appear to act additively (Fig. 2B). The presence of both elements (plasmids
pYH7, pYH10, pYH11, pYH15) resulted in a 4.4-fold activation, on the
average. When the GCE-A element was mutated, there was an average
activation of 2.6-fold (plasmids pYH8, pYH13, pYH14); when the GCE-B
element was mutated, a 2.9-fold activation was observed (plasmid pYH9).
and GCE sequences. Plasmids
pYH1, pYH7, pYH17, and pYH12 were introduced into strain XK53-31
and -galactosidase activities were measured as described under
``Materials and Methods'' in cells grown either under
starvation (solid blocks) or surfeit conditions (hatched
blocks). See legend to Fig. 2for definition of starvation
and surfeit. The numbers above the solid blocks indicate the -fold increase of -galactosidase activity under
starvation conditions. UAS
,
intact UAS element; UAS, mutated UAS element; GCE, all GCE sequences are
intact; GCE, GCE-A and -B sequences are
mutated (see Table 1), GCE-C and -D sequences are
intact.
The Proximal LEU4 Promoter Region Contains One Functional
TATA Element and a Bas2 Response Element
The two segments of the LEU4 promoter that, when deleted, led to basal level II or
near zero expression (Fig. 2A, plasmids pYH5 and pYH6)
were located in the downstream region of the promoter where one might
expect to find TATA elements. This portion of the promoter was
therefore subjected to DNase I footprinting and mutational analysis.
The DNase I protection assay was performed with a purified preparation
of yeast TATA box binding protein (yTBP). The footprint (Fig. 4)
showed protection between positions -250 and -271 on the
coding strand and between positions -250 and -270 on the
noncoding strand. This region contains a sequence, 5`-TATATA-3`, that
perfectly matches a TATA consensus sequence. The apparent K
for binding of yTBP, defined as the
concentration of yTBP required for half-maximal interaction, was
approximately 5 nM. This value is well within the range of
specific binding established earlier for yeast TFIID(29) . In
addition to the TATA box region, an element that is 67% identical with
the Bas2 response element (BRE) of the HIS4 promoter (20) was identified between positions -135 and -152
of the LEU4 promoter (5`-GAAAAATAACCAATAAAT-3`; noncoding
strand). To explore the functional significance of both the TATA box
and the potential BRE, we generated extensive mutations in both
regions. An appropriate strain (XK25-1B) was then transformed
with plasmids carrying either wild type or mutant element sequences in
the promoter of a LEU4-lacZ fusion. The specific
-galactosidase activity in the transformed cells (Table 2)
demonstrated that the TATA element is essential for the expression of
the LEU4-lacZ gene since destruction of the TATA box caused an
almost total loss of promoter activity. Mutation of the BRE caused a
reduction of promoter efficiency by about 40%. Simultaneous destruction
of TATA and BRE had the same effect as mutation of TATA alone. To
examine the significance of the -152 to -132 region for
basal control of LEU4, we measured the LEU4-lacZ expression from wild type and mutated promoters as a function of
the phosphate concentration in the medium. The results indicated the
following (data not shown): (i) LEU4-lacZ expression was
affected by the phosphate concentration. Under the conditions of the
experiment, increasing the inorganic phosphate concentration from a
low, depleted level to 5 mM caused a decrease of about 15%
when the wild type promoter was present (plasmid pYH1), and a decrease
of about 45% when a truncated promoter lacking UAS, GCE-A
and -B, and the TATA box (plasmid pYH5) was used. (ii) The phosphate
response was lost when LEU4-lacZ expression was directed by a
promoter whose BRE was eliminated either by truncation (plasmid pYH6)
or by site-specific mutagenesis (plasmid pYH19). (iii) Under phosphate
sufficiency, the LEU4-lacZ expression level obtained with wild
type promoter (pYH1) dropped to about half when the BRE region was
mutated (plasmid pYH19). We conclude that LEU4 expression is
subject to both phosphate and basal level controls. The fact that both
controls were absent whenever a Bas2 response element was missing
strongly suggests that both are brought about by the Bas2 protein.
) and
two Gcn4-binding elements (GCE-A and -B). These three elements are
responsible for the two major controls of LEU4 that had
previously been postulated to occur on the basis of physiological
studies and genetic manipulation of trans-acting
factors(5, 9) . The two proximal elements are centered
on position -260 and on position -143, respectively. The first of
these clearly shows the properties of a functional TATA box: it is
protected by and strongly interacts with a TATA-binding protein, and
its destruction causes a drastic decrease of LEU4 expression.
This TATA box is unusually far removed from the first major
transcription start site (about 195 versus 40-120 bp for
most yeast promoters). The element centered on position -143
possesses the features of a Bas2p-binding site(20) . Expression
from this site apparently proceeds without an additional TATA element.
This is not uncommon; TATA-independent transcription from a Bas2p site
was previously demonstrated in the HIS4 promoter(30) . , GCE-A, and GCE-B is unusual since in many other
eukaryotic promoters upstream elements activate transcription
synergistically(31, 32, 33) . Two major
models, the cooperative DNA binding model and the simultaneous contact
model, have been proposed to explain synergistic
activation(34, 35) . In the simultaneous contact
model, multiple contacts with the transcription apparatus would have a
multiplicative effect. We would like to propose a variant of the
simultaneous contact model to explain the additive effect observed with
the LEU4 promoter. We envision that Leu3p and Gcn4p might each
facilitate the assembly of a different subcomplex prior to formation of
the final preinitiation complex. This idea is in agreement with recent
findings suggesting that the preinitiation complex might not be
assembled by the stepwise addition of individual components but that
some of the components might exist in
subcomplexes(36, 37) . The in vivo partners
of Leu3p and Gcn4p are not yet known. However, in vitro experiments have shown that Gcn4p can interact directly with RNA
polymerase II (38) and that Leu3p can interact with
TBP(39) . These results are consistent with the notion that the
two regulatory proteins interact with the transcription apparatus at
different stages of assembly. Also of interest in this context is the
relative arrangement of the Leu3 and Gcn4 proteins along the promoter
DNA. If we assume that the UAS of LEU4, like
that of LEU2(11) , consists of two contact triplets 9
bp apart (center-to-center), then the downstream contact triplet
(centered on position -441) would be separated by approximately
two helical turns from the symmetrical center of GCE-A (assuming B form
DNA); GCE-A would be separated by about six helical turns from GCE-B.
This arrangement would allow the Leu3 and Gal4 proteins to bind to the
same face of the DNA double helix and might thus make simultaneous
contacts between the activators and components of the preinitiation
complex thermodynamically favorable.-IPM, general amino acid
control, and the phosphate level is complemented and augmented by the
regulation of the activity of -IPM synthase. Besides being subject
to feedback inhibition by leucine, -IPM synthase is reversibly
inactivated by coenzyme A, an effect that appears to be correlated with
the control of acetyl-CoA utilization and that can be reversed or
prevented by a high energy charge(45) . We believe that at
least part of the physiological reason for this tight and diversified
control of the first step in leucine biosynthesis is the fact that
-IPM also acts as a co-activator of genes that are controlled by
Leu3p. The recent discovery that Leu3p--IPM is an important
regulator of the main ammonia-assimilating enzyme in yeast (the GDH1-encoded NADP-dependent glutamate
dehydrogenase) strongly suggests that the yeast cell utilizes -IPM
as a signal molecule to feed information from the periphery back to the
center of nitrogen anabolism(13) . Just how far the
-IPM-controlled network extends remains to be established.
)-IPMS, -isopropylmalate
synthase; GCE, Gcn4p-binding elements; ss, single-stranded; bp, base
pair(s); yTBP, yeast TATA-binding protein; BRE, Bas2 response element.
We thank Karen Arndt, Harvard University, for a gift
of yeast TBP and Scott Emr, University of California, San Diego, for
plasmid pSEYC102.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. A. Davis and M. Ares Jr. Accumulation of unstable promoter-associated transcripts upon loss of the nuclear exosome subunit Rrp6p in Saccharomyces cerevisiae PNAS, February 28, 2006; 103(9): 3262 - 3267. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. B. Kohlhaw Leucine Biosynthesis in Fungi: Entering Metabolism through the Back Door Microbiol. Mol. Biol. Rev., March 1, 2003; 67(1): 1 - 15. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wang, F. Zheng, S. Holmberg, and G. B. Kohlhaw Yeast Transcriptional Regulator Leu3p. SELF-MASKING, SPECIFICITY OF MASKING, AND EVIDENCE FOR REGULATION BY THE INTRACELLULAR LEVEL OF Leu3p J. Biol. Chem., July 2, 1999; 274(27): 19017 - 19024. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |