The sequence and potential regulatory elements of the HEM2 promoter of saccharomyces cerevisiae

Material Information

The sequence and potential regulatory elements of the HEM2 promoter of saccharomyces cerevisiae
Schlaepfer, Isabel R
Publication Date:
Physical Description:
vii, 67 leaves : illustrations ; 28 cm


Subjects / Keywords:
Saccharomyces cerevisiae ( lcsh )
Yeast fungi ( lcsh )
Saccharomyces cerevisiae ( fast )
Yeast fungi ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 58-67).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Biology.
General Note:
Department of Integrative Biology
Statement of Responsibility:
by Isabel R. Schlaepfer.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
30696482 ( OCLC )
LD1190.L45 1993m .S35 ( lcc )

Full Text
Isabel R. Schlaepfer
B.A., Universidad de Navarra, Pamplona, Spain,1991
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Isabel R. Schlaepfer
has been approved for the
Department of
Gerald Auderisk
iJxj/ 9 -3

Schlaepfer, Isabel R. (M.A., Biology)
The Sequence and Potential Regulatory Elements of the HEM2 Promoter
of Saccharomyces cerevisiae
Thesis directed by Professor James R. Mattoon.
The yeast HEM2 gene encodes the second enzyme of the heme
biosynthesis pathway, delta-aminolevulinate(ALA)dehydratase. As part
of the effort to characterize the physiological and genetic regulation of
the ALA dehydratase, the nucleotide sequence of a 1890-bp DNA
fragment located upstream of the coding sequence and containing the
HEM2 promoter has been determined. The following potential
regulatory protein-binding motifs were found: ABFl-binding site, 2
REB1-binding sites, a putative cyclic AMP-responsive element
(CRE), yAPl-binding site, RAPl-binding site, and several HAP2/3/4-
binding sites. A reporter gene construction for the HEM2 gene
promoter was also made using the repressible yeast acid phosphatase
gene (PH05). This construct appears to be incompatible with E. coli,
so alternative strategies for inserting it into yeast must be considered.
This abstract accurately represents the content of the candidate's
I recommend its publication.
James R. Mattoon

1. Introduction 1
1.1. Regulatory elements of yeast promoters 1
1.2. Structure of tram-acting factors 3
1.3. Transcription factors found in yeast 5
1.4. Homology between mammalian and yeast regulatory genes 10
1.5 Objective of this work 19
2. Material and Methods 25
2.1. Plasmid constructions and DNA sequencing 26
2.1.1. pHEM22 (4.7 Kb) 26
2.1.2. Nested deletions from the 3' recessed end of the promoter using
exonuclease III 26
2.1.3. Sequencing 28
2.2. Construction of a transcriptional reporter system for the HEM2
gene 31
2.2.1. pNZ8P plasmid construction 31
2.2.2. Construction of the plasmid pRBl 32
2.2.3. Construction of plasmids pIRl and pIR2 32
3. Results 39
3.1. Sequencing experiments 39
3.1.1. Identification of the nested deletions from pHEM22 3 9

3.1.2. Promoter sequence of the HEM2 gene 41
3.1.3 .Analysis of the sequence and putative regulatoiy sites 44
3.2. Transcriptional reporter system for the yeast HEM2 gene 50
4. Discussion 52
References 58

1. Heme biosynthesis pathway 23
2. pHEM2 plasmid 33
3. pHEM21 plasmid 34
4. pGem7Zf(+) phagemid 35
5. Nested deletion strategies for the HEM2 promoter 36
6. pRBl construction 37
7. pIRl and pIR2 plasmids 38
8. Sequence of the 1890-bp region upstream of the HEM2 gene 42
9. Map of motifs in the HEM2 promoter 43

1. Transcription factors found in yeast and mammalian cells and their
target sequences 24
vi 1

Regulators of transcription have been found in many eukaryotic
systems. These proteins were first identified as factors contained in cell
extracts that would bind to regulatory sites of a promoter of a particular
gene (Dorn et al., 1987). Genetically, these regulatory sites can be
defined as regions of DNA that, when deleted, modified or substituted,
alter the expression from that promoter. These regulatory sites are
classified as positive or negative depending upon the effect they produce
on transcription; an increase or a decrease respectively.
The positive regulatory sites contain binding sequences for
transcriptional activators which are termed UASs,(Upstream Activating
Sequences) in yeast and enhancers in higher cells. Proteins that bind to
UASs or enhancers are themselves regulated by physiological signals in
the cell so that they convey these signals of the status of the cell to their
target genes.
1.1 Regulatory elements of yeast promoters
Four basic cri-acting elements have been found in yeast promoters:
UASs, operators or silencers (O) which mediate negative control, TATA
boxes and initiators (I) from which transcription starts and which
coincide with the DNA sequence encoding the mRNA cap site. Gene-
specific activators bound to UASs and the basal transcriptional

machinery bound at the TATA box, cooperate to bring about the
initiation of transcription by RNA polymerase II.
Virtually every yeast gene examined contains at least one UAS
between 1400 and 100 nucleotides upstream of the TATA box, and
deletion of this UAS greatly diminishes transcription. This element has
been compared to the mammalian enhancer because it can be inverted
and moved relative to the TATA box without its activity being affected
(Olesen, Hahn and Guarente, 1987). However, yeast UASs differ from
enhancers; they can not be placed downstream. Several models have
been put foward to explain how UASs can bridge distances to activate
transcription. The most commonly accepted model proposes that
activators bound at UASs and factors bound at the TATA box, and
possibly the I (initiator) site, touch each other by looping out intervening
DNA. This hypothesis is based on several experiments with prokaiyotic
systems that prove that such looping can occur, and suggests that
positive control involves protein-protein interactions (Guarente, 1987).
The study of negative regulatory sites in yeast has not been as
extensive as the study of UASs. However, in the last few years, the
understanding of the nature of these sites has increased. Some repressors
with DNA-binding activity have been identified, but they are not well
characterized. They appear to act through at least two distinct

mechanisms; by competing with positive trans-acting factors for binding
to the DNA target sequence or by interfering with activation by other
promoter elements. The latter has been shown to occur in the galactose
kinase (GALJ) gene promoter, which contains an upstream repression
sequence (URS) located between the UAS sites and the TATA box, and
is involved in glucose repression of galactose metabolism in yeast
(Flick and Johnston, 1990). Other known repressors bind to positive
trans-acting factors and inhibit their transcriptional activation
properties. This is the case for the GAL4 and GAL80 yeast proteins;
GAL4 is a positive regulatory protein that acts on the genes required for
galactose metabolism. When glucose is present in the medium, GAL4 is
complexed with the GAL80 repressor and cannot activate transcription;
however, when glucose is depleted and galactose is added to the
medium the repressor GAL80 dissociates from GAL4 and
transcriptional activation of the galactose genes occurs. This mechanism
emphasizes the importance of protein-protein interaction and shows that
negative-acting proteins do not have to be DNA-binding proteins.
1.2. Structure of fraws-acting factors
The /ram-acting factors that bind these positive and negative
regulatory sites in the DNA (cis-acting elements) are generally, but not
necessarily, made up of two independent domains:(i) a DNA-binding
domain with a very high affinity for its associated DNA binding site and

(ii) a transcriptional activating region (Verdier, 1990). Generally, these
functions are easily separable and allow the contraction of chimeras
with the DNA-binding domain of one protein and the activating region
of another. For example, the DNA-binding domain of GCN4 (activator
of amino acid biosynthetic genes) has been replaced by the homologous
c-jun proto-oncogene DNA-binding region to create a chimera that
activates transcription in yeast (Strahl,1988).
Site-specific DNA binding of transcription factors has been shown to
occur through various common DNA-binding motifs such as the helix-
turn-helix, zinc finger, leucine zipper and helix-loop-helix structures. In
some cases the transcription factor does not show common specific
binding regions, as is the case with the HAP2/HAP3 (Heme Activator
Protein) factor complex where it is not clear whether the binding to
DNA is realized by HAP2, HAP3 or both (Verdier, 1990).
Less is known about the structure of the transcriptional activating
regions. It is known, however, that they are often enriched in acidic
amino acids, glutamines and/ or prolines. Most transcriptional activators
posseses both DNA- binding and transcriptional activating functions in a
single polypeptide, but complexes of transcription factors including two
or three different protein subunits have been recognized and
characterized (Guarente, 1987).

1.3. Transcription factors found in yeast
The first tram-acting yeast regulatory loci identified encode proteins
that bind to DNA sequences upstream of the genes that they regulate.
These loci were identified by regulatory mutations that prevented the
proper expression of those particular genes. The first such locus
identified was GAL4 which is a gene required for the regulation of the
galactose-metabolizing enzymes.(Guarente, 1987).
As mentioned before, expression of the GAL genes is induced by
galactose and severely repressed in the presence of glucose (catabolite
repression). These genes provide a model system to study the
interaction of positive and negative regulatory pathways. GAL-gene
induction is controlled directly by two regulatory genes, GAL4 and
GAL80 When the yeast cells are growing in the absence of galactose,
GAL4 protein binds to the UASs of each of the GAL genes, encoding
catabolic enzymes, but is prevented from activating transcription by
interaction with another regulatory protein called GAL80. The
addition of galactose to the medium, in the absence of glucose, relieves
the GAL80-mediated inhibition of GAL4, and this results in a 1000-fold
induction of GAL gene expression (Flick and Johnston, 1992).
The GAL4 transcription factor binds specifically to a 17-bp sequence
of the GAL UAS which is partially dyad-symmetrical, suggesting that
GAL4 binds as a dimer. Extensive studies of this protein have shown

that it contains a cysteine-zinc finger domain that binds to the DNA as
well as two short acidic regions which constitute a transactivating
domain. Another well-characterized locus is GCN4, which is essential
for derepression of amino acid biosynthetic genes responding to general
control when cells are starved for amino acids. GCN4 is one of the best
characterized transcription factors in eukaryotes. The factor contains a
leucine zipper motif, binds to DNA as a homodimer, and its
transcriptional activation domain consists of a short, highly acidic
region. (Georgakopoulos and Thireos, 1992). GCN4 gene-expression is
translationally regulated by general control. The proper translational
activation of the GCN4 mRNA, by amino acid starvation, is required for
the factor to mediate transcriptional regulation to its full extent. The
consensus DNA binding sequence of GCN4 is a perfect palindrome,
ATGACTCAT, named general control nitrogen 4 recognition element
or (GCRE) (Verdier, 1990; Oliviero, 1992).
As mentioned before, most of the transcription factors contain the
DNA-binding domain and the transactivating domain in the same
polypeptide. However, a few years ago, HAP2 and HAPS factors were
genetically defined (Olesen et al., 1987) and later, HAP4 (Forsburg and
Guarente,1989). The identification of these HAP (Heme Activator
Protein) genes was made during the study of the heme and carbon
source regulation of the C.YC.l gene, which encodes for the iso-1-

cytochrome c apoprotein. Truncated promoter-CTC 1-lacZ fusions were
constructed and two tandem Upstream Activation Sites were found;
U AS land UAS2 which governed the expression of (3-galactosidase in
yeast carrying the fusions.
UAS1 is activated by the HAP1 gene-product, which binds to the site
in a heme- dependent manner in vitro. (Keng and Guarente, 1987).
Regulation through the UAS2 site requires HAP2 HAP3 and HAP4
gene-products. The HAP2/3/4 system appears to regulate respiratory
genes globally, since mutations in any of the three genes renders cells
unable to grow on nonfermentable carbon sources.
Using size variants of these HAP genes, it has been demonstrated that
all three are present in the regulatory complex, but the presence of
additional components has not yet been excluded. The requirement of
the HAP4 protein for binding of the complex to the DNA suggests that
HAP4 stabilizes the HAP2-HAP3 dimer or alters its conformation
(Olesen et al., 1987; Forsburg and Guarente, 1989). HAP2 and HAP3
proteins contain a DNA-binding domain, but lack similarity to other
proteins that bind DNA The HAP4 factor contains 553 amino acids,
and its carboxy-terminal region is extremely acidic, suggesting that it
functions in the complex as a transcriptional activation domain. These
three proteins have been shown to bind to the DNA as a heterotrimeric
complex The complex binds to the sequence TNRTTGGT (coding
strand) located in the wild type UAS2 element of S. cerevisiae CYC1

gene (Olesen, Hahn and Guarente,1987). A mutation in any of these
subunits abolishes transactivation from the UAS2 site.
The HAP2/3/4 complex also activates transcription through CCAAT
sequences (TNRTTGGT read from the complementary strand) located
upstream of the HEM1 gene that encodes for 8- aminolevulinate
synthase (ALA synthase). This enzyme catalyzes the first commiting
reaction in the heme biosynthetic pathway, Figure 1. (Keng and
Guarente, 1987).
Other genes involved in the mitochondrial electron transport chain are
also regulated by the HAP2/3/4 complex. This complex acts positively
on transcription when a non-fermentable carbon source replaces glucose
in the medium. S. cerevisiae ferments glucose preferentially even in the
presence of oxygen, and many of the respiratory genes are repressed 10-
to 20- fold when growing in glucose- containing media (Mattoon,
Caravajal and Guthrie, 1990; for a review see Gancedo, 1992). In yeast,
the regulation of cytochrome formation by oxygen and carbon source
has been extensively studied. Heme itself is the effector that mediates
induction by oxygen. When oxygen is absent, enzymes late in the heme
biosynthesis pathway (oxygenases) do not function and heme is not
made Figure 1.

The transcriptional activator HAP 1 seems to have a more general
function as a coordinator of the activity of genes encoding proteins
involved in various aspects of oxygen metabolism. Recently, it has been
shown that the expression of CTT1, encoding for catalase T, and HMG1,
encoding HMG-CoA reductase, is also mediated by the HAP1
transcriptional regulator ( Lodi and Guiard, 1991). The HAP1 gene
encodes a protein of 1,483 amino acids which can be divided into three
functional domains: a zinc finger domain involved in DNA binding to
specific sequences, a carboxy-terminal acidic region necessary for
transcriptional activation and a middle region involved in heme
induction. Deletion of this internal region converted the protein into a
heme-independent, constitutive activator, leading to the conclusion that
this region masks the DNA-binding domain in the absence of heme
(Pfeifer et al., 1987).
Mutational and footprinting analysis of the CYC1, CYC7 (iso-2-
cytochrome c), CTT1 and CYB2 (cytochrome b2) upstream activation
sites, has allowed the derivation of a consensus sequence, but some
reservation remains because the analyses were carried out with crude
extracts and other proteins different from HAP1 could bind to this
consensus site (Zitomer and Lowry, 1992);
A/T A/T - A/T C C G A/T A/T A/T A/T A/T C C G

The HAP1 factor is also required for the transcription of the HEM 13
(coproporphyrinogen III oxidase) gene in the absence of heme. This
observation adds more complexity to the function of the HAP1
regulator. It has been suggested that HAP1 could be acting indirectly,
perhaps inhibiting the expression of an activator. However, if it acts
directly on the DNA, intriguing questions about the nature of the UAS
and what makes it specific for heme-deficient activation are raised
(Verdiere et al., 1991). Structural and functional studies of the HAP1
factor should provide better understanding of the relationship between
DNA binding activity and transcriptional activation.
1.4. Homology between mammalian and yeast regulatory genes.
The number of genes encoding transcription factors found in yeast and
their homology to genes of higher eukaryotes is growing rapidly. In
several cases, genetic studies in yeast and molecular or biochemical
studies in mammalian cells have converged on highly related
transcriptional activators that bind to the same DNA sequences
(Guarente and Bermingham-McDonogh, 1992). This is the case for the
mammalian serum response factor, SRF, a transcription factor that
activates the fos gene upon serum starvation In yeast, a factor encoded
by the MCM1 gene was shown to bind in vitro to the same site as SRF.
The MCM1-binding site is adjacent to binding sites for the MATal and

MATa2 mating type regulators and apparently helps al and oc2
proteins bind to the promoters and activate a-specific genes and repress
a-specific genes. The SRF protein also cooperates with an adjacent
mammalian binding protein to cause induction in much the same way as
MCM1 cooperates with al and a2.
A second example of a type of factor that has been conserved is/are
the CCAAT box-binding protein(s) shared by yeast and mammalian
cells; in mammals, several distinct factors bind to different types of
CCAAT boxes that bear different flanking sequences (Dorn et al.,
1987). Several yeast genes contain CCAAT-box consensus sequences
and at least one CCAAT-box has been demonstrated to bind the
HAP2/3/4 complex, which regulates many different nuclear genes that
are involved in the biogenesis of mitochondria. The mammalian
equivalent of the HAP complex (CPI A, CP IB) functions as a more
genera] factor that works in combination with other highly regulated
activators that act at a distance. The yeast HAP complex can also
activate transcription of genes involved in mitochondria biogenesis, in
combination with other yeast factors (see below) in response to
appropriate carbon sources.
It is worth mentioning that studies of the regulation of the COX6
(cytochrome c oxidase, VI) gene promoter (Trawick et al., 1992) have

revealed an 84-bp segment, located about 300 bp from the transcription
initiation site, involved in the transcriptional control of the C.OX6 gene
by carbon source; this segment contains a consensus recognition site for
the HAP2/3/4 activator and a sequence that matches the consensus site
for yeast ABF1 factor. ABF1, or ARS-binding factor I, is a yeast
protein that binds specific DNA sequences associated with several ARSs
(autonomous replicating sequences). ABF1 acts as an activator of
origins of replication binding to ARSs sequences. It also acts as a
repressor at the silent loci HML and HMR (mating-type information
loci). In the last few years, more ABFl-binding sites have been
identified in the C7C7(iso-l cytochrome c ) gene promoter and in other
genes involved in the control of cell growth (Grivell et al., 1991).
Therefore, the yeast ABF1 factor is a multifunctional DNA-binding
protein which plays a triple role; it is a repressor at the mating-type loci,
an activator of DNA replication and an activator of transcription at
several yeast gene promoters (Diffley and Stillman, 1989; Buchman and
Komberg, 1990). Linker scanning mutagenesis assays have shown that,
the HAP2/3/4 responsive element in the COX6 gene promoter is
inoperative in the absence of the ABF1 site. Thus, in COX6, the HAP
consensus site does not function as an independent UAS-like element,
although it is necessary for full transcriptional activation through this
84-bp domain. However, no HAP complex has been found to bind to the
CCAAT box adjacent to the required ABF1 site. It has been suggested

that factor(s) that bind to this site may act cooperatively with ABF1 in
the induction of transcription. These findings, however, do not imply
that all CCAAT boxes necessarily require ABF1 or other adjancent
factors. The proximity and interdependence between these regulatory
sites (ABF 1/CCAAT) have not yet been studied in other yeast
Signal transduction pathways converge ultimately at the level of
transcriptional activation to produce specific patterns of gene expression
in response to enviromental stimuli. In the case of higher eukaryotes the
stimuli are hormones or growth factors and in the case of yeast, the
stimuli are primarily nutrients. The initiation of transcription mediated
by these signaling pathways is regulated by the coordinate expression
and/or activation of specific transcription factors that bind to the control
regions of genes. Hormonal activation of signal transduction pathways
results in the transcriptional stimulation of many cellular genes,
(Hoeffler, 1992). Studies of mammalian genes responsive to activators
of protein kinase A and C have identified similar DNA regulatory
elements that mediate responses to these agents; the octameric cAMP-
responsive element (CRE, 5-TGACGTCA-3') differs from the
heptameric phorbol ester (TPA) reponsive element (ARE, 5'-
TGAGTCA-3') by only a single base (see Table 1). Several studies have
demonstrated that the oncoproteins jun and fos dimerize, producing a

heterodimer via leucine zipper, and mediate transcription through
sequence-specific binding to the ARE (or TRE) motif when protein
kinase C is activated, ( Angel et al., 1987; Lee et al., 1987; Parker et al.,
1989;). The jun/fos heterodimer is called API complex. There is
evidence for transcriptional factors in yeast which bind to ARE
sequences in viral (SV40 ARE) and yeast (amino acid biosynthesis)
promoters and activate transcription (Harshman et al., 1988). Yeast
contains a family of ARE-binding factors, which contain leucine zipper
domains, with at least two characterized members, GCN4 and yAPl.
The GCN4 factor, which is a general activator of amino acid
biosynthesis, contains the same DNA-binding domain as the avian
oncoprotein jun. In fact, when c-jun was expressed in yeast, the DNA-
binding domain of jun was bound to the normal GCN4-binding sites in
yeast (Struhl, 1987). The yeast API factor (yAPl) was isolated and
characterized by its ability to bind to the ARE sequence and activate
trancription, (Parker et al., 1989).This novel yeast protein was named
yAP 1 because of the identical footprinting formed on the ARE by this
factor and the human API factor. The strong similarity between the
DNA targets of yeast yAPl and GCN4 factors (Table 1) leads to the
suggestion that GCN4 may also bind to the ARE sequence and activate
transcription in yeast. Harshman et al. (1988) have shown that GCN4
binds as well to the ARE motif as to the GCRE however, experiments

carried out by Parker et al. (1989) using AKE-CYCl-lacZ gene fusions,
showed that the levels of (3-galactosidase activity produced in a
yap7mutant strain were not higher than the same fusion lacking the ARE
upstream activation site, (S. Moye-Rowley, unpublished results). These
results suggest that the authentic yeast GCN4 protein can discriminate
between the ARE and the yeast GCRE. The presence in yeast cells of
several other ARE-recognizing proteins, not yet characterized, may
provide an explanation for the nonessentiality of the cloned YAP1 gene
(Parker et al.-, 1989).
As mentioned before, another DNA-regulatory element responsive to
stimuli activation of signal-transduction pathways is the CRE (Table 1)
or cAMP-responsive element found in mammalian cells. The cloning of
human placental CRE-binding protein (CREB) (Hoeffler et al., 1988),
marked the begining of a long series of reports describing proteins
related to CREB and to the activating transcription factor, ATF. ATF
factors bind the same consensus sequence as CREB, but are not
responsive to cAMP. Most of the ATF factors do not yet have a known
biological function. All these related factors are grouped in the
ATF/CREB family of transcription factors.
Several laboratories have demonstrated that the amino-terminal
sequence of CREB is responsive to activation by protein kinase A,
(Schutz et al., 1992; Hoeffler, 1992). Experiments that overexpressed

the CREB protein in mammalian cells demonstrated that the levels of
endogenous protein kinase A were insufficient to activate the
overexpressed CREB in order to see an increase in transcription of a
reporter gene driven by the human a-gonadotropin promoter (a cAMP-
regulated gene). There is also evidence that CREB is phosphorylated by
protein kinase C and by calcium calmodulin-dependent kinases, but the
functional consequences of these phosphorylations are not yet clear.
(Sheng et al., 1991). Another CREB-like factor, called CREM, has also
been identified in mammalian cells and seems to functionally counteract
the activity of CREB, (Vincent and Struhl, 1992). The palindromic CRE
(Table 1) is composed of two functional CGTCA units which overlap in
opposite DNA strands to form a symmetrical CRE (Schutz et al., 1992).
The CREB proteins bind as dimers to the CGTCA units by using a
conserved leucine zipper dimerization domain adjacent to a basic region
that interacts directly with the DNA (Abel and Maniatis, 1989). Despite
this, some well studied cAMP-regulated genes, like c-fos and human
proenkephalin genes, contain only one crucial CGTCA unit forming an
asymmetric CRE.
A major metabolic signaling pathway in yeast involves cAMP and
protein kinase A. By manipulation of the RAS-cAMP signaling complex
it has been shown that changes in intracellular cAMP concentrations
produce coordinated changes in all mitochondrial cytochromes when

cells are grown on a non-fermentable carbon source (Schroetke, 1987).
The capacity of transcription factors to distinguish between ARE and
CRE elements has been conserved from yeast to humans. Experiments
with mammalian CREs and yeast factors have shown that yeast has
several proteins with a DNA-binding specificity identical to that of
mammalian ATF/CREB ones. It has also been shown that an
ATF/CREB mammalian binding site can act as an efficient UAS in
yeast, that the yeast ATF/CREB binding protein(s) can be
phosphorylated by the cAMP-inducible protein kinase, and that
phosphorylation is important for efficient binding (Jones and Jones,
Interestingly, studies in mammalian cells have shown that AP-1
proteins bind efficiently to ATF/CREB binding sites but ATF/CREB
proteins do not bind well to AP-1 sites. Individual AP-1 and ATF/CREB
proteins have distinct dimerization properties and interactions within
and between these families generate a large number of potential
transcriptional regulatory proteins. Recently, a new yeast gene ^4CR 1,
has been isolated and characterized (Vincent and Struhl, 1992). Unlike
the positive factors observed by Jones and Jones, this novel gene
encodes an ATF/CREB repressor whose leucine zipper structure
strongly resembles that of the mammalian transcriptional regulatory
protein CREB. Moreover, extracts from acrl deletion strains contain

one or more ATF/CREB-like DNA binding activities. These genetic and
biochemical observations suggest that S. cerevisiae may contain a
family of ATF/CREB proteins that function as transcriptional repressors
or activators.
Studies with mutant his3 promoters have shown that ACR1 protein
appears to repress transcription by at least two distinct mechanisms; One
mechanism involves competition between ACR1 and transcriptional
activators for ATF/CREB sites, which would explain why ATF/CREB
sites are poor activating sequences. Another alternative is that ACR1
can repress transcription mediated by other promoter elements necessary
for activation; ACR1 may cause a promoter activation interference when
it binds in between critical promoter elements.
A number of examples have appeared that demonstrate that the
transcriptional machinery is more similar between yeast and mammals
than previously thought. Proteins analogous to mammalian CCAAT
factors as well as a TATA-binding protein (Cavallini et al., 1988) have
beeen detected in yeast. This conservation suggests that the basic
eukaryotic process of transcription by RNA-polymerase II may be
studied in yeast to yield results that can be extrapolated to higher cells.

1.5. Objective of this work
In this work, the complete promoter sequence of the HEM2 gene and
putative regulatory binding protein sites are reported. The HEM2 gene
codes for 5-aminolevulinate (ALA) dehydratase, which is the second
enzyme of the heme biosynthetic pathway (see Figure l).The enzyme
has been purified from genetically engineered yeast and biochemically
characterized (Borralho et al., 1990). The coding sequence of the HEM2
gene and part of the flanking regions have been previously reported by
Myers et al. (1987) and its similarity to the human ALA dehydratase c-
DNA permits successful expression of the human enzyme in
Saccharomyces cerevisiae (Schauer and Mattoon, 1990).
The interest in this work in the upstream sequence of the HEM2 gene
comes from the fact that there is considerable evidence that heme and
cytochrome biosynthesis can be regulated through the second enzyme of
the pathway; ALA dehydratase (porphobilinogen (PBG) synthase).
Studies have shown that ALA dehydratase is sensitive to catabolite
repression; this enzyme is substantially repressed by glucose, leading to
the suggestion that catabolite control of cytochrome biosynthesis is
mediated through ALA dehydratase (Jayaraman et al., 1971).There is
also evidence of repression-resistance of ALA-dehydratase in some
mutant strains (Mattoon et al., 1989) which is consistent with the

suggestion that ALA-D is rate-limiting for heme and cytochrome
A major metabolic signalling pathway in yeast involves cAMP and
cAMP-dependent protein kinase. Metabolic signals such as nitrogen
starvation and glucose are transmitted to this system by membrane-
bound complexes containing adenylate cyclase and regulatory proteins.
One of these regulatory components is the RAS protein which is a
GTP/GDP-binding protein, homologous to the mammalian RAS
oncoproteins. In fact, mammalian RAS proteins can substitute for those
of yeast encoded by the yeast RAS1 and RAS2 genes, and activate
adenylate cyclase (Powers et al., 1984). By manipulation of this RAS-
cAMP system, it has been shown that increasing intracellular cAMP
concentrations can produce a coordinated increase in all mitochondrial
cytochromes when the cells are grown on a non-fermentable carbon
source, (Schroetke, 1987; Mattoon et al., 1990). Because this effect is
observed only when cells are grown in non-fermentable carbon sources
but not in glucose medium, it has been suggested that the non-
fermentable carbon source normally signals a down-regulation of
adenylate cyclase so that cytochrome synthesis is cAMP-limited.
Studies with yeast cAMP-permeable strains and a CYCl-LacZ fusion
plasmid, demonstrated that the cAMP acts on transcription of the CYC1
gene which encodes apo-isol-cytochrome c. In glucose medium, cAMP

decreased (3-galactosidase about 3-fold but, in glycerol medium, the
activity of the reporter enzyme was stimulated about 2- to 3-fold
(Mattoon et al., 1990).
Uncoordinated production of cytochromes and heme has been
observed when yeast cells bearing a mutation in the BCY1 gene, which
encodes a non-functional regulatory subunit of the protein kinase A,
were grown on glycerol. Some porphyrin accumulation has also been
observed in certain mutant yeast strains growing in glucose. These
strains are r/w-minus mutants, which lack mitochondrial DNA, and
bear a ras2vall9 mutation also, which over-stimulates the adenylate
cyclase enzyme. These results suggest that cAMP stimulates heme
biosynthesis by way of protein kinase A (Mattoon et al., 1990). In fact,
it may be that cAMP coordinates heme and apocytochrome
Based on this background, a search for potential regulatory binding
sites in the promoter region of the HEM2 gene is presented in this work.
In order to identify a possible cAMP responsive element(s) within the
promoter, a transcriptional reporter system for the HEM2 gene was
constructed by attaching the HEM2 to the gene PH05, which encodes a
repressible yeast acid phosphatase (AP). The reason for choosing such a
reporter is due to the simplicity of the AP activity detection by specific

staining of intact cells, because the enzyme is localized in the
periplasmic space. Other useful characteristics of the system are that
the enzyme expression is regulated in response to inorganic phosphate
(Pi) and that there is a 500-fold difference between the repressed (high
Pi) and derepressed (low Pi) levels of AP activity, which makes the
system very suitable for studing gene regulation in S. cerevisiae.

Fig. 1 Heme Biosynthesis Pathway.

Transcription factors found in yeast and mammalian cells and their target sequences.
ORGANISM Os-element SEQUENCE Trans- FACTOR(s) Heterologous Trans- FACTOR(s)
Mammalian ARE TGACTCA API (jun/fos) yAPl
Mammalian CAAT-box CCAAT CP1B/CP1A HAP2/HAP4

Bacterial strains: DH5aF':(F'/endAl hsdR17 supE44 thi-1 recAl gyrA
relAl A(lacZYA-argF)U169( 80 dlacA (lacZ)M15))
GM2163: ( F-ara-14 leuB6 thi-1 fhuA31 lacYl txs-78 galk2 galT22 supE44
hisG4 ipsL136 (Str ) xyl-5 mtl-1 daml3::Tn9 (Cam) dcm-6 mcrBl hsdR2
mcrA.) This strain lacks an endogenous adenine methylation at GATC
sequences. Therefore, it could be used for making DNA susceptible to
cleavage by the restriction enzyme Bell.
Competent cells; The bacterial strains used in this experiment were made
competent following the protocol of Inoue, Nojima, and Okayama (1990):
Media and assays; LB liquid media and plates were prepared as described
in Current Protocols For Molecular Biology (Asubel et al., 1989). Selective
media contained 100 ug/ml ampicillin, 15 ug/ml tetracycline and 17 ug/ml
chloramphenicol as final concentrations.
The blue/white color screening was carried out on LB plates supplemented
with ampicillin, 0.5 mM IPTG and 40 ug/ml X-Gal.
DNA manipulation; Standard procedures for the manipulation of plasmid
DNA were followed as described in Current Protocols in Molecular Biology
(Asubel et al., 1989).

2.1. Plasmid constructions and DNA sequencing
2.1.1. pHEM22 (4.7Kb.)
This plasmid was constructed using pHEM21 (6.2 Kb.), obtained from a
Nsi 1 digestion of the pHEM2 plasmid, and the phagemid pGem7ZF(+)
(3.0Kb.) Figures 2, 3 and 4 Both molecules were digested with BamHI and
Hindlll and the 1.7Kb fragment of the promoter was purified from a 1%
agarose gel (BioRad), following the Gene-Clean purification procedure fr om
Promega, Madison,Wi.. The fragment was ligated into the BamHI / Hindlll
poly linker sites of pGem7ZF vector. The ligation mixture containing the T4
ligase was incubated at 16 C overnight as described in the New England
BioLabs protocol (N.E. BioLabs). Transformation of DH5aF competent cells
was earned out as described above, and transformants were selected on LB
plates supplemented with ampicillin, 100 ug/ml.
2.1.2. Nested deletions from the 3' recessed end of the promoter using
pHEM22 was digested with Kpnl and Hindlll (Figure 5) and treated with
ExonucleaseUI (N.E. BioLabs) as described in Current Protocols (Asubel et
al., 1989). ExoIII catalyzes the stepwise release of 5'mononucleotides from
the 3' hydroxyl recessive end of duplex DNA. The digested Kpn I site
confers resistance of the four-base 3' extension to ExoIII. The exonuclease

reaction was carried out in a 30 ul volume; 25 ul of digested plasmid DNA
(100 pmole/ml of 3'-generated ends), 3 ul of 10X Exonucleaselll buffer, and
3 ul of enzyme. Aliquots of 10-ul reaction mixture were stopped at one-
minute intervals, using a pH shift (from pH 8.0 nuclease buffer, to pH 5.0
mung bean nuclease buffer) and incubating at 65 C for 10 minutes. Mung
bean nuclease (N.E. BioLabs) was used to create blunt ends in the molecules
. This enzyme is a single-strand-specific DNA and RNA endonuclease
isolated from mung bean sprouts. It removes any single-stranded regions from
DNA and RNA, resulting in ligatable blunt-ends. After the incubation with
mung bean enzyme at 37 C for 30 minutes (3 ul mung bean nuclease 10X
buffer, 3 ul DNA from exoIII reaction, 20 ul of doubly deionized sterile water
and 4 ul of mung bean nuclease), the reactions were stopped with 1 ul of 0.5
mM EDTA, and the DNA was ethanol-precipitated after phenol-chloroform
treatment to eliminate the proteins.
The DNA was resuspended in doubly deionized sterile water (10 ul) and a
ligation mixture containing T4 Ligase (N.E. BioLabs) was made to religate
the molecules. The ligation mixture was used to transform competent DH5aF
cells after treatment of the mixture at 65 C for 10 minutes to inactivate the
ligase. The selection of the deleted plasmids was made on LB plates
supplemented with 100 ug/ml of ampicillin.

Single-stranded DNA from pHEM22-derived vectors was made by
treatment with 0.2 N NaOH and 0.2 mM EDTA at 37 C for 30 minutes.
Neutr alization was carried out with 0.1 volume of 3 M sodium acetate (pH 5)
and DNA was precipitated with 3 volumes of ethanol (-70 C for 15 minutes).
The pelleted DNA was washed with 75% ethanol and resuspended in 7 ul of
doubly deionized sterile water.
The sequencing reactions were carried out by the dideoxy-chain
termination method (Sanger et al.,1977) using a modified form of T7
polymerase (Sequenase Kit from U.S.Biochemical,Cleveland ,OH.) and [a-
3^S]dATP as the isotope (Amersham, Arlington IL.).
Three different primers were used in the reactions; a 17-mer (-40) M13
sequencing Primer (included in the USB kit), a 17 mer reverse primer kindly
provided by Dr. G. Bajszar, and a Specific 17 mer Oligonucleotide purchased
from Oligos etc. Inc. The primers were all used at the same concentration of 3
ng/ul in the sequencing reactions.
2.1.3.a. Gel preparation
The sequencing glass plates were cleaned with a non-abrasive soap, rinsed
with 3-4 ml of 70% ethanol to ensure that all the soap was removed and
wiped diy with lint-free Kimpwipes tissues. Both of the plates were silanized

with "Sigmacote" from Sigma Chemicals: 2 ml of silanizing agent were
spread over the plates, using Kimpwipes, and allowed to dry for a few
minutes. 0.4 mm spacers were placed at each side of the longer plate before
the assembly. The assembly of the plates was carried out using brown
packing tape to seal the bottom of the plates and clamps along the sides to
keep the plates tightly together.
The following recipe was used to prepare the gel solution;
urea (BRL)
10X TBE (tris-borate-EDTA)
40% acrylamide (19:1 ratio) stock
45 g
10 ml
15 ml
25 ml. (approx.)
Acrylamide stock solution: 380 g/1 acrylamide (BRL) and 20 g/1
bisacrylamide (BRL) were dissolved in water with stirring (under the hood),
brought to a final volume of one liter and filtered through 3MM Whatman
paper in the dark. The solution was stored at 4 C in a dark bottle.
One ml of 10% APS, ammonium persulfate, was added to the gel solution
and 10 ml of this solution were poured in a 25 ml beaker. Five ul of the
polymerizing agent TEMED (N,N,N',N'-tetiamethylethyllenediamine) were
mixed with the 10 ml of gel solution, containing APS, and the mixture was
rapidly poured in the sides of the assembled plates to seal the sides and
bottom of the plates. The pouring had to be done quickly because of the large

amount of TEMED used in the small fraction of gel solution, 10ml. After 15
minutes, the rest of the gel solution plus 15ul of TEMED was poured in a 60-
ml syringe. Holding the plates in a 45 degree angle, the gel was injected
along one side of the plate. The angle of the plates and the rate of the addition
was adjusted to avoid air bubbles. After the gel addition, the plates were
placed at a 5-degree angle, and a shark's tooth comb was clamped in place to
prevent gel from coming between the comb and the bigger plate.
Polymerization was completed in about 30- 35 minutes. The gel cassette was
then mounted on the sequencing apparatus and a IX TBE running buffer
was added to both anode and cathode chambers. A 0.6 X TBE running buffer
was used in some cases to decrease the running time without compromising
the resolution.
A BRL power supply Model 300/3000 was set for the electrophoresis; The
optimum power setting is found by multiplying the gel volume by 0.8-1.0
watts/cm In the present, for 70 cm of gel volume, 55-60 watts were applied
being the constant voltage of 1800 volts. After a 20 minutes of prerun, the
DNA samples (previously heated at 80 C for 2 minutes) were loaded: 2 ul of
sample were placed in the wells after washing them with a Pasteur pipette to
remove the urea. Timing of multiple sample loads was determinated by
monitoring the bromophenol blue dye (BPB), which migrates at the same
speed as oligonucleotides of 40 bases. The second loading was added when
the BPB was 5 cm from the bottom of the gel.

Once the run was completed, one of plates was separated carefully, with
the gel remaining attached to the other plate. The gel (still attached to the
plate) was then treated with 10% acetic acid/10% methanol solution for 15
minutes, peeled from the glass plate with 3MM chromatography paper and
dried with a vacuum-heater overnight. The film used for the autoradiograph
was Kodak XAR-5 film (14"X17"), and the exposure time was 48 hours at
room temperature.
2.2 Construction of a transcriptional reporter system for the HEM2 gene
2.2.1 pNZ8P plasmid construction
Plasmid pNZ8B (4.2Kb) obtained from George Bajszar, University of
Colorado, Colorado Springs (Zvonok et al.,1988), was digested with BamHI
and blunt ends were made using T4-polymerase as described in New England
BioLabs protocol (Figure 6). The linear molecule was then treated with
Calf Intestinal Phosphatase (CIP) for 1 hour at 37 C according to
N.E.BioLabs. The phosphatase reaction was stopped with 2 ul of 0.5 M
EDTA followed by an incubation at 65 C for 10 min to inactivate the
enzyme. After phenol\chloroform purification, the DNA was ethanol-
precipitated and resuspended in sterile deionized water. A Pstl
phosphorylated linker (GCTGCAGC) was ligated into the 4,170 bp fragment
to create pNZ8P plasmid (Figure 6) which now contained two Pstl sites
flanking the PH05 gene (1.5 KB).

2.2.2. Construction of the plasmid pRBl
pNZ8P was digested with Pstl, and the 1.5Kb fragment corresponding to
the PH05 gene was isolated from a 1% agarose gel (Figure 6). pHEM21 was
digested with Nsil and treated with calf intestinal phosphatase as described
The 1.5 fragment containing the reporter gene was ligated into the Nsil site
of pHEM21 using T4 ligase as described before. pRBl was digested with Sal
I to check for the right orientation of the insert which is downstream of the
2.2.3 Construction of plasmids pIRl and pIR2
pRBl was cloned into the GM2163 strain to make the plasmid susceptible
to Bell digestion The modified plasmids were cleaved with Bell and BamHI
restriction enzymes. The 3.4-Kb. fragment, obtained from the digestion and
containing the promoter fused to the reporter gene, was isolated from a 1%
agarose gel using the GeneClean purification protocol from Promega,
Madison, WI. The fragment was ligated into the CIP-treated BamHI site of
the shuttle vectors YCp50 and YEp24 The ligation mixtures were used to
ttansform E. coli (DH5aF', competent cells) and tetracycline-sensitivity and
ampicillin-resistant transformants were selected. Restriction enzyme
digestions were earned out on the DNA extracted from the transformants to
confirm the correct plasmid construction containing the promoter and the
reporter gene Figure 7.

Fig. 2 pHEM2 plasmid. This plasmid contains the HEM2 gene and its promoter.

Fig. 3 pHEM21 plasmid .

3000 bps
Fig. 4 pGEM7ZF(+): Phagemid used for sequencing.

Linearized by HindWl and
Kpn\ digestion
| >4/711,306
| | Pad,377
Sad, 1707
Exolll/Mung bean nuclease
This end is
Fig. 5 Nested deletion strategies for the HEM2 promoter.

\ Nar\
i I
i \SphI
' !Pst\
Spe I
Nhe I
l San
Ban t insert Psfl :
|i^iH ~ linker, then
Kpn\ v cleave. Ji
ffci mrjWrvPsnsm
Ndel i
mi i
U Sail
with Nsli
>1-11 I H//7dlll,105
7177,4 Srt304
5628, Mfel
Seal, 1051
4270, Eapl
3982, San
3893, Sph
Fig. 6 pRB1 construction.

fig. 7 pIR1 and pIR2 plasmids.

3.1. Sequencing experiments
3.1.1. Identification of the nested deletions from nHEM22
The pGEM single-stranded system is designed to allow the production of
cloned DNA as single-stranded circular molecules suitable for mutagenesis,
sequencing and other applications. Single-stranded DNA can be produced
from any of the pGEM-Zf(+) and pGEM-Zf(-) series of plasmids because
they contain the origin of replication of the filamentous bacteriophage fl.
However, the generation of single-stranded circular molecules was carried
out following the alkali denaturation procedure described in Materials and
These chimeric phage-plasmid vectors, or phagemids, also contained SP6
and T7 RNA polymerase promoters flanking a region of multiple cloning
sites within the a-peptide coding region of the enzyme p-galactosidase. The
inactivation of the P-galactosidase enzyme by insertion of a DNA fr agment at
the poly linker site, allows direct color screening of the recombinant clones on
plates containing IPTG (inducer) and X-Gal (substrate). Clones containing
plasmids with the inserts give white colonies, whereas those containing the
vector produce blue colonies. The identification of the transformants
containing pHEM22 and its progressive deletions at the promoter site, Kj
through K \ q, were carried out on ampicillin plates containing inducer and

substrate, to eliminate the possibility of isolating molecules with a complete
deletion of the HEM2 promoter. Most of the plates contained a few white
colonies except for those containing cells transformed with the DNA
corresponding to the 8- and 9- minute fractions of the ExoIII digestion, where
no colonies were found. This could be due to an extensive nuclease digestion
that removed the whole promoter (insert), and also affected the following
ampicillin resistance gene, which didn't allow the growth of the transformants
on the ampicillin plates. Each of the white colonies obtained was used to
inoculate a culture which was grown overnight in 5 ml of LB media
supplemented with ampicillin. The amplified plasmid DNA from each
culture was purified using a DNA purification resin in 7M guanidine-HCl
from Promega, Madison, WI. The amount of DNA obtained from each
culture was approximately 10 ug.
In order to determine the extent of the nuclease digestion, 2 ul (0.4 ug) of
DNA obtained from each time point was treated with EcoRI and BamHI
enzymes for 3 hours and finally analyzed in a 1% agarose gel. The results
from the agarose gel indicated that the clones K4, K5, K7 and K10 contained
600 bp, approximately, of the promoter and that clones K1 and K6 contained
a promoter segment of about 1000 bp.

3.1.2. Promoter sequence of the HEM2 gene
The results of the gel indicated that the best candidates for sequencing were
the plasmids obtained at the time points corresponding to 1,4,5,6,7 and 10
minutes of digestion. Using the computer program GELCAP and a digital gel
reader, Gelmate 2000, the different sequences obtained were introduced in
the computer for further comparison.
The overlapping regions of the sequences and the ambiguities found in
some bases of the overlapping area were mostly resolved using another
computer program, ASSEMBLE (PC gene), which facilitated putting
together all the sequence data from the gels. The first 500 bp of the proximal
promoter had already been sequenced by Steven Quigley at the University of
Colorado at Colorado Springs, (Quigley, 1990, unpublished.). Our sequence
data overlapped with his and the end result of all the sequences obtained is a
promoter 1895 bp long, Figure 8.

Naml Xctl
ABF1 HAP2/3/4
alndlll Hhel
Fig. 8. Sequence of the 1890-bp region upstream of the HEM2 gene
Restriction sites and potential regulatory motifs (ABFVCREB, HAP2/3/4, REB1, YAP) are underline*

Map of motifs in the HEM2 promoter
w W4

, REB1
Jatg (HEM2)


3.1.3 Analysis of the sequence and putative regulatory sites
The DNA sequence analyzer Clone Manager was used to obtain a
restriction map of the promoter sequence. The results coincide with the
restriction analysis carried out on the pHEM21 plasmid (this work) with the
commonly used restriction enzymes. The promoter does not contain many
useful restriction sites (Figure 8) except those already mapped.Using other
computer programs (PC gene, Intelligenetics, CH.) which search for
eukaryotic regulatory elements in promoters, the following potential protein-
binding motifs were found (Figure 8 ): ABF1, REB1, RAP1, yAPl, CREB
and a putative TATA-box centered at position -73 and correlated with a
HAP2/3/4 complex binding-site 11 bp upstream of the putative TATA-box.
This putative HAP2/3/4 is almost a perfect match to the consensus sequence
TNRTTGGT (N= any nucleotide, R= A or G) established by Guarente, 1987.
Another three HAP2/3/4 putative binding-motifs are also found along the
promoter sequence. In contrast, the HEM1 (5-aminolevulinate synthase) gene
promoter contains only one known consensus HAP2/3/4 binding-site that is
required for HEM1 transcription (Keng and Guarente, 1987).
3.1.3.a ABF1: a yeast regulatory protein
This factor is a multifunctional DNA-binding protein that, depending on
the context, can act as a transcriptional activator, as a transcriptional
repressor or an activator of DNA replication at the origin sites (Buchman and
Komberg, 1990). The ABF1 gene has been cloned, sequenced and shown to

be essential for viability. The predicted amino acid sequence contains a
structure related to a zinc finger DNA-binding motif (Diffley and Stillman,
Results of recent experiments demonstrate that ABF1 protein functions as a
gene activator in vivo; when a DNA segment containing the ABF1 binding-
site was placed in a plasmid upstream of the CYC1 TATA-box, the
transcription of the reporter gene ((3-galactosidase) was increased with
respect to a construction lacking the ABF1 site, which acts as a UAS. It
appears, however, that ABF1 is intrinsically weaker than other well-
characterized yeast regulatory factors like GAL4 or GCN4. Buchman and
Komberg (1990) have demonstrated that ABF1, alone, functions as a weak
yeast activator, but it can act as a strong transcription factor in conjunction
with a T-rich promoter element. This has lead to the suggestion that ABF1
factor activates transcription synergistically in combination with other weak
activating factors.
The HEM2 promoter contains a perfect consensus ABFl-binding site
(RDCNYNNNNNACG; D = A,G or T; R = A or G; Y = T or C; N = A, G, T
or C) located about 1,000 bp from the start point of transcription.
Interestingly, this ABF 1 motif is flanked, at each side, by two putative
HAP2/3/4 binding-sites which could be acting cooperatively with the ABF 1
site in activation of transcription. Evidence of cooperativity between

HAP2/3/4 and ABF1 regulatory sites in yeast, has been recently shown in the
C.OX6 gene by Trawick et al., 1992.
3.1.3. b RAP1 yeast factor
Another regulatory consensus site found in the HEM2 promoter sequence is
the binding site for the repressor/activator protein, or RAP1, factor. Like
ABF1, RAP1 is an essential, multifunctional DNA-binding protein implicated
in transcriptional repression at the silencer sites of the HMR and HML loci.
It also acts as a transcriptional activator for a wide variety of genes involved
in RNA processing, glycolysis, cell growth and yeast telomere function,
(Verdier, 1990; Buchman and Komberg, 1990). The function of the RAP1
factor, if it binds to the consensus site found in the HEM2 promoter, may be
that of activator of transcription analogous to the ABF1 factor. However, the
RAP1 consensus site (AYCCRNRCA; R=purine, N=any nucleotide) found in
our sequence is quite far from the transcriptional initiation point (about 1.5-
Kb) raising some doubt about the importance of this site in the HEM2
transcriptional activation.
3.1.3. C REB1 yeast factor
The REB1 DNA-binding protein, or yeast rRNA enhancer-binding protein,
is a factor known to bind within the yeast rRNA enhancers, (Remade et al.,
1990). It now seems likely that REB1 may play a more general role in the
organization of the cell's DNA. The REB1 gene was cloned recently, and it

has been shown to be an essential gene reflecting its important function
within the cell. The REB1 consensus binding site, CGGGTARNNR, is found
in the promoter of many unrelated genes ACT1, RAP1, TRP5, SIN3, SWI5
and TOPI (Stillman et al., 1990). REB1 was also shown to bind within the
GALJ-GAL10 UAS, leading to a localized exclusion of nucleosomes.
Although this site is not essential for the GAL4-dependent transcription of
GALJ and GAL JO genes, it may be involved in GAL4-independent
transcription changing the structure of the surrounding chromatin when it
binds to the GALJ-GALJ 0 UAS site. However, the role of REB 1 in
transcription by RNA polymerase II is not very clear, and the physiological
role that this factor plays in the cell has yet to be established. Two putative
REB 1 binding sites, each with a perfect match to the consensus sequence,
have been identified in the HEM2 promoter. One of the sites is located about
222-bp from the translational initiation point and the other potential site is
placed 1.5 kb upstream from this point. It is worth mentioning that all the
promoters tested which contain REB 1 consensus sites bound the REB 1
protein when gel-shift assays were done. Preliminary experiments with this
factor have shown that, on its own, it exerts a modest effect on transcription
when tested as a UAS placed upstream of a reporter gene. This factor is
therefore similar to RAP1 and ABF 1 factors, and may play a role, in the
HEM2 activation, as an enhancer of the effect of another neighboring weak
activator. In fact, the REB1 consensus site that is further upstream, overlaps
in one base with the potential RAP1 binding site.

3.1.3.d vAP-1 and ATF/CREB regulatory proteins in yeast
DNA-regulatory elements responsive to stimuli activation of signal
transduction-pathways are the CRE or cAMP-responsive element, and the
ARE or AP1 responsive element. Both elements were first identified in
mammalian cells but, a few years ago Parker et al. (1989) reported the
isolation and characterization of a new yeast factor (yAPl) able to bind to the
mammalian ARE sequence and activate transcription in yeast. The ARE
motif is present in the HEM2 gene promoter, 500 bp upstream of the
translation starting point, where it might act as an activator of transcription
depending upon the nutritional conditions in the medium. However, there is
not yet a well- characterized signalling pathway in yeast associated with
yAP 1 activation of transcription.
Multiple ATF/CREB-like binding activities have been detected in nuclear
extracts from Saccharomyces cerevisiae. Experiments with ATF oligo-
affinity (chromatography) purified yeast extracts have also shown
dependence on protein kinase A activity. The binding activity of partially
purified yeast ATF factors (yATF) can be enhanced in vitro by the addition
of protein kinase A, suggesting a possible link between cAMP signaling and
ATF/CREB protein function in yeast, (Jones and Jones, 1989). The DNA
sequence of the HEM2 gene promoter contains a putative ATF/CREB
binding site located 464 Bp (figures 8 and 9) from the translation starting

point. This putative CRE (ACACGTCA) found in the HEM2 promoter is
different from the thoroughly characterized consensus CRE (see Table
l).This putative CRE only contains one CGTCA crucial unit necessary for
the binding of CREB. However some well studied mammalian genes known
to be regulated by cAMP like c-fos, human proenkephalin and tyrosine
aminotransferase genes, have functional asymmetrical CRE's in their
promoters and therefore contain also only one CGTCA unit. This findings
suggest that an ATF/CREB-like protein could be involved in the
transcriptional regulation of the HEM2 gene in response to cAMP levels.
As mentioned before in the introduction, a new yeast gene called ACR1
has been recently isolated and encodes a novel ATF/CREB repressor. This
factor binds to ATF/CRE sites and is similar to the mammalian CREB- and
CREM- regulatory proteins (Vincent and Struhl, 1992). ACR1 factor contains
a leucine zipper domain, binds to the HIS3 promoter as a homodimer and
interacts poorly with the related API site. However, the role of protein kinase
A in the transcriptional regulation mediated by ACRl is not clear.
3.1.3.e. HAP1 yeast transcription factor
Heme activator protein (HAP1) consensus binding sites (as found in
UAS1 of the CYC] gene promoter) have not been found in the HEM2
promoter region. The HEM1 promoter also lacks a HAP 1-binding site (Keng
and Guarente, 1987). This could explain the observed porphyrin

accumulation in hapl mutant strains,(Mattoon, Caravajal and Guthrie, 1990;
Mattoon et al., 1990) where all cytochromes decrease without a proportional
decrease in porphyrins (heme). However, heme could regulate HEM2
expression through HAP 1-independent sites, as happens for the COX6 gene,
where two novel DNA sites mediate regulation of COX6 by heme in a HAP1-
independent manner. Analysis of the activity of different segments of the
HEM2 promoter will be necessary to understand what, if any,effect of heme
and the putative regulatory factors on the ALA-dehydratase expression may
3.2.Transcriptional reporter system for the yeast HEM2 gene
The principle of applying the PH05 reporter gene to the study of yeast
promoter activities is similar to that introduced by other authors, using the E.
coli lacZ gene fused to yeast promoters. The PH05 gene is one of the natural
yeast genes whose product is efficiently secreted out of the cytoplasm, so that
there is no need to employ special approaches to permeabilize the cells for
the enzymatic assays. The acid phosphatase expression can be detected easily
on agar plates by a quick colony staining method, (Zvonok et al., 1988).
As described in "Materials and Methods", the HEM2 gene promoter
encoding ALA-dehydratase has been attached to the promoterless PH05 gene
encoding a yeast acid phosphatase. The pRBl construction (see Figure 6)
contains the HEM2-PH05 fusion cassette (aprox. 3.5 Kb) flanked by the

BamHI and Bell unique restriction sites. Since the Bell site is blocked by
Dam methylation, the pRBl plasmid was transformed into a dam strain and
the transformants were selected by growth on selective medium containing 17
ug/ml of chloramphenicol.
Several E. coli-yeast shuttle vectors (YCp50, YEp24, pRS316) were used
in an effort to introduce the HEM2-PH05 cassette into the BamH I site of
each shuttle vector. The ligation reaction conditions were tested with
different amounts of vector, insert cassette, ligase enzyme and ATP. Time
and temperature conditions were also varied. However, the only
transformants obtained contained only the shuttle vector plasmids, suggesting
that no ligation reaction took place. In contrast, agarose analysis of the
ligation mixtures prior to transformation of E. coli, indicated that
recombinant molecules having greater molecular weight than that of the
vector were in fact formed. However, whenever such a ligation mixture was
used to transform E. coli, no recombinants were obtained. The reason for
these negative results is not known. Since the PH05 gene have been
successfully used as a reporter gene with other yeast promoters (Bajszar et
al., 1988) it appears likely that the HEM2 promoter sequence itself may be
responsible for the negative results. It may be suggested that this particular
construction is not tolerated by E. coli. Yeast integration experiments using
the HEM2-PH05 expression cassette are now being considered in order to
introduce this reporter system into the yeast genome.

4. Discussion
As a way of investigating the regulation of the yeast HEM2 gene, we have
determined the nucleotide sequence of a 1890-bp DNA fragment containing
the HEM2 promoter and additional upstream flanking sequence. The DNA
sequence was determined by preparing nested-deletion subclones and using
the di-deoxy chain termination technique (Asubel et al., 1989).
From physiological studies it has been shown that the yeast HEM2 gene,
encoding ALA-dehydratase, has a regulatory role in the synthesis of
poiphyrins and heme. It functions as the pace-maker of the heme biosynthesis
because its enzymatic activity is limiting for the pathway (Jayaraman et al.,
1971; Borralho et al., 1983). The expression of this enzyme is repressed by
glucose (Borralho et al., 1983, 1990) and preliminary experiments strongly
suggest that cAMP regulates its gene expression (Mattoon et al., 1990). The
potential regulatory sites found in the promoter suggest that HEM2 is subject
to a complex set of regulatory mechanisms.
The ABF l, RAP1 and REB1 transcription factors, whose consensus
binding sites have been found in the HEM2 promoter, share several
characteristics: they are very abundant, essential for cell viability, bind to
many sites in the genome and have related functions within the cell. When
bound to DNA, these proteins may influence the chromatin structure to
facilitate transcription, replication and other essential chromosomal functions.

As evidence supporting this suggestion, it has been shown that the GAL1-
GAL10 regulatory sequence contains a REB1-binding site which influences
the chromatin structure of the DNA, leading to a localized exclusion of
nucleosomes (Remade and Holmberg, 1992). ABF1 and REB1 act as weak
transcription activators when combined with the CYCl TATA-box but they
can act as powerful auxiliary transcriptional activators when associated with
other weak promoter elements, like T-rich sequences. These observations
lead to the suggestion that under inducing conditions of HEM2 expression,
these factors could bind to the promoter and facilitate transcriptional
activation of HEM2 through other more specific activators like yAP-1,
CREB-like proteins and the HAP2/3/4 complex.
Sequence searches for CRE-binding sites in the HEM1, HAP2, HAP3
HAP4 and CYCl known upstream sequences were negative. This result
favors the suggestion that cAMP regulates heme biosynthesis through a
CREB-like protein which will bind to the HEM2 promoter and activate or
repress transcription in a protein kinase A-dependent manner, depending on
the carbon source; Mattoon et al. have shown that a CYCl-lacZ reporter gene
fusion is stimulated by increased levels of cAMP and non-repressible carbon
sources, but repression of transcription of the lacZ gene is observed with high
levels of cAMP and glucose-repressing conditions (Mattoon et al., 1990).
These results, together with the evidence of the presence of an ATF/CREB
family of proteins in yeast and a putative ATF/CRE consensus-binding site in

the HEM2 promoter, could provide an explanation of the carbon source
regulation through cAMP in porphyrin biosynthesis and subsequently in
cytochrome formation.
The presence of an AP-1 consensus site (ARE) in the HEM2 promoter also
indicates that protein kinase C could be implicated in the regulation of the
HEM2 expression, as occurs in higher eukaryote genes which contain ARE
sites in their promoters By using in v/7ro-synthesized mammalian proteins, it
was demonstrated that formation of either heterodimeric complexes between
the fos and jun proteins or homodimeric complexes between jun proteins,
facilitates their binding to the ARE (TGAC/GTCA) consensus sequence. It
was also found that mammalian AP-1 factor activity could be modulated
through two mechanisms; (i) postranslational modification of the jun protein
by protein kinase C and (ii), changes in the expression of various jun and fos
proteins (jun/fos family) which form homodimers or heterodimers of different
stabilities (Trejo et al., 1992).
The yeast factors yAP-1 and GCN4 are the only two yeast ARE-binding
proteins so far characterized. GCN4 protein levels are regulated by
translational control of its transcript. Under amino acid starvation conditions,
translation of the GCN4 mRNA is activated by mechanisms not yet well
understood. Previous work (Harshman et al., 1988) has demonstrated that
GCN4 binds as well to the ARE as to the GCRE of the HIS3 promoter. These

data support the idea that GCN4 could, under derepressing growth
conditions, bind to the AP-1 site found in the HEM2 promoter and activate its
expression at the same time that it activates genes of amino acid biosynthesis
required for apocytochrome formation. Under repressing and amino acid
starvation conditions, GCN4 will only activate the amino acid biosynthesis
genes since the HEM2 gene expression will be blocked by a strong repression
regulation mediated by glucose. Alternatively, yAPl factor could be the
regulator that binds to the ARE sequence in the HEM2 and/or the GCN4
genes and activates their expression under non-repressible growth conditions.
Gel-shift experiments using the ARE and flanking sequences of the HEM2
promoter and nuclear extracts from yap-1 and wild type strains, with GCN4
factor present or absent, will be necessary to confirm the above hypothesis.
One common feature of eukaryotic promoters which contain CREB/ATF
and/or API binding sites is that they respond to environmental stimuli; for
example, mitogens, phorbol esters, viral infections, hormones or nutrients. In
the case of the yeast HEM2 promoter, the fact that it contains all the
mentioned putative regulatory sites (RAP1, ABF1, CREB, yAPl, REB1,
HAP2/3/4) indicates an already predicted complex regulation of this promoter
(Mahler and Lin, 1978), which makes the ALA-dehydratase expression a
limiting factor for heme and cytochrome formation. From this work, it can be
concluded that the HEM2 gene may be subject to a composite of regulatory
mechanisms: (i) The general and abundant ABF1, REB1 and RAP1 factors

may be involved in altering the chromatin structure and therefore producing a
localized exclusion of nucleosomes, allowing the transcription factors to bind
to their corresponding sites on the promoter. Interestingly, different
phosphorylated forms of the ABF1 factor have been found in nuclear extr acts
from glucose-repressed and derepressed yeast cells (Trawick et al., 1992).
This suggests that phosphorylation of yeast nuclear transcription factors is a
way of linking the nutrient stimuli with the transcription factor profile, (ii) the
HAP2/3/4 complex, which acts as a potent activator of transcription under
derepressing and heme-sufficient growth conditions, is modulated by the rate-
limiting HAP4 subunit; northern-blot analysis has revealed that the HAP4
mRNA levels increase fivefold in a shift from glucose to lactate (Olesen and
Guarente, 1990). The HAP2/3 complex may bind weakly to the HEM2 site(s)
under glucose-repressing conditions, but no activation of HEM2 expression
occurs because the HAP4 subunit is present in veiy low levels under these
conditions. Studies of the HAP4 promoter in search for regulatory sites, like
ARE and CRE, is necessaiy for understanding the mechanism of HAP4
carbon source regulation. If the HAP4 subunit is regulated by cAMP and/or
AP-1 factors, this could explain the fact that an abnormal increase in
intracellular cAMP coordinately enhances production of all mitochodrial
cytochromes in glycerol medium but not in glucose medium (Mattoon, et al.,
1990). Interestingly, by computer analysis of the HAP4 gene, we have found
a putative protein kinase A phosphorylation motif within the amino acid
coding region of the gene. This putative phosphorylation site is located in the

amino terminus domain of the protein required for interaction with HAP2 and
HAP3 factors, (iii) Evidence of the presence of CREB-like (Vincent and
Struhl, 1992) and AP-1-like (Parker et al., 1989) proteins in yeast and the fact
that the HEM2 promoter contains consensus DNA-binding sites for both
factors, leads to the suggestion that ALA-dehydratase is subject to regulation
by nutritional signaling transduction mechanisms, mediated by protein
kinases A and C respectively.

Abel, T., and T. Maniatis. "Action of leucine zippers." Nature 341(1989):24-25.
Angel, P., M. Imagawa, R. Chiu, B. Stein, R.J. Imbra, H.J. Rahmsdorf, C.
Jonat, P. Herrlich, and M. Karin. "Phorbol ester-inducible genes contain a
common cis element recognized by a TPA-modulated trans-acting factor." Cell
Asubel,F.M., R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith
and K. Struhl (Eds). Current Protocols in Molecular Biology. John Wiley &
Sons, N.Y.(1989) pp 7.2.1-7.6.13.
Borralho,L.M.,C.H.Ortiz, A.D. Panek, and J.R. Mattoon. "Purification of delta-
aminolevulinate dehydratase from genetically engineered yeast." Yeast 6
Borralho, L.M., A.D. Panek, D.R. Malamud, H.K. Sanders, and J.R. Mattoon.
"In situ assay for 5-aminolevulinate dehydratase and application to the study of
a catabolite repression-resistant Saccharomyces cerevisiae mutant." J.Bacteriol.
156 (1983):141-157.
Buchman, A.R., and R.D. Komberg. "A yeast ARS-binding protein activates

transcription synergistically in combination with other weak activating factors."
Mol.Cell Biol. 10 (1990):887-897.
Cavallini, B., J. Huet, J.L. Plassat, A. Sentenac, J.M. Egly, and P. Chambon.
"A yeast activity can substitute for the HeLa cell TATA box factor." Nature
334 (1988):77-80.
Diffley, J.F., and B. Stillman. "Similarity between the transcriptional silencer
binding proteins ABF1 and RAP1." Science 246 (1989): 1034-1038.
Dom, A., J. Bollekens, A. Staub, C. Benoist, and D. Mathis. "A multiplicity of
CCAAT box-binding proteins." Cell 50 (1987):863-872.
Dorsman, J.C., W.C. Heeswijk, and L.A. Grivell. "Identification of two factors
which bind to the upstream sequences of a number of nuclear genes coding for
mitochondrial proteins and to genetic elements important for cell division in
yeast." Nucl.Acids Res. 16 (1988):7287-7301.
Flick, J.S., and M. Johnston. "Two systems of glucose repression of the GAL1
promoter in Saccharomyces cerevisiae." Mol.Cell Biol. 10 (1990):4757-4769.

Flick, J.S., and M. Johnston. "Analysis of URS-mediated glucose repression of
the GAL1 promoter of Saccharomyces cerevisiae." Genetics 130 (1992):295-
Forsburg, S., and L.Guarente. "Identification and characterization of HAP4: A
third component of the CCAAT-bound HAP2/HAP3 heteromer." Genes Dev. 3
Gancedo, J.M. "Carbon catabolite repression in yeast." Eur.J.Biochem. 206
Geargakopoulos, T., and G. Thireos. "Two distinct yeast transcriptional
activators require the function of the GCN5 protein to promote normal levels of
transcription." EMBO J. 11 (1992):4145-4152.
Grivell, L.A. "Nucleo-mitochondrial interactions in yeast mitochondrial
biogenesis." Eur.J.Biochem. 182 (1989):474-493.
Guarente, L. "Regulatory proteins in yeast." Ann.Rev.Gent. 21 (1987):425-
Guarente, L. and O. Bermingham-McDonogh. "Conservation and evolution of
transcriptional mechanisms in eukaryotes." TIG 8 (1992):27-32.

Hai, T., F. Liu, W.J. Coukos, and M.R. Green. "Transcription factor ATF
cDNA clones: an extensive family of leucine zipper proteins able to selectively
form DNA-binding heterodimers." Genes Dev. 3 (1989):2083-2090.
Harshman, K., W. Moye-Rowley, and C. Parker. "Transcriptional activation by
the SV40 AP-1 recognition element is mediated by a factor similar to AP-1 that
is distinct from GCN4." Cell 53 (1988):321-330.
Hoeffler, J.P. "Structure/function relationships of CREB/ATF proteins". J.
Invest Dermatol. 98 (1992):21s-28s.
Hoeffler, J., T. Meyer, Y. Yung, J.L. Jameson, and J. Habener. "Cyclic AMP-
responsive DNA-binding protein; structure determined from a cloned placental
cDNA." Science 242 (1988): 1430.
Inoue, H., H. Nojima, and H. Okayama. "High efficiency transformation of
Escherichia coli with plasmids." Gene 96(1990):23-28.
Jayaraman, J., G. Padmanaban, K. Malathi, and P.S. Sarma. "Haem synthesis
during mitochondrogenesis in yeast." Biochem. J. 121 (1971):531-535.
Jones, R.H., and N.C. Jones. "Mammalian cAMP-responsive element can
activate transcription in yeast and binds a yeast factor(s) that resembles the

mammalian transcription factor ATF." Proc.Natl.Acad.Sci.USA 86
Keng, T., and L. Guarente. "Constitutive expression of the yeast HEM1 gene
is actually a composite of activation and repression." Proc.Natl.Acad.Sci.USA
84 (1987):9113-9117.
Lee, W., P. Mitchel, and R. Tjian. "Purified transcription factor AP-1 interacts
with TPA-inducible enhancer elements." Cell 49 (1987):741-752.
Lodi, T., and B. Guiard. "Complex transcriptional regulation of the
Saccharomyces cerevisiae CYB2 gene encoding cytochrome b2: CYPl(HAPl)
activator binds to the CYB2 upstream activation site UAS1-B2." Mol.Cell Biol.
Mahler, H.R., and Chi-Chung Lin. "Molecular events during the release of
c/e//tf-aminolevulinate dehydratase from catabolite repression." J.Bacteriol. 135
Mattoon, J.R., E. Caravajal, and D. Guthrie. "Effects of hap mutations on heme
and cytochrome formation in yeast." Genetics 17 (1990): 179-183.
Mattoon, J.R., D.E. Oliviera, M.N. Tenan, A.D. Panek, D.R. Malamud, and

L.M. Borralho. "Parallel changes in catabolite repression of haem biosynthesis
and cytochromes in repression-resistant mutants of Saccharomyces cerevisiae."
Journal of General Microbiology 135 (1989): 1217-1227.
Mattoon, J., W. Schauer, R. Schroetke, A. Plummer, D. Calabrese, D.R.
Malamud, L.M. Borralho, E. Carvajal, and A.D. Panek. "Coordinate regulation
of mitochondrial cytochrome formation in Saccharomyces cerevisiae". In:
Genetics of respiratory enzymes in yeasts. Edited by Lachowicz, T.M.
Wroclaw: Wroclaw University Press, 1990, p.16.
Myers, A.M., M.D. Crivellone, T.G. Koemer, and A. Tzagoloff.
"Characterization of the yeast HEM2 gene and transcriptional regulation of
COX5 and COR I by heme." J.Biol.Chem. 262 (1987): 16822-16828.
Oeschsner, U., H. Herman, A. Zollner, A. Haid, and W. Bandlow. "Expression
of yeast cytochrome cj is controlled at the transcriptional level by glucose,
oxygen and heme." Mol.Gen.Genet. 232(1992):447-459.
Olesen, J., S. Hahn, and L. Guarente. "Yeast HAP2 and HAP3 activators both
bind to the CYC1 upstream activation site, UAS2, in a interdependent manner."
Cell 51 (1987):953-961.
Olesen, J.T., and L. Guarente. "The HAP2 subunit of the yeast CCAAT

transcriptional activator contains adjacent domains for subunit association and
DNA recognition: model for the HAP2/3/4 complex." Genes Dev. 4
(1990): 1714-1729.
Oliviero, S., G.S. Robinson, K. Struhl, and B.M. Spiegelman. "Yeast GCN4 as
a probe for oncogenesis by AP-1 transcription factors:transcriptional activation
through AP-1 sites is not sufficient for cellular transformation." Genes Dev. 6
(1992): 1799-1809.
Parker, C.S., K.D. Harshman, and W.S. Moye-Rowley. "Yeast YAP] encodes
a novel form of the Jun family of transcriptional activator proteins." Genes
Dev. 3 (1989):283-292.
Pfeifer, K., T. Prezant, and L. Guarente. "Yeast HAP 1 activator binds to two
upstream activation sites of different sequence." Cell 49 (1987): 19-27.
Powers, S., T. Kataoka, O. Fasano, M. Goldfarb, J. Strathem, J. Broach, and
M. Wigler. "Genes in Saccharomyces cerevisiae encoding proteins with
domains homologous to the mammalian ras proteins." Cell 36 (1984):607-612.
Remade, J.E., and S. Holmberg. "A REB 1-binding site is required for GCN4-
independent ILV1 basal level transcription and can be functionally replaced by
an ABF1 binding site." Mol.Cell Biol. 12 (1992):5516-5526.

Roesler, W.J., G.R. Vandenbark, and R.W. Hanson. "Cyclic AMP and the
induction of eukaiyotic gene transcription." J.Biol.Chem. 19 (1988):9063-
Sanger, F., S. Nicklen, and A.R. Coulson. "DNA sequencing with chain-
terminating inhibitors." Proc.Natl.Acad.Sci.USA 74 (1977):5463-5467.
Schauer, W.E., and J.R. Mattoon. "Heterologous expression of human 5-
aminolevulinate dehydratase in Saccharomyces cerevisiae." Genetics 17
(1990): 1-6.
Schroetke, R. The effects of a RAS protein-modulated regulatory cascade on
the cytochrome biosynthetic pathway: a possible role for cAMP-dependent
protein kinase. Honors Thesis. University of Colorado at Colorado Springs.
1987. (UnPub)
Schutz, G., F. Weih, W. Schmid, C. DeVack, E. Kowenz-Leutz, B. Luckow,
M. Boshart, and M. Nichols. "Phosphorylation of CREB affects its binding to
high and low affinity sites: implication for cAMP-induced gene transcription."
EMBOJ. 11 (1992):3337-3346.
Sheng, M., M. Thompson, and M. Greenberg. "CREB: a Ca2+-regulated

transcription factor phosphorylated by calmodulin-dependent kinases." Science
252 (1991): 1427-1430.
Stillman, D.J., P.R. Nicholson, and H. Wang. "Identification of a
Saccharomyces cerevisiae DNA-binding protein involved in transcriptional
regulation." Mol.Cell Biol. 10 (1990): 1743-1753.
Struhl, K. "Regulatory sites for HISS gene expression in yeast." Nature 300 (1987):28-
Struhl, K. "The jun oncoprotein, a vertebrate transcription factor, activates
transcription in yeast." Nature 332 (1988):649-650.
Trawick, J.D., N. Kraut, F.R. Simon, and R.O. Poyton. "Regulation of yeast
C-OX6 by the general transcription factor ABF1 and separate HAP2- and heme-
responsive elements." Mol.Cell Biol. 12 (1992):2302-2314.
Trejo, J., J. Chambard, M. Karin, and J. Brown. "Biphasic increase in c-jun
mRNA is required for induction of AP-1 mediated gene transcription:
differential effects of muscarinic and thrombin receptor activation." Mol.Cell
Biol. 12 (1992):4742-4750.
Verdier,J.M. "Regulatory DNA-binding proteins in yeast: An overview." Yeast
6 (1990):271-297.

Verdiere, J., G. Mauricette, and R. Labbe-bois. "CYPl(HAPl) is a determinant
effector of alternative expression of heme-dependent transcribed genes in
yeast." Mol.Gen.Genet. 228 (1991):300-306.
Vincent, A.C., and K. Struhl. "ACR1, a yeast ATF/CREB repressor." Mol.Cell
Biol. 12 (1992):5394-5405.
Young-Sun Lin, and M.R. Green. "Identification and purification of a
Saccharomyces cerevisiae protein with the binding specificity of mammalian
transcription factor." Proc.Natl.Acad.Sci.USA 86 (1989): 109-113.
Zitomer, R.S. and C.V. Lowry. "Regulation of gene expression by oxygen in
Saccharomyces cerevisiae." Microbiol.Rev. 56 (1992): 1-11.
Zvonok, N.M., E. Horvath, and G. Bajszar. "Construction of versatile
Escherichia co/z-yeast shuttle vectors for promoter testing in Saccharomyces
cerevisiae." Gene 6 (1988):313-318.